The Most Important Things a Linux Desktop Newbie Needs to Know

[This article was most recently edited 23 Aug, 2013]

I am making this particular document available for public consumption, under the Creative Commons Attribution-ShareAlike 3.0 Unported License.  You are free to re-distribute, possess, add-to, and otherwise propagate this document, under the terms of the Creative Commons Attribution-ShareAlike 3.0 Unported License.  Other documents/pages featured on L. Lucas blog, materials, or other outlets, may be subject to other licensing-schemes.

This document is for those of us new to Linux-based desktop operating systems for personal computers.  HOWEVER, this document assumes that you, as a person interested in perhaps migrating to such a system, have done at least a little research of your own, before you got here, and that you have at least * some * basic grasp of personal computing, generally.  In the text, I often assume you are already acquainted with certain basic terms and concepts, such as the difference between software and hardware, and the meanings of some common acronyms, like RAM and CPU.  Most other things will be at least partially explained.

[This is essentially the ** condensed ** version of this information.  This file has about 40 entries, which—believe it or not—I have tried to make as brief as possible—while also taking-into-account that many of us folks trying to migrate from Windows * are * not * so * tech-savvy *.  And we may not be very clear, for example, on the difference between, say, our BIOS menu screen, and our BIOS boot-screen; so things like this are explained here (at least to * some * extent).  For the * expanded/full * version, look for the file “newbie get started and migrate to linux desktop tip by tip”; it has 80 + entries, more detail, and is oriented toward those who ** like ** to ** read **.  MY PRIMARY PURPOSE IN POSTING THIS INFORMATION IS SO THAT OTHERS DON’T HAVE TO REPEAT MY MISTAKES, OR DUPLICATE MY RESEARCH, wasting time by “re-inventing the wheel”. ]

Extreme Condensation:

The Entries Below, by Number:

1. Entry-1:  Disclaimer     c. pp 3
2. Entry-2:  Disclaimer Addendum:  all desktop Linux performance varies where some hardware is concerned     c. pp 6
3. Entry-3:  What Does it Mean:  32-bit Linux and 64-bit Linux     c. pp 7
4. Entry-4:  Use of “Live Media”:  booting desktop Linux as a “live environment”, which will not touch your harddrive, so you can try it out.    c. pp 7
5. Entry-5:  Always set the Burner to the Slowest Speed, when burning a Linux distro to cd or DVD     c. pp 11
6. Entry-6:  Do Not Throw Away your MS Windows Install.  Although Modern Desktop Linux is very complete, there might still be a few times when you still need WINDOWS.     c. pp 11
7. Entry-7:  The Best Way [for us Noobs (?)] to Solve the Problem of THE OS NOT CONNECTING to the Web/Internet     c. pp 14
8. Entry-8:  Most Reliable Equivalent to ctrl + Alt + Del to re-boot Linux    c. pp 15
9. Entry-9:  Another Important Manouvre of Desktop Linux, equivalent to ctrl + Alt + Del     c. pp 17
10. Entry-10:  There is a Setting you can Enable in Most Desktop Linux, which will cause the system to try to shut itself down properly if someone just quickly “taps” the power switch    c. pp 19
11. Entry-11:  The Restricted Extras Package Must be Downloaded and Added to Regular Ubuntu:  see this entry for the easy way        c. pp 19
12. Entry-12:  Why Linux is Harder to Install to a Laptop or Mobile Device [and some (comparatively) easy workarounds]      c. pp 20
13. Entry-13:  BIOS Booting your Desktop Linux “Live” cd, DVD or USB Thumb-key:  a (basic) walk-through         c. pp 21
14. Entry-14:  At Least Skim-Over the Official Release-Notes for your Distro    c. pp 23
15. Entry-15:  What do they mean by a Linux Desktop “Release” ?   c. pp 23
16. Entry-16:  Sometimes you can use an older release, to get desktop Linux to work well on your hardware   c. pp 25
17. Entry-17:  What’s the Best Desktop Linux for a Noob/Newbie (a person new to Desktop Linux) ?   c. pp 25
18. Entry-18:  Dual Boot and “Dual Boot Headache”      c. pp 28
19. Entry-19:  You might wanna watch-out for Microsoft/Windows’ * “Secure Boot” *………          c. pp 30
20. Entry-20:  The Desktop Linux Equivalent for Windows’ Start > Run:  Alt + F2……     c. pp 30
21. Entry-21:  Don’t use the Linux * Computer Janitor * feature—it isn’t perfected yet       c. pp 30
22. Entry-22:  A Word About Safe Web-Usage Practices:  there aren’t any Linux viruses “in the wild”, at the time of this writing, BUT……..      c. pp 30
23. Entry-23:  Linux is Easier to Run as a Live, Compressed File-System, than Windows Or Apple/Mac         c. pp 30
24. Entry-24:  Installing Apps in Modern Desktop Linux:  There * is * no single method to install a software in desktop Linux.  But this is rather a technicality:  there are Wizard-like, graphical programs to help us    c. pp 32
25.Entry-25:   What is Linux “Swap” ?  It is the equivalent of Windows’ Virtual Memory.  Also, a ‘walk-through” of harddrive partitioning here, and some CAVEATS     c. pp 34
26. Entry-26:  If you can’t Live with the Default Desktop Environment:  The End of the GNOME Deal, Ubuntu’s New (Crummy) Unity Interface, the Contemporary Linux Desktop Environment Mess, and 3 Easy Outs for us Mere Mortals      c. pp 40
27. Entry-27:  You Install a Software (“App”) from Control-Panel in Linux:  Linux “Control Panel” (Software Center, a.k.a. Software Manager) is a * two-way * street—whereas Windows’ CP only lets you * un-install *     c. pp 43
28. Entry-28:  Most of us would usually just re-install desktop Linux, if the system were to become “broken”. This is rare, and Modern Desktop Linux is much easier and quicker to re-load, than Windows.   c. pp 44
29. Entry-28:  Modern Desktop Linux still has great difficulty supporting certain hardware, even at the time of this writing (2012).     c. pp 45
30. Entry-30:  Linux has considerable difficulty in supporting certain peripheral devices (i.e. certain printers, certain USB web-cams, & tc.:  BUT there are some workarounds).    c. pp 45
31. Entry-31:  Linux lacks equivalent programs for a few of the lesser-known (“obscure”) OFFICE software programs.  [That’s why we preserve our Windows install (one reason, anyway)—for that “twice a year” OFFICE occasion.  But because of recent developments, this situation seems rapidly to be changing….]         c. pp 46
32. Entry-32:  There are many * very nice * audio / music editors available for modern desktop Linux, but……   c. pp 46
33. Entry-33:  The Linux Platform has been somewhat opened-up to Gaming    c. pp 46
34. Entry-34:  As to Un-Installing Desktop Linux     c. pp 46
35. Entry-35:  When do you have to de-frag in Linux?  Well, * NEVER *, basically.  Linux doesn’t need it.  c. pp 47
36. Entry-36:  Linux is CaSE-SeNsiTivE”—but this really only applies to command-line operations   c. pp 47
37. Entry-37:  Linux counts from Zero ( 0 ), NOT from one     c. pp 47
38. Entry-38:  You * can * try dragging your cherished Windows apps with you, over to your Linux install, and running them with a compatibility-layer, like the Linux WINE program.  But you’ll like a native Linux app better.     c. pp 47
39. Entry-39:  “RIVER” can be better than “POD”    c. pp 48
40. Entry-40:  Can’t find an instruction on how to do something from the Terminal, even though you’ve Googled it repeatedly?   c. pp 48
41. Entry-41:   If you’re stuck, and really having difficulty getting things to work, Googling around for “Post-Install Checklists” for your distro (i.e. “Ubuntu 10.04 post install checklist”) can help shed some light.  Read this entry first, though.    c. pp 50
42. Entry-42:  Copying and pasting into and out-of a Terminal window    c. pp 50
43. Entry-43:  Sometimes the “paste-special”, or “paste as unformatted text” feature in Open Office Writer (or even Libre Office) remains greyed-out……..    c. pp 51

44. Entry-44:  As to Upgrading a Linux Distro that you already have Installed to your Harddrive            c. pp 51
45. Entry-45:  As to Naming Files, for Compatibility on Linux and Dos-Based Systems     c. pp 52
46. Entry-46:  Need to Edit a System text-file?     c. pp 53
47. Entry-47:  DOCUMENTS IN LINUX      c. pp 53
48. Entry-48:  SAFE REMOVAL OF EXTERNAL USB HARDDRIVES and THUMB-KEYS:  c. pp 54

Expansion:

1. Entry-1:  DISCLAIMER:  I am * not * trying to be in the business of sowing F.U.D. (“Fear, Uncertainty, and Doubt”) about Linux:  I really * like * Linux, and I like its concepts.  But I feel one of us ought to be more frank about the “facts on the ground”, so to speak.  A few of the drawbacks might be ascribed to Linux itself; and many of these are * ”social” *, rather than “innate”—i.e. Linux programmers (who are often doing it for free—or, more to the point, for their own fame and sense of accomplishment—which is just fine):  well, these are not always disposed to creating some utility that is lacking, but which just “isn’t  ‘sexy”.  In terms of developing, say, a GUI front-end for some not very often done operation, which most Linux “initiates” seem able to do from a command-line anyway.

Other of the “drawbacks” can be put squarely in the province of the hardware makers, who have little incentive to make their products “Linux-compatible”, Out-Of-The-Box.  Yet a surprising number of them do just that.  And there are often simple hacks and work-arounds available on the web, sometimes available even at the major hardware vendors websites.  So maybe the hardware makers are trying to tell us something. (?) ]

Linux * is * a better system, in many technical ways.  But most of these would not matter to the average person, who just wants their computer (laptop, tablet) to actually work, when they want to do something with it.

Even so, I have to say that at this juncture, my thinking is that the Linux desktop will always be a little harder to use than the two main commercial platforms (WINDOWS and Apple/MAC).  [Though I have to say right here, that desktop Linux is often * remarkably * easy *, where it comes to most common tasks.]  So why is desktop Linux (Ubuntu, Linux Mint, et al) a little more difficult, once we get away from just browsing the web, listening to music, or downloading videos?  * Because * it’s * not * bundled * software * [except in some comparatively rare circumstances, such as System76 computers, or some other marques (hardware vendors) that offer PCs with a Linux desktop already installed, or perhaps some larger machines that are ordered with only RHEL in mind].  Linux seldom comes as the default operating-system on any personal system you buy, for ordinary home- and on-the-go use.  Yes, there are some.  But most machines come with WINDOWS (or MAC).  And why is this?  Probably because the other two are for-profit concerns, in the end-user market.  Both of them got their foot in the door, so to speak, earlier-on in the personal computing revolution.  And both platforms and the corporations that own them have concentrated on market penetration, and steadfastly maintained the grip of the bulldog’s jaws, upon the bull-neck of the consumer pc market, if you will allow me a klunky metaphor.  Desktop Linux, on the other hand, is only a for-profit concern in the server and industrial markets, and this only in certain circumstances.  Yes, there is ANDROID.  Which * does * use a Linux kernel:  but * that * I often feel has created its own category.  And ‘DROID is still a system mostly for smart phones, at the time of this writing (late 2012).  Not that many people seem to be booting it on a netbook, or a laptop.  The rest of the time, a Linux desktop distro is something that may be installed at a later time—after the device has left the retail product-stream.  And this from a variety of motives.

It is also true that many (though by no means * all *) of the people who create many of the open-source apps Linux uses are just—well, they’re a * different * crowd, in a way:  often the FOSS software writers are working from a somewhat different motive than when programs are written for the other two platforms; people write these Linux desktop apps * for * themselves *, as much as for the general consuming public, and they’re often doing it on a volunteer basis.  And this is wonderful.  It makes you feel better, in a way, about “human nature”, and the world generally.  And I must say, just about every open-source app I’ve yet used has worked * very * well *.  But it also reflects some of the difference in philosophy between desktop Linux and its distributed community, and the two major “commercial” platforms.

People who create many of these FOSS (Free Open Source Software) APPS often times are making the application for their own use first, and then also for the world.  The Linux platform itself is * powerful *, and this is made-use-of.  But because the person is writing first for himself, and the power of Linux is not something that he or she wants to go to waste, oft times the resulting program turns out to make at least a little more demand upon the end-user:  the interface of many Linux desktop GUI apps is often a little more demanding, than, say, some of the simpler “free-to-download” apps in Windows.  The controls in some open-source app are frequently less “intuitive”.  And the Linux desktop community * chafes * at the idea of mixing Free/Open-Source and pay-for applications.  Yes, some considerable gestures have been made in this direction.  Especially in recent months and years (relative to the time of this writing).  But if you think about it for a little while, you can probably understand the reluctance of Linux’s Community, to “let the camel’s nose under the tent”, so to speak.

It might also be tempting to ascribe some of this to the formation of UNIX, which was the Main-Frame operating system upon which Linux is based, and which is still in use in the computer industry today.  But differences in the Microsoft-type and UNIX-type file-systems don’t seem to have much hampered ‘DROID users.  So it wouldn’t seem to be the “learning curve” of getting used-to the underlying file-system.  At least, not wholly.  Yes, “DROID, as I’ve said, * is * not * a Desktop * system *—like the type you want on a machine larger than a phone or tablet—especially to use for productivity.  At least it is not * intended * as such.  But then funny things have happened in software development, in its (relatively) brief history.  By the time I’ve finally gotten around to uploading this paper to my blog, a fair number of people may have figured out how to use ANDROID * as * a * desktop *.  Never say never.  Really, I suppose some folks are using it as a sort of * limited * desktop * now *.  ANDROID * can * be booted on a netbook, or even a full size laptop.  But I don’t know if you can run Libre Office or Kompozer from it—let alone port something like Final Cut-Pro to it.  I wouldn’t think so, especially given that most builds of freely available ‘DROID seem to be on the order of only 80 Mb or so.  But then I remind myself that GrafPup Linux was only about that size.  And some people * did * use that as a desktop, though unfortunately it seems now defunct.

Apparently the programmer who built GrafPup got tired of maintaining it.  This is another issue in desktop Linux:  the developer or college student who made some variant or re-hack of Fedora, Knoppix, or whichever of the major distros—that person will occasionally get tired of his-or-her labor of love, and abandon the distro.  Or perhaps he or she just * graduates *, and that’s the end of the distro they had been offering the rest of us.
Fates such as these have befallen even a few Linux desktop distros of note.  The much loved DSL (“Damn Small Linux”), Feather Linux, and GrafPup, and some others that had achieved a following.

Often, though, a new distro will rise-up, to fill the void.  SliTaz Linux, for example, seems destined to be our replacement for DSL.  Now, before some people out there reach for that mouse (or a gun), let me add very quickly that I am * NOT * saying that SliTaz * IS * DSL—or that it is a * clone *, or some perfect analog for DSL.  I am just trying to illustrate my point:  desktop Linux is in flux, and changes more frequently (or at least more “radically”), than ms Windows.  But I will add rather quickly also, that—as I write this—Windows 8 Power Shell is poised to hit the stores for Christmas (2012); and indications, based-on preliminary tests, seem indicative that the interface is going to be a mess for productivity and work-flow—just as seems to be the case with the current major Linux GUI desktop-environs (GNOME 3/GNOME-Shell, KDE 4, and the new Unity DE).  Microsoft’s hands are far from clean, where it comes to “pulling the rug out from under” users.  Witness the changes to the user interface in Word from the 2007 version to the 2010 version.  Some who thought they new how to use the product were left scratching their heads.  And this is far from the only such instance.

It is basically due to the “destabilization” of the GNOME Desktop Environment/GUI that I am now using Linux Mint 13 XFCE Edition as my workhorse-os.  It comes with the XFCE Desktop Environment by default, instead of Gnome * 3 * (yuck!) or KDE 4.  The new KDE Desktop Environment (KDE 4) is better now than in the first months of its release, but you probly won’t be able to run KDE 4 on hardware that’s as lower-powered as we could with GNOME 2.x; KDE always did have the highest processing-speed requirements of pretty much any Desktop Environment for Linux.  For those with at least a Pentium 5 or equivalent and at least, say, 512 Mb. of RAM, this probably won’t be an issue.  Also, KDE is Qt toolkit oriented—Qt toolkit being the major building-engine for KDE apps, and a software set that’s owned by a corporation (Nokia), whereas GNOME’s gtk + software-framework is under the aegis of the GNOME * foundation *—just like GNOME itself.

XFCE used to be half-jokingly referred-to as “GNOME-Light”, and it supports apps built with the intention of running on * either * GNOME 2x and GNOME 3, as well as KDE.  XFCE also has a collection of its own “native” apps.  But really, this “orientation” of Linux desktop apps seems not to have mattered in years, where it comes to running which type of app on what Desktop Environ—if it ever really did matter.  I find I am able to install “K-apps” just fine in Ubuntu 10.04 with the default GNOME 2 desktop—often installing just from Software Center, with just a couple of mouse-clicks.  And I am told KDE will run gtk + stuff just fine.  The other Desktop Environments and window-managers in Linux are all capable of running either, as far as I can tell.  XFCE can also be installed to regular Ubuntu, or to other Linux desktop systems (though to install to regular Ubuntu we’re supposed to use something called “Xubuntu package”).  Once you get used-to XFCE’s few quirks (which I’ve tried to detail in this document), you should be able to use it for productivity.  See my doc “l xfce desktop environ”—especially the section “XFCE Basics and Caveats”.

I will add right here, at once, that the “Majors” (Major distros of desktop Linux) are much more stable, in terms of staying-power, at least.  In saying “staying-power”, my meaning is “is the distro likely to still be around and in a state of current-maintenance in the future”.   These would be:  distros such as Fedora, CentOS, probly also PCLinuxOS, Ubuntu, Linux Mint, MEPIS, Knoppix, Slackware, probly also Gentoo Linux, and the Puppies.  At least according to my dumb a%%#!&, that I could think of, off the top of my (pointed) head.  Apologies in advance to those who are about to flame me, because I didn’t mention their favorite distro.  I should also say quickly that entry No. 17, below,  is devoted to which distros are the best for a newbie/n00b—which just means a beginner.

There’s also the * other * kind of stability:  by which people usually mean “it doesn’t crash or freeze-up”.  Linux historically has had a reputation for this kind of stability.  Throughout much of the past two decades, Linux users were able to boast of how Linux desktop distros like Slackware, Yggdrasil, and then Ubuntu never crashed, while ms Windows installs would sometimes have frequent crashes and freeze-ups, and were fodder for virus-writers.  Well, boys and girls, sad to say the proverbial shoe is now on the other foot, thanks to the wandering-off-from-purpose of the GNOME Project.  Yes, the Linux * kernel * itself is still pretty dog-gone stable.  (So is Windows’ NT-kernel, for that matter—NT having replaced the MS-DOS underlying infrastructure in the late 1990s.)

But that  which is * just * as * important *—the Desktop Environment (by which—in this context—we mean the GUI)—has unfortunately gone through a recent period of turbulence, and Linux DE s are only now beginning to emerge from re-mixing and its attendant bugginess.  XFCE always * did * have some quirks and “bugs”.  And these are still present in XFCE 4.8, basically, although the XFCE developers have done many other sorts of things to the DE, to make it more powerful, and to make it easier to use where it comes to things like settings and work-flow/productivity.  [See the entry below, as to “If you can’t Live with the Default Desktop Environment,….” No. 26]

Nevertheless, I notice I am able to use Linux Mint 13 XFCE Edition on a daily basis, and get all my work done.  It depends, I suppose, on just what you need to do.  I find that, for everyday tasks (and some that are not everyday-type-of-things), Linux Mint 13 XFCE Ed. performs well within the bounds of acceptability, and it is noticeably faster on my 10” netbook than the Windows 7 install—even though the manufacturer stares plainly that this “mini-laptop” will support Windows 7 Home Premium.  I can say the same for my install of Ubuntu 10.04 on my full-size laptop, and for various desktop Linux installs on my old Thinkpad T-20 with a Pent. 3 and only 384 Mb. RAM [though I notice the only Linux Mint build I have been able to successfully run on this low-end machine (so far), has been Linux Mint Fluxbox 9].  I did try Linux Mint 10 GNOME version on this old box.  It booted, but it was difficult to get the cursor to budge.  Eventually, I could do a task, but it would take forever.  Ubuntu 10.04, however, runs well on this old machine, although more slowly.

And I’ll add that I had (some) trouble installing desktop Linux (Ubuntu 10.04) to my full-size laptop (a Toshiba L515 Satellite):  but this was in large-part due to my lackadaisical approach—had I but been more diligent in my reading of the existing online documentation and help pages, I would have arrived at a usable system much sooner.

Linux has been criticized as being “underdocumented”. i.e. not having enough documentation—which is just a fancy way of saying “lacking in instructions and manuals”.  Well, having been on desktop Linux for about two years now—almost never booting my Windows install, except maybe to let it update once in a while—I have to attest that nearly all of this “documentation problem” has been fixed.  There is a huge amount of free support available, mostly online, but from other sources too.  Linux desktop stuff is now remarkably well documented, except perhaps in the case of some of the more “exotic” and “obscure” distros.

Occasionally, though, modern desktop Linux will still throw us a curve.  “GRUB 2” is a good example.  Cannonical, Ltd., switched their distro—Ubuntu—to this new version of the bootloader, as soon as it was in “final release”.  Meaning “as soon as GNU Grub 1.98 (the first iteration of “Grub 2”) was ‘out of Beta-testing’, and had been granted the stamp of ‘final release’ by its makers”.  Which is fine, I guess.  “Grub 2” (alias Grub 1.98, Grub 1.99) has some new features which, depending on your point-of-view, would qualify as * improvements *.  And I must say I have yet to have a problem with * any * version of Grub.  But the * manual * for Grub2 that you could access online was seemingly * incomplete *, and this situation prevailed for a while.  Even so, I found it did not hamper my migration to the Linux desktop.  Most of us probly use Grub just as it comes, and aren’t much concerned with customizing it.  I’m speaking of ordinary desktop uses, at least.  But for some of the more advanced people out there, these “undocumented changes” to the Linux bootloader handed them a difficulty.  In any case, Grub 2—at least the 1.98 version—is now pretty up-to-date, where documentation is concerned.

When all is said and done, I have to say that I find desktop Linux well documented, generally, and well-supported, in terms of free online resources, and related help.

There will always be the occasional curve-ball, though.  This is just inevitable, in a group of free desktop systems that are relying-on an “alternative business model” [read donations, volunteers, and failed (well, semi-failed) concepts like the Linux “man-month”].  But I feel compelled to add, very quickly, that such an occasional “curve” hasn’t stopped my newbie b*tt from continuing to use a Linux desktop as my daily os.  For productivity and also for “fun”.  Not even a curve like the “Linux Desktop Environment Mess”, which I just walked-through in the above paragraphs.  Once you “get your feet wet”, so-to-speak—the more accustomed you become to working with Linux as a desktop, the easier it gets.  And the better you get at hitting these occasional “curve-balls”.  And, I will add, without much resorting to a command-line.

I suppose that brings to mind one more thing:  Linux desktops are notorious for forcing you into a command-line, in order to do certain tasks that “aren’t GUI ready”.  Or this * has * been * the reputation.  Well, I can attest that I’ve been using a Linux desktop almost exclusively, for more than two years now.  And I find my (increasing) use of the command-line / Terminal-emulator to be at least 95 % out of things like * curiosity *, and * experimentation *—and * NOT * because there wasn’t a GUI way to do something.

Desktop Linux * is * going to make more demands on you, generally, than, say, MAC OS/X, ANDROID, or Microsoft Windows.  But you get the possibility of greater system privacy and security, and the opportunity (if you want it) to become more competent in computers, generally, and better understand how they work.

2. Entry-2:  Just because, Bubba, the things I describe have worked on * my * own * machines, does not necessarily mean they will work for  * you *,  on  * your *  machines.  Ubuntu, Puppy Linux, and Knoppix, as well as certain specialized tool-sets that come as their own (small) Linux desktop (Parted-Magic, System Rescue boot disk), will boot and run on a huge variety of equipment.  If you’re determined, you can often get it working on “stubborn” equipment. Linux is not like ms Windows, which is tested specifically for bootability on every computer that enters the product-stream, in the * whole * world * (except, of course, for Apple-MAC, which is rather in its own category, respective of “bootability”).

The bottom-line, though, is that Ubuntu, most of its “variants” (like Linux Mint, PinguyOS, &tc., &tc.), as well as Knoppix, and Puppy Linux * ARE * VERY * BOOTABLE *—in that they will often boot-up without issue * on * a * very * large * variety * of * computers *.

There are some other pitfalls, too, about which I have tried to be frank, in the entries below.

Some of us invoke what Linux-heads half-jokingly call “the E-Bay Patch”.  What this means is just that a component or two in my laptop doesn’t want to support Linux, so I take a screwdriver to the computer’s case, and remove the problem hardware, then sell it on E-Bay (or Amazon, whatever); then I use the money (plus probably some out-of-my-pocket) to purchase a supported card * from * E-Bay/Amazon, etc.  This is alright for those who know what they’re doing.  Otherwise, you may have to research the hardware carefully (online, public library, etc.), and then order and pay-for the needed components up-front, and then have a reputable computer shop make the change for you, and * then * get what money you can out of the exchanged cards, by posting them on the “ ‘Bay”.

I am also come to the conclusion that * some * machines just * will * not * run Linux—no matter how hard you try.  But this is comparatively few.  Modern desktop Linux can be gotten to work * on * most * hardwares *, even if you’re just someone who’s * approaching * the level of “Advanced User”.  (* I * was able to do it as a * n00b * -user, barely knowing anything.)  EVEN ON “ * STUBBORN * ” MACHINES, one can often get desktop Linux to work, * if * you A) have at least a bit of a pioneer-spirit, B) have disposable time, C) are willing to experiment, and get people online to help you, D) you * like * to * read *, E) are stubborn as hell, F) can perhaps join a LUG (Linux Users Group), and G) are on good-terms with / sleeping with a geek/computer-nerd.

3. Entry-3:  Know that most computers today (well, at the time of this writing) have CPUs that have a 32-bit address-bar.  But the 64-bit ones are coming-along.  A lot of these are already in use.  And most Linux desktop distros now offer * both * the “traditional” 32-bit build, and a 64-bit build.  Most of us desktop users seem to stay with the 32-bit version.  At least for now—relative to the time of this writing—in 2012.  It is (somewhat) easier to install new apps to the 32-bit builds.  And it’s the same principle as with potatoes:  you can put 32 ponds of potatoes in a 64-pound potato bag; but you cannot put 64 pounds in a 32-pound bag.  A 32-bit build of desktop Linux will almost always run fine on a machine with a 64-bit CPU—it just won’t make real efficient use of the computer’s available (huge) processing-power.  But with Linux, this usually isn’t much of an issue anyway—Linux is capable of doing some big tasks on a feeble CPU, so this is kind of a wash.  Depends, I guess, on the sort of tasks you’re interested in performing on a Linux desktop system.

4. Entry-4:  IF YOU ARE GONNA MESS WITH LINUX ON CD s OR DVD s (as opposed to making bootable thumb-keys with the kind of GUI programs now freely available), then YOU SHOULD KNOW THE FOLLOWING:  Not all cd trays on same standards.  This is something that is really too bad for everybody, and affects things like music-file sharing, as well as the bootability of Linux on optical disks (cd s and DVD s).  Knoppix has nearly un-paralleled hardware recognition (detection and support).  So does Ubuntu.  A Knoppix cd or DVD will boot and run on a huge variety of computer equipment.  This is also true for Ubuntu and its variant-distros, that use Ubuntu as a starting-point/basis, and then build on top of it.  Puppy Linux also.  BUT UNFORTUNATELY, SOMETIMES A LINUX CD OR DVD THAT YOU BURNED ON ONE COMPUTER JUST WILL NOT BOOT ON SOME OTHER CERTAIN COMPUTER. There are several vendors in the business of making the cd-drives, for the major hardware manufacturers.  Unfortunately, not all use the same hardware/software standards, where it comes to play-back of the disc.

People on some of the various online forums love to tout to you the wonders of having a totally portable Linux desktop operating system, self-contained, on a DVD.  Well, these * are * nice, and I’ve used  ’em myself.  Unfortunately, if you use one long enough—on more and more computers, I’m saying—you start to come upon machines that just won’t boot the distro, or which won’t boot it after the first time.  This is also true, by the way, of many laptop-created music cd s.  Because not all the cd-drives use the same standards-sets—not even at the time of this writing, in 2012.  It’s not as noticeable with music files, or even video.  But things get touchier, where it comes to getting some machines to boot * an * operating * system *, where that operating system wasn’t burned to disc using * that * computer’s * cd-tray.

The cure is, well, simple enough, though not particularly convenient:  If you wanna run desktop Linux on some machine by means of optical disc (cd/DVD), and you had difficulty booting it on that machine, and it is also true that you burned the cd on * another * computer—then try downloading and burning the same Linux distro on * that * machine—on the * same * machine you intend to run it on.  The differences in the standards of cd-trays can be subtle, and can require considerable Googling to unearth.  Modern computers (in this context, equipment manufactured after, say, 1995 or so) just are not built with the intention of the user changing the operating-system any more:  that’s more of a 1980s thing (you know, back when there were no harddrives, but only two floppy slots in the front, and Ronald Reagan was in the White House.  If you don’t remember, or weren’t alive then, ask someone who had to do a lot of work in an “automated” office back then.)

Often this quick-fix will produce results.  If you * still * can’t get Linux to boot, try doing some research on the make and model of BIOS that is installed in the target computer, and perhaps its other major hardware too.  Take a look at the Release Notes of the desktop Linux distro you are trying to use.  BIOS firmware cannot be swapped-out for another firmware:  * but * the BIOS * can * have its firmware updated (“re-flashed”)—and sometimes this will then allow Linux to go ahead and boot.  NOTE that * only * knowledgeable * persons * should attempt to mess with the install of BIOS firmware.  If this is gotten out-of-whack, it may take a good computer shop to straighten it out, and it will not be able to boot anything until that is done.

Be aware, too, that many public library and/or kiosk computers (i.e. “public computers”—like at an internet cafe’), and computers at the company where you work, have had their BIOS “adjusted” by the system-administrator (that is, the “boss” who’s in-charge of the maintenance and up-keep of those machines).  In this context, by “adjusted”, I mean that certain things may have been done to the individual machines or their LAN that will prevent a non-whitelisted operating system from being booted from the cd-tray, or other location.  So your portable Linux os on a DVD that you made may still not be usable at the public library, even though the cd-tray is compatible with the one in the computer you used to create the disc.  It depends.

Alternately, a person can nowadays make a bootable USB thumb-key, and boot Linux from * that *.  If you like the distro and how it behaves, you can invoke the distro’s built-in installer, and install it to your harddrive.  Or just continue to run Linux “live”, from the thumb-key.  There are pros and cons to both scenarios.  Some research will shed light on this.  And I have tried to make it clear elsewhere in this document, and other docs.  Note that there is also the “Virtual Machine” method, to boot and run desktop Linux.  Modern desktop Linux can be installed and run this way * from * inside * ms Windows *—just as if it were any other application.  This way also makes it easier to un-install.  See entry Number 18.

NOTE that any live Linux disc (cd or DVD) is gonna be  S – L – O – W – E – R  than when/and/if Linux is installed to and running from some other media, like an internal harddrive, or an external/add-on harddrive, or even a USB thumb-stick.  So optical discs (cds and DVDs) will run Linux * SLOWLY *, and not at full speed.  Not having enough hardware resources/ “computing- ‘horsepower” can also cause Linux that is being tried from a live cd to be slow, or to “hang”.  This is just something you have to suss-out.  Check the “system requirements” for the Linux distro in question against the hardware (CPU model and clock-speed, amount of installed RAM) of your computer:  see the official website of the distro’s maker, for the hardware requirements.

Puppy Linux, however, is rather an exception to this, because Puppy is one of the comparatively few distros that is small enough to have ALL of the File System loaded into the computer’s RAM as the boot process finishes—and Puppy is pre-configured to do so.  It is because of this feature that Puppy Linux will run just as fast from a live-cd as from any other boot-media.  There are other distros that are like Puppy in this way, but they are comparatively few.  Knoppix also—though it does not load all of its FS into the RAM—is somewhat exceptional in this regard, as it normally runs as fast from a cd/DVD, and the creators discourage harddrive installation anyway, because Knoppix is made to run from an optical disc or a thumb.

It may help you to first try some releases of desktop Linux (Ubuntu, KNOPPIX, Puppy, PinguyOS, etc.) that were released in about the same time-frame in which your machine was manufactured. Often this is helpful. Let’s remember that a Windows computer is generally associated with the time-frame in which its version of the Windows operating system was current. A Windows 2000 computer will usually have been manufactured around that time; an XP computer will usually have been built somewhere between 2000 and 2008 or so. Machines sold with Vista were usually built around 2007 or 2008.

A VERY IMPORTANT NOTE/UPDATE TO TACK-ON HERE, IS THAT MOST USERS (by the time I’ve had a chance to insert this) ARE NOW PROBABLY SKIPPING THE * BURN * THE *CD * PART, BECAUSE THE USE OF WINDOWS-BASED PROGRAMS LIKE UNIVERSAL USB CREATOR from pendrivelinux.com, and others (free!), ARE INCREASINGLY PERFECTED, AND SO YOU CAN TRY-OUT DESKTOP LINUX FROM A BOOTABLE USB KEY YOU MAKE IN ABOUT 15 MINUTES, AND IT WILL RUN AT VERY NEAR FULL-SPEED!  This is now a * better * way to test-run desktop Linux, and it will usually install just fine from the USB-drive, if you decide to do so.  Further, it usually won’t hurt the USB-thumb drive (“pen-drive” / “jump-drive”), and you could re-use it later, for other stuff, just by re-formatting it back to FAT32.  One caution about this method, though, is that you do run (some) risk of formatting the Linux distro to the wrong drive, if you’re   A)  really inexperienced, or   B) really tired.  So be careful!  Or just stick to trying desktop Linux from a cd/DVD.

If you have difficulty figuring-out how to format / re-format a USB thumb-drive from Windows XP, there is the HP Formatting-Tool For Windows, a nice GUI tool, which can be downloaded free of charge.  I think it may also work in Vista/7, but I’d have to check that.  Vista & + up allow you to format / re-format a drive by just going to Start, > my Computer, > RIGHT-click the drive in-question, and click “Format”.  From there, you can select “FAT32”, or NTFS.  These are the only two * Disk * Formats * which ms Windows understands, by default.  But for us, it is * enough *.  Yes, Windows cannot go into Linux and read/write to files—at least not without the installation of a third-party app (though there * are * several of these available, free).  But it’s still okay, because desktop Linux can read/write to all Windows formats, by default.  Disk Formats * and * files-formats.  Even .docx is now supported—and it has been since roughly 2010—depending on exactly which Linux software.  By the time I write this, however, all up-to-date versions of the popular Linux documents-programs seem to support .docx as well as .doc.

* Just * be * dang * careful * ** which ** * drive * you’re * selecting *, if you find yourself formatting / re-formatting a partition or device with ms Windows operating system.  Or ** ANY ** software.  If you accidentally re-format the wrong drive, you could accidentally erase precious data, or even your install of ms Windows itself.  And it might then take a professional to retrieve it.  So watch what is going-on.  It’s a ** computer **, not a toaster.  It’s not as easy as a * toaster *.  And it probably never will be.  This also is one more thing that illustrates the importance of good data-backup practices.

CDs and DVDs are still good, though, to make backup-copies of all your text, documents, batch, audio and video files (and your operating system(s) too), once you’ve installed the systems and programs you want to your harddrive.  Always make backup-copies of your Windows operating system on cds/DVDs as soon as you get a new (or used) computer, and then once again before you mess with Linux, BSD, or any other system.

WHEN IT COMES TO MAKING BOOTABLE THUMB-DRIVES:

I am about the last person who wants to disparage much of anything to do with personal computing—either hardware or software; but I will say that it may be well to stay away from using a Kingston thumb drive (for making a Linux bootable thumb):  from what I can glean, it seems they sometimes have real issues booting this type of os, after the first time.  Or even to boot Linux at all.  Remember that these thumb-drives were not invented for the purpose of booting Linux, or any similar booting.  Natty Narwhal 11.04 is having issues with USB flash drives from SanDisk that have U3 Launchpad.  You can either use another brand or use either u3-tool from Ubuntu Repositories or SanDisk’s U3 Launchpad Removal Tool to remove U3.

Ubuntu 10.04.3 is having issues.  You might get a segmentation fault if used from command line. There are many launchpad bugs regarding segmentation fault (eg: 572611).

Personally, I have successfully created more than 8 or 9 bootable Linux thumbs in the past several months.  I must say I have had very reliable results with SanDisk Cruzers.  Whether I re-format them first to remove manufacturer-installed softwares or not, these seem to work, creating the bootable thumb by a variety of methods.  Thumb-drives nowadays are just not that expensive:  I recently bought a 4 Gb. one for a measly ten (10) USD (U.S. Dollars).  Just 2 Gb. is usually more than adequate, where it comes to making a bootable Linux USB thumb-key.  Back when these jump-drives were first becoming widely available, you only got something like 512 Mb. of space, and they cost * a * lot * more.  Universal USB Installer from pendrive Linux.com is a good tool, and can configure persistence at the time of the install—for certain distros.  This nice free tool is being continually improved.  This tool runs from Windows.

Be sure to checksum your download of a Linux distro, before you try to use it.  What is a checksum?  It is a mathematical way (automated) to verify the integrity of your Linux operating system, that you have downloaded from the internet.  You use checksum from a graphical program, * before * attempting to boot or install Linux.  Checksum verifies that the system files did not get damaged, as they flowed through the millions of miles of information “pipeline” that is the modern internet.  You compare an alpha-numeric “hash number” that is the product of a calculation you can make in either ms Windows or Linux, respective of your recent download, against a “known good value”, which you obtain from the official webpage of the distro in question.  Frankly, you can probably use just about any Linux-oriented webpage as a source for the “known good value”; if the product of the calculation you run from your end matches at all, then the download is almost certainly good:  the odds of a bad download accidentally matching a hash-string from the web are probably about one-in-a-million.  So doing this protects you from accidentally booting a corrupted os filesystem, when you boot the distro from live-media.  You only need perform the calculation once, for * that * download of a Linux distro.

Where it comes to this checksum business, I like HashCalc 2.02 for Windows.  It is a GUI/graphical program that will run in ms Windows 2K, XP, Vista, 7.  Simple and reliable, and it has a built-in “browse files” button [you know, that inconspicuous little button with the elipsis (three little dots) in it].  It looks sort of like this:  “ […] ” .  Look for it.  You might note also that this button is pre-programed (in * this * app) to only show files it thinks are your Linux download.  So if you sometime find yourself dealing with some Linux .iso you’ve re-mastered yourself, you may have to find the exact file-path, and enter it manually.  HashCalc 2.02 is simple and lightweight.  In other words, HashCalc apparently contains no files that aren’t necessary, and doesn’t hog the computer’s available processing power when you’re using it.  Download it to Windows from CNET/Download.com, TuCows, or other reputable mirror-site.  You should probably remember to create a Windows’ Restore-Point first, just to be safe.  As with * any * file you intend to install to a Windows system, there exists the potential for malware, so use the same precautions that are advised to Windows users.

You’ll have to know how to highlight the results-field with your mouse, in order to copy/paste the result into a text-file (or just use a .doc), so you can compare the alpha-numeric string with the “known-good” one from the distro’s webpage.  HashCalc 2.02 supports the md5 algorithm, which is the most used at the time of this writing; SHA 1 and SHA 256, as well as many others are also supported by default, or can be enabled by ticking the check-box next to that algorithm.  It will take up to several minutes to complete the algorithmic-calculation(s), once you have clicked the “calculate” button, so some patience is in order.  Yes, there * are * more full-featured such hash-calculators available, on the world-wide-web—many of them also free.  But I can find nothing * wrong * with HashCalc 2.02; it is simple, and reliable.

5. Entry-5:  If you are about to burn a Linux distro to an optical disc (cd or DVD), set the burner program to the slowest speed it will go to.  This helps insure a “quality burn”.  You might get away with a “poor burn” of a music disc (Mp3 s), and it still sound okay; but where it comes to burning an * operating ** system * itself *, you need to be * picky *.  You cannot burn an .iso file to disc with Windows built-in disc burning, nor with most cd/DVD creator software for ms Windows (especially the free versions).  You’ll probably have to download and install a different kind of program, known as an “iso buster”.  There are many good free ones available on the web, for installation to Windows XP, Vista, and 7.  Myself, I like Active@ iso burner.  A word of caution:  as with any of the tremendous metaverse of freebie programs available for ms Windows, some of these “iso busters” have doubtless gotten contaminated (by black-hat hackers), with viruses and/or malware.  Use good sense, in anything you install to a Windows system.  Remember to create a restore-point first.  Download from mirror-sites that have a good reputation.  Do some research first.  Look for reviews of the program * from * actual * users * of the program—not just the staff writers at a website.  RIGHT-click the downloade3d file, and scan it with A) your anti-virus program, and then B) Malwarebytes (which you have first manually updated—usually just the act of opening Malwarebytes will cause it to update, but if not, do what you have to).  And * then * install your iso-buster.

6. Entry-6:  DO NOT THROW-AWAY YOUR INSTALL OF Microsoft Windows.  NEVER, NEVER, NEVER, NEVER!  There will be times when you will still need to boot a Windows install, no matter how much you get to like Linux.  Because you’ll be out somewhere—like at a public library of coffee-house, and need to print something, and maybe all they’ll have is a Lexmark printer—and Lexmarks and Linux have a troubled history.  Or you’ll need to do some operation with your tower computer that you don’t yet know how to do in Linux, and the work is due in the morning, and you know you’d better hurry-up and get it done.  You can learn the Linux app that does the process next weekend, when you’re not under a deadline.

Because of recent (crummy) developments in pc s, I have updated this entry:

I have recently begun to consider acquiring a new mobile hardware platform for myself, because i want to keep pace with technology, and for other reasons I won’t go into right here, as these would be beyond the scope of this article.  Suffice it to say that I am contemplating replacing my trusty AAO netbook with some sort of  heavy-duty tablet or what device may be able to fill its role as my “workhorse-from-public-wifi-and-coffee-shops” device.  I’m not sure if I am up to wrangling with WINDOWS 8 Secure-Boot just yet.  And I’ll mention right here as a caveat that much hardware that comes with WIN 8 * will * NOT * let you wipe-off WINDOWS 8, and install WIN 7 or XP (D’Oh! D’Oh! D’Oh!).  So I’ll probably look for something that has WIN 7–or that * had * it.  Why buy a machine with no working ms WINDOWS install?  What if I find myself in need of some lesser-used ms Office app, while at an airport far from home?

My strategy here is just to abandon the idea of a WINDOWS install on the new device altogether.  I’m tired of needing to update it when I occasionally boot it–like to use Universal USB Installer from Pendrivelinux.com–which is built to only run from WINDOWS.  This and other factors of having a traditional Dual-boot (actually multi-boot–I have 4 os available from this hdd) type of arrangement have gotten exasperating, at times.  And though I have not had “dual boot-headache” on this machine, I wouldn’t be confident of this also obtaining on the new piece of equipment until I actually * installed * Linux Mint.  You can’t always tell, unfortunately, just from booting a live cd.

What to do?

I’m thinking seriously of just etting a ‘DROID device (which we know * can * already boot Linux, since ‘DROID runs on-top of a Linux 2.4 kernel.  This of course is two series behind the current kernel series:  2.4.x, then 2.6.x; and now 3.x.x (at the time of this writing).  But it tells you that the device is perhaps more Linux-friendly.  Or I might go with something else.  And I don’t like the factor of WINDOWS 8 Secure-Boot being in the product-stream.  Yes, I know the official rationale is to boost system-security for WINDOWS.  But I just feel that perhaps the consumer should be able to * choose * this feature.  If I fork-out the money for the dern thing, shouldn’t it be up to * me *?

My point is that what I am considering is to just dual-boot Linux with ‘DROID–no ms WINDOWS install.  OR just wipe the drive (hybrid drive) and just boot Linux–either one distro or several, as I think I’ll use.  What about that occasional WINDOWS Office moment?  I’m thinking of just installing a VM container to my Linux desktop on this new machine, and downloading some fixed-image of XP or 7, which can be booted that way.  I’ll have to research this some more, before I make up my mind.  And it is IMPORTANT to understand here, that such a “fixed WINDOWS image” A) is pirated, so I’d think I’d have to get it off The Pirate Bay, or Megaupload, or some such P2P file-sharing site (not such a big deal to me, at this juncture, because I’m angry over the Secure-Boot thing); and B), any “fixed-image” is NOT gonna come with that Office app you need in a pinch.  Even if it does, when the pinch actually occurs and you’re at an airport at three o’clock in the morning–the work being due at 9 am that morning–you’ll find the situation has changed, and you need a * different * WINDOWS app.  Happens all the time.  You can just go online, and download the app.  It’ll be installed to the virtualized WINDOWS, and run just fine (you pray).  Often this will work.  If not, you can plead “trouble with my computer”(?).  But it’s one way of coping.  BUT KNOW that the downloaded app that “installed” to the virtual WINDOWS is just “installed’ to the RAM-disk.  The app is going to be virtual, and if you want to use it again after the computer has been re-started, you’ll have to download it again (or maybe load it from a backup?).  AND you’ll have to have enough RAM on hand to run it.  Not such an issue for most apps.  But it could be, depending on what it is.  SO ATTEMPTING TO UPGRADE THE RAM IS A GOOD IDEA IN THIS CONTEXT.  If you plan to go with this type of strategy.  RAM is not as expensive as it used to be.  If you plan to do it this way, I think I’d try to install as much RAM as the machine will accept.

So, * if *, on the other hand, you are goning to try to preserve the stock (“O.E.M.”) install of ms WINDOWS (98, 2000, XP, Vista, 7) to a computer, some pretty good general advice is given BELOW, in the remainder of this entry.  Remember that you need to take your own unique situation and use-habits into account–at least somewhat.

SO, AS FAR AS PRESERVING A MANUFACTURER’S STOCK INSTALL OF WINDOWS:

It is probably best to use Windows Disk Manager to reduce (“shrink”) the size of your Windows partition on your harddrive, then use Linux’s own installer to install Linux from a live media—such as a cd or USB thumb-key.  NOTE that I am not sure if Windows Disk Manager is available for versions of ms Windows prior to Vista.  If such is the case, I’d think it’d be safe to just use G-Parted—or let one of Linux’s own installer programs use it for you.  Windows Disk Manager is GUI.  BE SURE TO DO FULL BACKUP OF ALL YOUR PERSONAL FILES, AND THEN YOUR WINDOWS OPERATING SYSTEM ITSELF, WITH WINDOWS’ OWN BACKUP-AND-RESTORE-CENTER.  It is best to backup Microsoft Windows to one set of discs * before * you shrink its partition, and again * after * having done so.  This is a fact, because MS Windows is often much more difficult to re-install than is Linux.

This backing-up of the Windows operating system itself will take a long time, but you can be doin’ other stuff while you’re running backup-and-restore-center, pausing when it is necessary to insert another blank DVD into the tray (if you’re backing-up Vista or 7/8; XP will often back-up to a single disc).  And it’s worth it.  Because Microsoft is no longer in the habit of allowing computers to ship with that actual Windows-installer cd—like we used to get with a new computer.  Too many of us just copied/pirated it, it seems, and didn’t pay.  So the actual ms Windows installer-cd isn’t supplied much anymore.  (Though it is possible, of course, to go out and * buy * a copy of ms Windows on DVD, and install that to pretty much any pc.)

So instead of the Windows 2000 or XP cd—which used to be supplied with the new computer—you’re supposed to make a set of “backup discs”, and these will automatically orient themselves to your machine’s harddrive, its own particular serial number, UUID, & rel.  Which makes it harder to pirate ms Windows.  This doesn’t seem to have been much of effect, however, where the professional * pirates * are concerned.  The effect it * has * had, it seems, is to make it * harder * for us not-so-geeky Windows * users * to * re-install * ms Windows.  Which is too bad.  Because if you * use * ms Windows long enough—with or without messing with Linux—you’ll find it necessary to re-install it.  This is a FACT, and is related to the underlying files-system schema and structure of data-storage, which even Windows Vista and 7 inherit from earlier versions.  And nobody knows what virus might be released onto the web next week.

DO NOT SHRINK AN INSTALL OF MS WINDOWS TO TOO SMALL A SIZE.  Microsoft Windows requires more disk space than desktop Linux.  This has always been, generally speaking.  It is this way in-part because of the underlying type of kernel and file-system that Windows uses.  And you’d be advised to leave sufficient room for Windows updates.  I’m not sure about good ‘ol XP (I learned on Vista!); but Windows 7 nags you about the waiting updates for awhile, and then it will * make * you install them.  I shrank the install of Windows 7 Starter version on my new netbook to 29.5 Gb.  This was fine for the first 6 months or so—and be advised I wasn’t booting into Windows much anyway—my install of Linux Mint was working satisfactorily for me.  But eventually I found myself running critically low on Windows disk space.  My only way out was to backup everything, and completely destroy all three of my installs of Linux, and then the Extended-type partition that held them.  Then I used Windows Disk Manager to re-expand Windows 7 Starter to 41 Gb.  Then I booted it, just to be sure all was well.  Then I used G-Parted (from a live Ubuntu 10.04 thumb-key) to create a new Extended partition, and then some logical-volumes inside this, and then I ran a Live thumb-key of each Linux distro, to re-install my Linux desktops.  No data loss, but it took me like 12 hours, all totaled.  Time I could’ve spent doing something else.

When I used the default option to install Ubuntu to my full-size laptop (a Toshiba Satelite), I found that Ubiquity installer had left something like three-fifths of the harddrive to Windows 7 Home Premium.  This was for the best—at the time—as I was still learning Ubuntu (10.04) and about things like having a large, separate vFAT partition.  And so I still needed to do daily work from W7 for some weeks.  (I was still pretty * green *, where it came to computers, generally—otherwise it wouldn’t have taken me * weeks * to learn to work from Ubuntu 10.04:  It’d have been more like * days *).

That computer is still in a hall closet:  I’ve since migrated to an Acer netbook/”mini-lappy”.  Eventually, I intend to install a more current Linux desktop to the full-size laptop, as 10.04 will be reaching EOL in April of 2013.

When I do, I intend to make myself a vFAT partition in the middle of the big laptop’s harddrive.  And I’ll probably shrink Windows considerably—as for one thing I’ll be able to store all files I might ever need to work with in Windows on the vFAT.  Win 7 Home Prem on that machine is currently at c. 100 Gb., if memory serves me correctly.  But I probably won’t re-size Win 7 Home Prem to less than, say, 65 Gb.  Even 65 Gb. seems a little excessive, on the face of it.  (Remember when an operating system loaded from a * floppy * * disk *, and was measured in * kilo-bytes * ??)  There will be times when you’ll still be in need your install of ms Windows—if for no other reason than to access a printer at the Public Library.  Because certain printers * still * have a problem with Linux.  And there may also be times when you need to do some process with a file that Linux handles very differently, and the work is due on Monday.  Get it done in stinky ole ms Windows, and learn the Linux way later.  This might require that you copy the file to Windows’ own partition—depending on what the process in question is.  And so you might need some Windows space for that, too.

I suggest you do some research, Dude.  Read some of the forums-posts, and other info that’s available for free, online.  Because it’s better in more ways than one, if you don’t find yourself having to mess with this * twice *.

SPECIAL NOTE:  Unlike Linux (and most other free/FOSS os), ms Windows * insists * on only booting from a Primary-type partition.  (This can be got-around with techniques like “disk-swapping” in the bootloader; but we don’t need to bother with learning that, normally, just to enjoy some desktop Linux.  So I won’t cover that ground right here.)

You are allowed to have up to four (4) Primary-type partitions on a data-disk (harddrive, USB thumb-key, external harddrive, SSD, hybrid-drive, etc.).  Or you can have up to three Primaries, and what they call an “Extended” partition, and an Extended-type partition can be very large, and can be divided-up into many “sub-partitions”, called * logical-drives *, or * logical-drive * partitions *.  You will not be allowed more than a total of fifteen (15) partitions all together, on one disk.  If you leave your harddrive with two Primaries, then an Extended partition that you can make (with an easy graphical program like G-Parted, run from a live Linux disk—cd or USB thumb), well, this Extended can then have up to twelve (12) logical-drive “slices” in it.  Each of these is effectively a * disk * partition *, in its own right.  But if you need to have three Primaries, then any Extended partition that you make can have only eleven (11) logical drives.  Because the Extended-type partition that will be used to hold the so-called logical-drive partitions will * itself * be counted * as * a * partition *.  Linux (and most other free/FOSS operating systems) don’t care whether you’re booting them from a Primary or a Logical-drive.  Nor do they much give a crap whether the partition has been made “active”, or whether a boot-flag has been set (though I notice many installer programs for Linux still set a boot-flag anyway).  Somewhat of an exception to this is FreeDos + GEM, which I remember used to require that the partition be “active” (at least in the case of booting it from a Primary-type partition), and that the boot-flag be set.  In version 1.0, anyway.  And when I was installing it to an * old * machine.

If you think you need to set a boot-flag, this can be done with G-Parted; run it, make sure you have the intended operation set to the right “target”; then just use the “manage boot-flags” feature.

But the rest of the free/FOSS operating systems don’t seem to mind what the heck kind of partition they’re booting from, or whether it has been set as “active”, or about a boot-flag..  Usually only * Windows *  minds.

FURTHER—* VERY * IMPORTANT * :  Any time Windows’ own partition is re-sized, what you find is that its own boot-stuff needs to “register” the new size of the partition, the next time it boots.  So if you should re-size Windows, and then change your mind and re-size it again:  unless you boot it in-between, you will find that it will not boot again at all, until you re-install it from “bare-metal”, and this will require you to wipe-out everything else on the disk—your Linux install, your FAT32 partition with your saved videos and personal files (if you’ve made such a space); and everything else.  So the rule, really, is: “MS WINDOWS SHOULD ALWAYS BE BOOTED AT LEAST ONCE, ANY TIME ITS OWN PARTITION IS MESSED-WITH, JUST TO MAKE SURE WINDOWS IS OKAY, AND TO MAKE SURE IT REGISTERS A NEW PARTITION SIZE”.

7. Entry-7:  The best way [for us Noobs (?)] to solve the problem of the os not connecting to the web/internet, is probably to just try other distros, as live cds or live, bootable USB thumb-keys. Learning to figure-out things like DHCP and ARP, etc., can easily turn into a huge time-sink. You might also try installing a different (better ?) networking-app—like MadWifi, or *wicd*. The latter also handles Ethernet.  I’m not sure about the former.

NOTE THAT YOU CAN OFTEN USE THE ETHERNET-TYPE HARD-WIRED CONNECTION TO ACCESS THE INTERNET / NETWORK from the Linux desktop, until you figure-out how to get wifi working, or decide to just try another distro or build.

AND LINUX SELDOM REQUIRES THAT YOU USE SOME COMMAND TO CONFIGURE NETWORKING (such as the “ipconfig” command in Windows).  Most Linux distros will “ping” themselves to an * Ethernet * connection “automagically”, without intervention from you.
If you find you need it, the Linux networking command to research is probably “ifconfig”.

Recently, I came across a web-page that gives us indication of * which * WI-FI cards are supported in Linux.
https://help.ubuntu.com/community/WifiDocs/WirelessCardsSupported
This page may also link to tables and information as to * other * hardware * with descent or often * very * good * Linux support.  It is still good practice to refrain from taking this or * any * web-page merely * at * face * value *.  Even hardware issues can vary.  Do some confirmation of your own.  But pages like this are a * help *.

8. Entry-8:  ctrl + Alt + F1
This is a Linux manouvre well known to experienced users; it is as important in Linux as ctrl + Alt + delete is in ms Windows.  And it serves very similar purposes.  Yes, ctrl + Alt + delete will work in Ubuntu, and most other desktop Linux, to let you re-boot, or perhaps to reach your task-manager.  [Now, in Linux Mint XFCE 13, this command * locks * the * screen *—so that, for instance, you can be away from your workstation for a minute, and no passer-by can mess with it, without your password.  So ctrl + Alt + delete IS NOT, apparently, set to perform like it does in ma Windows, in each-and-every distro of desktop Linux.  Check the distro’s free online documentation and manuals, before you find yourself in need of this command:  that’s the best advice.  The command ctrl + Alt + F1 * will * almost * certainly * work * in * any * distro * of * desktop * Linux *.  So, if you can master * this *, you don’t have to bother with whether or not your distro’s default key-binding for ctrl +Alt + delete will take you to a task-manager, or what. ]

So a lot of us desktop Linux users know the use of ctrl + Alt + F1.  It’s important, because sometimes, desktop Linux doesn’t load/boot completely or properly.  This is also true of every other operating system on the planet—including ms Windows.  But in a Windows environment, a user deals with the situation a bit differently.

If you boot your Ubuntu or other Linux desktop, and the login-manager didn’t load; or you’re surfing the web and desktop Linux freezes-up:  the culprit is usually that the GUI didn’t load right, or has frozen because of ACPI issues (see below), or other hardware-related problem.  DO NOT JUST HOLD YOUR FINGER DOWN ON THE POWER-SWITCH, to shutdown so you can re-boot.  This is called—among other things—a “hard-reset”.  Now, I did many such “hard-resets” to my big new laptop (a Toshiba L515 Satellite) because I was too green to understand the hardware issue I was having with the machine, where it came to running Ubuntu (10.04), in a “dual-boot” arrangement with my existing Windows 7 install..  And I was irresponsible in not doing a sensible amount of research * beforehand *, online, from my Windows system, as to running Ubuntu on that make and model of equipment.  To this day, I cannot detect any damage—to the machine’s harddrive, or other componentry.  But I guess I could’ve damaged my harddrive—though this seems pretty rare nowadays.  Actually, I have yet to encounter anyone who’s damaged his-or-her computer by doing one of these “hard-resets” from a Linux session.  But it is still warned-against.

A MUCH BETTER WAY TO DEAL WITH A BAD-BOOT OR GUI-FREEZE OF DESKTOP LINUX IS  ctrl + Alt + F1.  (In most of the Puppies-Linux, I think it’s   ctrl + Alt + BACKSPACE.  )  This takes you out of the GUI, to an x-window—and to the Root-level (“Administrator-level”) of control.  It is a lower-level, more basic black-and-white environment, where you type letters, words, and symbols onto the screen to control the computer.  This allows for a “graceful shutdown” of the remaining Linux kernel.  You’ll then be able to start your machine back up, and boot into either Windows or Linux.  So what do we do in this black-and-white “x-window”?  Look at what it is prompting you for.  Type your login nickname (that we usually see when booting-up, and we see the graphical login-manager dialog window).  Hit Enter.  You’ll then be prompted for you password.  If you mis-type such things as your login-name or password in this environment, or any-such that the system doesn’t understand, you’ll be allowed to try again.  You’ll probably be allowed to try again any number of times.  Unless you’re using a Linux distro that came with more stringent security settings (like, say, maybe BackBox Linux, Backtrack, or Network Security Toolkit).  Just pay attention to the questions it’s asking you (the “prompts”).  Just as when you use Terminal-emulator, when typing a ** password **, your keystrokes will not be displayed—not even represented as dots.  They call this “the environment does not ‘echo’ my keystrokes”.  This is of course a security feature, to keep people from shoulder-surfing you, and memorizing your password.  So you just kinda have to mentally keep track, when typing your password, when in this x-window environment.  If you mis-type, you’ll probably be allowed to try again.

Of course, you have to remember your login-name—or have that info written-down somewhere, where you can easily retrieve it.  And then you’ve gotta type that password—without it displaying (“echoing”) your blasted keystrokes, and then type a command to shutdown or re-boot (which you just have to know, ahead of time—though there are only a few, and I’ve included these in the following paragraphs—and I am also uploading a file that covers this).  AND, once you’ve successfully done all these……the system will require you to type that password ** again **, so it will execute your command.  And THEN it will go ahead and shutdown / re-boot.

Now, we don’t have to hassle with this on a regular-basis—just maybe once in a while, when the Desktop Environment/GUI hung, or froze-up.  If you find yourself needing a procedure such as this one on more than a comparatively rare basis, then I’d say that you should probably try other distros, to find one that’s more suited to your hardware—or else investigate how you could make appropriate hardware changes.  Most distros of desktop Linux I’ve tried run just fine on my mini-lappy here [an Acer Aspire One—or “AAO” for short (type d255e)].  But it might not be the same with what hardware you’re trying to use.  And of course, the complexity/”geeky nature” of this ctrl + Alt + F1 procedure is a * MAJOR * REASON * more people don’t replace Windows with a free Linux desktop.  If the various hardware-makers (“vendors”) were able to be more forthcoming as to certain specs and other info, various desktop Linux entities (such as, say, Cannonical, The Linux Kernel Project, or The Linux Mint Community, et al) might be able to solve many of the remaining issues, so that users would be even less likely to have to resort to ctrl + Alt + F1.  Nevertheless, Linux is always developing (and more dynamically than Windows).  So it is perfectly conceivable that by the time you read this post, the issue will have been mitigated.

Once logged-in in x-window, you can type a command to shutdown the system, so you can boot-back-up from cold.  In Ubuntu, usually “sudo halt” (without the quotes, of course), or just “halt” (& then Enter, of course), will usually cause the system to “unmount” (to shutdown)—and in a proper fashion.  Without risk of damage to the harddrive.  You may have to type that password a second time, though.  Other commands to try would be:
“shutdown”, “shutdown -h”.  One of these will usually do the trick.  Consult your distro’s documentation, online (manuals), or research it with a search-engine (such as Google).

There are also some commands that’ll let you try to get the system to just re-load the GUI.  Sometimes this works.  Sometimes it doesn’t, and you’ll still have to shutdown the system, and boot-back-up from cold.  Linux isn’t perfect.  It has its quirks, and these have been around for quite some time.  If you can’t learn to live with them—via work-arounds like those I’ve attempted to describe here—then desktop Linux may not be for you.  For such commands, you could perhaps consult my document “L START THE DESKTOP FROM COMMAND LINE”, or your distro’s free online documentation.

9. Entry-9:  R.E.I.S.U.B.

This is also known as “shutdown with magic SysReq keys”.

Another possible way to avoid a hard-shutdown (holding down the power-switch) can be:
“Alt + sysrq (often this  ‘sysrq’ appears on the “print screen” key) + REISUB”.
The Linux kernel includes magic system request keys.  It was originally developed for kernel hackers.  However, you can use this hack to reboot, shutdown or halt the computer safely (remember safe reboot/shutdown == flush filesystem buffers and unmount file system, and then reboot so that data loss can be avoided).

This is quite useful when Linux based system is not available after boot, GUI has frozen, or after a X server crashed ( svgalib program crashes).  Or no display on the computer’s screen.  Sysrq key combo forces the kernel to respond it regardless of whatever else it is doing, unless it is completely locked up (dead).

Using further extension to iptables called ipt_sysrq (new iptables target), which allows you to do the same as the magic sysrq key on a keyboard does, but over the network.  So if your network server is not responding you can still reboot it.  Please note that Magic SysRq support needs to be compiled in your kernel.  You need to say “yes” to  ‘ Magic SysRq key (CONFIG_MAGIC_SYSRQ) ‘ when configuring the kernel.   (NOTE that this is already done in many builds of Linux Mint, and other variants of Ubuntu.  Ubuntu 10.04’s online documentation says that the Magic SysReq keys are already fully enabled, but * I * have never been able to make this command work in 10.04.)

Enable sysrq keys

By default it is not enabled on many Linux distributions.  Again, Ubuntu’s online documentation says that this feature is enabled by default.  Please check your distro’s documentation/manuals, to see if R.E.I.S.U.B. is enabled by default, or if you are required to config it manually.

In Linux Mint, I have found this “safety parachute” to be effective.

Your fingers almost have to be acrobats, though, to perform the required key presses.
I hold down Alt, then also hold down SysReq (“PrtScr”), and then, with those two held down, press r, then e, then i, then s, then u, then b.  Press these letter keys at one second intervals or so—the Linux kernel (which this is calling directly) needs time to register each command in the series, and then execute it.

Another Method:

Add or modify following line (as soon as new Linux system installed) /etc/sysctl.conf:
# vi /etc/sysctl.conf
Append following config directive:
kernel.sysrq=1
Save and close the file. Reload settings:
# sysctl -p
Save and close the file and reboot system to take effect
How do I use the magic SysRq keys in emergency?
You need to use following key combination in order to reboot/halt/sync file system etc:
ALT+SysRq+COMMAND-KEY
The  ‘SysRq’ key is also known as the  ‘Print Screen’ key.  COMMAND-KEY can be any one of the following (all keys need to hit simultaneously) :
•     ‘b ‘ : Will immediately reboot the system without syncing or unmounting your disks.
•     ‘o ‘ : Will shutdown your system off (if configured and supported).
•     ‘s ‘ : Will attempt to sync all mounted filesystems.
•     ‘u ‘ : Will attempt to remount all mounted filesystems read-only.
•     ‘e ‘ : Send a SIGTERM to all processes, except for init.
•     ‘h ‘ : Show help, indeed this the one you need to remember.
So whey you need to tell your Linux computer to reboot or when your X server is crashed or you don’t see anything going across the screen then just press:
ALT+SysRQ+s : (Press and hold down ALT, then SysRQ (Print Screen) key and press ‘ ‘) -Will try to syn all mounted system
ALT+SysRQ+r : (Press and hold down ALT, then SysRQ (Print Screen) key and press ‘ r ‘) -Will reboot the system.
If you wish to shutdown the system instead of reboot then press following key combination:
ALT+SysRQ+o
ipt_sysrq is a new iptables target that allows you to do the same as the magic sysrq key on a keyboard does, but over the network. Sometimes a remote server hangs and only responds to icmp echo request (ping). Every administrator of such machine is very unhappy because (s)he must go there and press the reset button. It takes a long time and it’s inconvenient. So use the Network Magic SysRq and you will be able to do more than just pressing a reset button. You can remotely sync disks, remount them read-only, then do a reboot. And everything comfortably and only in a few seconds. Please see Marek Zelem page to enableIP Tables network magic SysRq function.

For more information read official Documentation for sysrq.c version 1.15 stored in /usr/src/linux/Documentation/sysrq.txt and read man page of sysctl, sysctl.conf.

10. Entry-10:  There is a setting you can enable in most desktop Linux, which will cause the system to try to shut itself down properly if someone just quickly “taps” the power switch on the computer’s case.  In classic Gnome, I think it was under System > Preferences > Power Management.  XFCE desktop has a configuration manager window, accessible from the main menu (equiv of ms Windows’ Start button).  A little research (Google), or else a look at your distro’s manuals, will tell you how to access this.  I always run with it enabled.

11. Entry-11:  Ubuntu Restricted Extras Package:
THIS IS IMPORTANT.

Most current Linux distros come with about all the apps you’re probly gonna need—more-or-less—except sometimes for some multi-media codecs and coding.  [Remember that you probably had to install Java and Adobe stuff to any new WINDOWS computer, once you got it home from the store, and got it out of the carton.  Modern desktop Linux is no different.  Ubuntu, for example, has the “Restricted Extras” package in it’s default repositories, and this will cover all of what an ordinary desktop user will ever need, in the way of Flash, Java, and other “proprietary codecs”.  Linux Mint, and most of the other “Ubuntu variants” (i.e. PepperMint Linux, MoonOS, PinguyOS, AriOS, &tc. &tc.) * ALREADY * come with this “restricted” stuff installed.  Usually.  The Ubuntu people are prohibited from pre-loading it, because of copyright/re-distribution issues.  The way to check, to see if your distro already has these codecs, is to A) try to watch a video on You-Tube; if that works without the CPU reving-up to top speed, and the fan starting to run hard, then Adobe Flash is probly already installed.  B), next, check for the presence of OpenJDK (the Linux Java coding), g-streamer programs, and “ms fonts”.  Just use Software Manager/Software Center, using the “installed” tab.

[Some folks on forums will still tell you that Flash-playing functionality in Linux is always rough (and maybe even MAC also) :  well, * I * have never experienced this issue—at least not after I’ve installed the “Restricted Extras” to Ubuntu.  And I am one who has watched * a * lot * of video—both live in web-pages of all kinds, and as downloaded files.  It ** is ** still possible, though, that you may have problems with Flash-playing in web-pages.  Like I say in entry No. 2—sometimes there can be a hard-to-pin-down issue with Linux and your particular hardware.  It is also worth mentioning that sometimes there are just “Gremlins from the Kremlin”—mysterious bugs that can effect * any * operating system—even MAC, even * Windows *. ]

Note that you need not use desktop Linux’s Terminal * emulator * if you’re one of those geeky-types that is used to * real * msdos in a Windows environment; ctrl + Alt + F2 will take you to Linux’s x-window as * user *; ctrl + Alt + F1 takes you there * as * Root *.  And Root is equiv. to Windows’ Administrator.  No files with “openjdk” in their file-names?  Then your Java stuff is probly not installed.  If these codecs are installed, it will likely be all you need.  Playback of encrypted DVD requires a different piece of coding, easily checked-for, and easily installed from the web, if it is not already included.  There are some other, more “obscure” codecs.  I don’t think I’d bother with chasing after these, unless I had a related performance issue.  In that case, often a search for “post-install checklists” for your distro—or else Google:  “thenameofyourdistro perfect desktop”.  These search queries often yield informative results.

IF YOU’RE GONNA USE UBUNTU, OR XUBUNTU, OR LUBUNTU, YOU WILL NEED TO OPEN THE SOFTWARE CENTER, AND TYPE “RESTRICTED” INTO THE SEARCH-FIELD, AND THEN INSTALL THIS PRE-PREPARD PACKAGE (“Kubuntu Restricted Extras package”, or “Ubuntu Restricted Extras package”, or “Xubuntu Restricted Extras”, or which ever one is ** appropriate **), OR UBUNTU/KUBUNTU/XUBUNTU/LUBUNTU WON’T WORK RIGHT.

12. Entry-12:  ACPI:  Why Linux is Harder to Install to a Laptop or Mobile Device [and some (comparatively) easy workarounds]   Well, Daddio, we come to good ‘ol ACPI.  Blessed ‘ol ACPI.  What can I say?  This is a feature of laptops, netbooks, notebook computers, and anything that has * batteries *, and therefore requires * battery management *.  ACPI is “Advanced Control Power Interface”.  Or some jazz like that.  It is why desktop Linux is harder to install to a laptop or other mobile devices.  NOT TO MENTION THAT ALL LINUX DISTROS SEEM TO HAVE RATHER POOR BATTERY-MANAGEMENT–EVEN WATTOS.  Which is too bad.  A person can get around some of the battery-management issue by upgrading the laptop’s battery-pack, as most laptops only come with about one-half the batt capacity the manufacturer has created room for.  External power-supplies are also an option (especially the kind that have photo-voltaic attached, so that they continuously charge with solar/ambient light).  Frankly, i find I can almost always find a restaurant or coffee-house that has a seat near an electrical outlet, so I can plug-in my laptop’s power-cord.  ACPI is a more modern replacement for APM (“Advanced Power Management”), which mobile devices used to use.  Most problems with getting Linux to configure and run right on a laptop/mobile device are related to this.  [Although there are other hardware issues that can occur, too—particularly with attempting to install a Linux desktop distro to a laptop, or other device that is “mobile”.]

A good first step is to go to the site “Linux on Laptops”
http://www.linux-laptop.net/

Select the vendor/manufacturer of your laptop/mobile device from the menu, then follow/navigate to your specific model/series—or as close to that for which there is a page. This will help you to not have to repeat other people’s mistakes, because the people who write these pages have actually installed desktop Linux to that make/model.

BUT JUST AS IMPORTANTLY, a lot of us use “boot parameters”.  Just to confuse you, these also go by the names “boot arguments”, “boot options”, “boot switches”—and the Knoppix people call them “Cheat-Codes”.

“noacpi” (without the quotes) is one.  You type that at the boot prompt—the blinking cursor-bar, on the non-GUI (text-only) screen—when booting from a live Linux cd of Ubuntu desktop, or any desktop Linux cd.  But   noacpi   is like using a sledge-hammer to kill a fly; it turns Advanced Control Power Interface completely off (for that Linux session).  This renders your laptop as a tower computer:  there will be no battery management at all, should you not be able to find a place to plug-in at the airport, or your favorite coffee shop.  And battery level icon will be unavailable during a use-session booted with noacpi.  So when the battery runs out of electricity, the computer will just crash, it will “brick”.  Often the Linux kernel, and modern programs—such as DropBox 1.4.7 or 1.4.0 for Linux—will be able to prevent data-loss.  But not in every case.  The param   noacpi   can help you get going with your distro, but for a more permanent solution, research and experiment with other params.

Many machine’s BIOS will allow you to turn-off ACPI from there; but if you do, it will also disable ACPI for every other operating system currently installed to that machine—even Windows—until and unless you re-enable it.  So this will mean no battery management, in any os on that box.

Desktop Linux, historically, HAS NOT been as good at battery management as ms Windows.  A great deal of this seems to have been remediated in the versions of the modern Linux kernel—2.6 and now 3.x.  You may still not find it to be up to par with Windows.  However, this situation is changing.  Check your settings.  Often, desktop Linux distros * still * come with the setting appropriate to a * tower * computer *, by default.  Some tweaking may be in order.  Plenty of info for that, on the web.  Depends, I guess, on your usage habits, and physical circumstances.  I notice * I * can almost always find places to plug-in.  But you may not be in the same travel time-line as myself.

13. Entry-13:  BIOS boot:  Find out all you can about your hardware.  There’s the BIOS ** menu ** screen, and the BIOS *** boot ***-screen **.  Boot your computer from a cold-start, and go into the BIOS ** menu ** screen.  Your mouse won’t probably work in there, but you can navigate by using the arrow keys, and hitting Enter to confirm something.  Take care not to make any unintended changes, although most BIOS make it difficult to make a change to BIOS settings by accident.

The BIOS ** menu ** is usually reached by tapping the DEL, F1 or F2 key with your finger (at more-or-less one second intervals), early-on in the boot process (as almost immediately after pushing the power-switch, to turn the computer on).  Which key it is is usually indicated by the first or second boot screen, usually along the bottom.  This screen usually displays the name of the computer’s manufacturer, and often the logo as well. Often this screen is only displayed for a split-second or so.  On some machines, this “manufacturer’s screen” will not appear at all during boot, so you may have to resort to searching online for documentation about that make-and-model of computer, or resort to trial-and-error.  So some manufacturers/hardware-vendors don’t make this very easy for us (E-Machines is sometimes this way).  On many computers you can hit the Pause/Break key, and this will pause the boot process long enough for you to scrutinize the screen.  Hitting pause/break again will usually cause the boot to continue.
On some machines, it is a key other than the ones I just named.

The F- key that leads to the BIOS ** menu ** is usually (but not always) of lower numerical value (I.e. F1, F2, F3, F4, F5);  the key you tap to go to the BIOS *** boot ***-screen ** (from which you actually BIOS-boot) is usually of higher numerical value—F12, F11, F10, &tc.  (with Apple products it’s usually the “Alt” key—a.k.a. the “Option” key). BUT NOTE THAT IN * SOME * BIOS, * THIS * COMMAND IS NOT READY TO USE, BY DEFAULT.  It may need to be enabled from the BIOS * menu *.  On some HP computers, it is F9.  Where it comes to a lot of modern equipment (i.e. made after about 2008), it is gonna be F12.  But really, this seems pretty scatter-shot, from one make/model to another.  On some machines it can also be other keys, like DEL, ESC, &tc.  Sometimes you can hit the pause/break key to temporarily stop the boot process at this screen, and then tap that key again to let it continue.  This boot screen may flash on your monitor for only about ½ second—or perhaps not at all.  With some [mostly * older *] machines, the default BIOS settings don’t give you much of a time-window for hitting the stoopid key(s) to take you to the boot-menu screen.  And with some of ’em, you also gotta be attempting to hit the right key * at * just * the * right * time, while the bloody machine is booting-up.  Hallelujah.  So some of the BIOS out there are * tricky *:  this is mostly on older or cheap machines—but you never really know in advance, no matter what kind of info you think you have from stickers on the computer’s case, etc. You might consult your manufacturer’s home-page on the internet for information about adjusting your machine’s BIOS settings, if you can’t find the right keys to tap during boot-up.  On some really cheap machines, you’re sort of on-your-own—you may just have to experiment.

Some BIOS, bubba, are * more * difficult * than others, when it comes to booting Linux.  Some will only accomplish it with the aid of a geeky-person.  Some are just about impossible.  (q.v. The entry pertaining to Windows Secure Boot.)  Knowing what make and model of BIOS your computer has can be a tremendous help.  There are ways to find this out, from ms Windows.  (Go to Start > My Computer.)

When you are able to access BIOS menu screen, navigate around in there, and make notes about your BIOS version and manufacturer, your CPU make and model, whether or not it is single-core, dual, or quad-core, and other CPU specs, like the clock-speed.  Make notes of these facts, because even though you probly won’t need many of them to start using modern desktop Linux, it is good to write them down while you have the opportunity, just in case.  Note the size of the RAM, and any other specs that are available.  There is nothing like getting this information straight from the computer’s BIOS, because this is the most likely to be accurate.  When you are done in there, you should be able to find some button or tab that will let you “exit” the BIOS, at which point the machine will probly go ahead and boot into WINDOWS.  Once in there, you should be able to go to Start > Control Panel > System, and check this information against what is displayed there.  It will be a lesser amount of information, probably, less specific, less detailed, perhaps, than what was offered in BIOS, as to your hardware specs.  There are other ways, too, to find-out about your hardware specs.  A good way is to just run the Ubuntu live-cd, and open “Hard Info” from the menus.

The BIOS of some machines (especially older ones) does not have the option to boot from their USB hardware port/software interfacing.  This can be overcome by learning to make a special type of “boot-helper cd”.  You could look at the file “l boot from usb when bios is not able”, or Google the issue if it crops-up for you.  Sometimes this lack of USB-connection bootability can be remedied by just updating the BIOS’ firmware (the miniature operating system that drives BIOS).  This is sometimes referred-to as “re-flashing the BIOS”.  Better let a professional tackle this one, unless you’re already WELL on your way to being a hardware tech professional.  If a BIOS re-flash procedure doesn’t come off right, the machine will probly not be able to boot anything again, until-and-unless it is straightened-out.  An older-generation BIOS, that lacks the option to boot the machine from its USB hardware socket, can sometimes also be “helped-out” with various “bootloader tricks”; many of these are known to the Linux community at-large, and are discussed in forums.  There is a technique known as “disk-swapping”.  I think that I’d first perhaps check-out the   PL o P   boot-manager for Linux, as to its potential to enable BIOS to boot from USB, when the BIOS does not have that capability installed:  the   PL o P   site on the web has said that   PL o P   will let you make a “boot floppy” (on a cd, if you want), and then use * that * to enable BIOS-boot from USB.  Research it—you can Google something like “how to use   PL o P   boot manager to boot Linux from USB, when BIOS does not have the capability”.

Many Apple machines seem NOT to allow booting another os from the BIOS.  One should probably research this.  I myself have never had an Apple machine, and I don’t intend to mess with them.  My own (limited) experience seems to indicate that it is harder to move to a really good open-source, Linux usage-model, from most Apple hardware as a jumping-off-place (well, actually maybe this isn’t true anymore—if it was to begin with).  Yes,  there is the Yellow Dog Linux distro, but I am unclear as to whether or not one could single-boot it, if, say, you bought the machine second-hand (or were recycling it), and the previous owner had wiped OS/X off the hdd.  I’m lost here.  Other, more knowedgable ppl will post.  Then we’ll both know—you the reader, and I the (Lamebrain) writer.

14. Entry-14:  At least skim-over the official release-notes for your distro.  This is often only about a page-and-a-half, and is usually written in pretty plain English.  Your distro’s release notes are another thing Linux furnishes, to * help * you use it.  Release Notes tell you about things to watch-out for, which are * specific * to that particular distro, or “build” of the distro.

15. Entry-15:  WHAT IS A LINUX DESKTOP * RELEASE * ?
“Distro” is Linux slang for “distribution”—a distribution refers to who made it—Ubuntu, or the Linux Mint people, or the EasyPeasy Linux community, which makes EasyPeasy Linux—etc.  A release of your distro (Ubuntu, Linux Mint, whatever) is a specific model of that distro, just like you could buy a Chevrolet Caprice or a Mercedes Benz SL or E series, and it would have a further designation attached to the model-name (such as “Caprice 2007”) that denotes the time-frame of its manufacture.
So Ubuntu 9.10 tells you that this particular edition of Ubuntu was built around the spring of the year 2009, and released out of its beta-testing in October (the tenth month) of that year.  Hence, “9.10”—meaning 2009 + 10th month of that year for final, de-bugged version of this release.  [Really, dude, I prefer to still wait about 6 weeks or so, before trying to really * use * some latest release of Ubuntu—or pretty much * any * Linux desktop, for that matter.  Many (or maybe even * most * ) of the devs are * volunteers *, doing all this for us * for * free *.  Which is very nice of them.  Give ’em a little time, even after the release is officially made “final”, just to create and upload some more bug-fixes.  This is just common-courtesy. ]

Now, there are essentially two (2) main types of Linux desktop release:  A) the “rolling release” type, and B) the “fixed release” type.

A “fixed release” has no set schedule for when it is to be replaced by the maker’s next version of their product.  Automobiles from the major American car companies would, then, be said to be of the “rolling release” type, as the new models are always unveiled in the Autumn of every year.
Knoppix, and TexStar’s PCLinuxOS, then, are of the “fixed release” type—because the makers do not release a new build until it’s “ready”.

Cannonical, Ltd., and its distro Ubuntu (and many other of the various distros) have been criticized for following the “Microsoft way”, and releasing new versions “before they’re really ready”, just to meet their officially announced release-schedule.  There is some truth in this: * however *, I have been using Linux as my only daily desktop for more than two years now, and I have to say I find the * LTS * versions of Ubuntu more than adequate.  Which gets-around this “problem”.  LTS means “Long Term Support”.  These “LTS” releases of Ubuntu are every third release, and so come-out every two years.  Historically, each of these is supported with available updates (both software and security) for 36 months (three years), from the day the release was made “final” (that means certified as no longer in the  ‘ beta ‘ phase of development).  This, then, means that there is a 12-month period of overlap, between the end of updates-support in one LTS version, and the beginning of another.  This allows you a whole * year * to play-with the live-cd of the new LTS Ubuntu release, so you can adjust to any changes (like changes the makers may have had to make to the GUI, other stuff).

The most recent LTS Ubuntu (12.04) will be supported for a whole * five * (5) years, which is a really long time in computer-years, so none of us should complain about that.  The LTS versions of Ubuntu are often more stable, better documented, and just “better”, than the “tweeners” (my term for those that fall * in-between * the LTS releases).  And these advantages are often transmitted, and carried-over, to the various “variants” of Ubuntu (a’ la Linux Mint, PinguyOS, &tc.), which use Ubuntu as a “basis” (or “upstream” as we say in Linux-lingo).  And so a release, or a “build” of a release, of, say Linux Mint, that has been based not just on Ubuntu—but on an * LTS * release of Ubuntu—will often be more stable, and have fewer bugs, than a release based on one of the “tweener” releases.  And it is easy to find out which version of Ubuntu a particular variant release is based-on, just by Googling.

Modern desktop Linux has come a long way.  But when many of us try to accomplish something more complex than the “usual” tasks—i.e. browsing the web, listening to music, or even software development—which Linux is known-for—into such things as gaming, music production or video production….. things can start getting complicated in a hurry.  Occasionally, there are still glitches (usually minor) even in the most polished distros out there.  And this is embarrassing, in this day-and-age.

If you’re used to Windows XP (or, perhaps more-to-the-point, Windows 2000), I don’t think you’ll be much bothered.  But if Windows 7 Home Prem has got you spoiled (or ‘Droid, or OS/X), you might be taken a little aback.  Well, desktop Linux is still—for the most part—FOSS.  That is the mind-set:  Free Open Source Software.  “Stallman-ism” (after Richard Stallman, founder of the GNU project, upon which the Linux desktop still depends for many of its apps:  that’s why it’s properly referred-to as “GNU-Linux”).  And that’s the principal reason desktop Linux can be said to be “dynamic, yet unchanging”.  Distros and releases come-and-go, but desktop Linux * itself * seems never to much change its character:  even the best releases have some minor glitches.  And they probably always will.

NOTE that there is a further dot + numeral at the end of the release number, where it comes to LTS releases of Ubuntu:  some people call this a “point-update”.  Ubuntu 10.04.1 denotes that this is the April (4th month of the year) release, in the year 2010; the “.1” tells us that this is the first “build” of this release, but 10.04.2 will be the build for the second year of this one, because it’s supported for three (3) years.  Cannonical, Ltd. may add further of these “build” designations, where it is deemed appropriate by them.  You will also notice (if you check it by your Terminal) that after so many updates are installed by you, your distro’s release designation will have changed (like from, say, 10.04.1, to 10.04.2).  And eventually 10.04.3, 10.04.4, and as high as the Ubuntu developers take it, before that release reaches EOL (“End Of Life”), and support for it finally is discontinued.

Ubuntu will, of course, continue to function, even after software support for that particular release has been discontinued.  You will be occasionally prompted (well, just at a boot-up) to upgrade to a current release, because that’s what’s recommended.  See the entry pertaining to upgrading your release—No. 44.   (Do not use the “Upgrade button”; I and others feel it is not perfected yet.  And for godssakes, backup everything first, just like you’d do if upgrading from Windows XP  to 7.)

16. Entry-16:  Sometimes you can use an older release, to get desktop Linux to work well on your hardware. More than a few people resort to this.  But you won’t be able to get the current security-updates for your system.  And in order to install the newer versions of many apps, you may need to do the “compile-from-source” install method—though this is not as daunting as it sounds (see entries 24 and 27).

17. Entry-17:  WHAT’S THE BEST DESKTOP LINUX FOR A NOOB/NEWBIE (that means “a person new to the experience”) ?

I tend to think (firstly) in terms of trying to stay with the Debian Linux-based distros.  The Debian “upstream” makes things easier in a number of ways, though I won’t go into these right here, because it’d be beyond the scope of this article.  This is why (arguably) the developers at Cannonical, Ltd. start the “stream” that is their flagship product (Ubuntu) with Debian—because Debian is already easier to use (arguably, I say) than the other “base” Linux distros (like Slackware or even Fedora).  Debian Linux is * Debbie + Ian *; the founder (a guy called Ian) named the distro for his girlfriend (Debbie), and himself.  Debbie + Ian = Debian.  Debian Linux is headquartered in California, and is a non-profit organization.

Cannonical, Ltd. downloads this Linux system, and then adds its own improvements; the result is called Ubuntu—an African word that means something like “peace and goodwill”.  The improvements they see fit to add make Ubuntu more “bootable”—easier to boot on a wider variety of equipment, and also easier to use and command from a graphical (point-and-click) environment.  And easier to use and configure, generally, than even Debian.  Easier for us not-so-tech-savvy types to use.

Most desktop Linux distros are, in turn, base3d on Ubuntu [Linux Mint, PeppermintOS, PinguyOS, AriOS, and the pre-Statler versions of Crunch Bang, and others too numerous to mention].  Ubuntu is also perhaps the best documented [that means * online * free * instructions * and * manuals *—but ** I ** also consider the definition to include ** a ** large ** user ** community **, with ** lots ** of ** online ** forums ** posts **.  Otherwise, I might consider * Arch * to be the best documented.  But * online * manuals ** plus ** forums * communities * ** to ** me **, means ** “best documented” **.  (Sorry, Arch.)  But, as with all things Linux, this is (at least somewhat) subjective, and a matter of personal experience and preference.

There are also other distros that are arguably good for beginners.  (And which perhaps are just as good with which to continue).

MEPIS Linux is Debian-based, but is NOT based also on Ubuntu.  And it makes a good desktop. Also in the same category is Knoppix—though, while it can make a nice desktop (especially the DVD-build), it’s really supposed to be run * live *—as a cd or DVD, or as a * frugal/PMI * install; full, un-compressed install of Knoppix to a harddrive is discouraged by the makers—though this * can * be done, with Knoppix’s on-board default installer:  this “Knoppix installer program”, however, is non-GUI, and should probably only be tried by a person * knowledgeable * in computers and the Unix/*nix filesystem.  So most of us run Knoppix from a live-disk, not as a “full install”.

Also there is Puppy Linux.  This one is especially good to run as a live, compressed, read-only filesystem, without making changes to the harddrive.  It comes with a * lot * of custom, “Puppy-original” wizards and (GUI) configuration-tools, to make your Linux desktop experience easier.  Like Knoppix, it can be installed to the harddrive, but this is not necessarily encouraged.  Frankly, since say somewhere around 2008 or so, it probably doesn’t matter so much, because it has been tweaked to make this more, uh, well, “Linux/Unix-Kosher”.

Puppy Linux has no upstream-basis; it is not really based-on Ubuntu, Debian, or anything else.  Puppy Linux is the original creation of Barry Kauler, a retired professor, and it is based on no more than the Linux kernel, and many original programs written by Mr. Kauler himself.  One last comment upon Puppy, here:  in trying to use it as a desktop, I have to say I find the range of software available in its “default, ready-to-download-and-install” repos (repositories) to be in rather a narrow range.  Yes, Lucid-Puppy * can * install from Ubuntu’s repos; but I found that I could only access a rather limited number of the softwares in there—not all 30,000, like Ubuntu and many of its variants let you do.  Nor could I ever manage to get apt-get working fully, so I might install some of these programs by means of the command-line.  This might just be * my * own incompetence and lack of “tech-savvyness”.

I’ll finish my comments respective of Puppy by saying that it is a distro “like no other”—its filesystem is layed-out in a most innovative manner, though this is not readily apparent to the user; indeed, it is * transparent * to users:  it’s made that way for the * safety * and * convenience * of users.  Puppy has pretty good free online documentation, and descent forum-support—by which I mean there are enough users of it that regularly post (answers) on Puppy forums, that you are reasonably likely to find an answer, if you have a question.

We come next to the RPM-oriented distros.  RPM is “Redhat Package Manager”.  In Linux, a package manager is the method to install a software and/or update the system, and RPM differs (internally) from Debian/Ubuntu’s “apt-get”/Software Center paradigm.  This difference used to be rather a ** bigger ** deal **, too.

But not so much anymore.  Nowadays, you can often even install some software coded for Debian (perhaps with some help from a small “MiTM” program) to an RPM distro, and vice-versa.

Some of the “RPM-oriented” distros can be fine for a desktop Linux noob/newbie.  I guess if I had to try to name them in order of ease-of-install and configuration, maybe I’d say, starting with the “easiest”:  PCLinuxOS (many sub-builds available from the maker, with many different GUI s (Desktop Environments) from which to choose; myself, I think I’d go with the LXDE); then, Mandriva/Mandriva One; then, maybe, Fedora.

CentOS * can * make a good desktop (with some configuring), and has a * seven-year * life-cycle (release cycle); but it is really more of a “professional’s tool”.

I guess that about covers the “noobie-friendlies”—at least ** in-brief **, without getting too long-winded (this particular document is supposed to be the “bare-minimum” a noobie should know, before getting into desktop Linux).  I apologize in advance if I left-out anybody’s favorite desktop Linux distro.  I’ll close this entry by saying that —** for ** newbies **—I think I’d stay away from Slackware & related, as well as Gentoo, and even Arch Linux; these are probably best suited to a person with some Linux experience, and a fair amount of ** computer ** experience, generally.  This is also probably true for the various versions of BSD operating system, which is the * other * big Unix-based (and free/FOSS) operating system out there.

Google’s   ** Android **   system * does * run * on a Linux kernel, but it’s a proprietary fork, made by a corporation known for * other * things.  Android’s version of the Linux kernel is not fully open-source:  not all of its code is available for public inspection, unlike Ubuntu, and the other “major desktop Linux distros” I have mentioned.  Further, if you read its EULA (End User License Agreement) carefully, I think you’ll find that it says somewhere in there that your online movements and activities will be tracked, or some such; but I’d have to check that.  Android, as you probably know, was built for the smart-phone market.  But yes, you * can * boot and run it on a netbook, tablet, or even a laptop—if you can get the internal hardware of the machine to co-operate—which is also an issue with regular (FOSS) desktop Linux.

If you decide to use true Ubuntu, instead of one of its many variants (“forks”, i.e. Linux Mint, PinguyOS, etc.), I don’t think I would expect it to work as smooth as one of the variants that has had a reputation for smoothness (like for instance Linux Mint).  Regular Ubuntu is apt to be a bit buggy—even the LTS releases.  But it’ll be faster.  Most of this “bugginess” has to do with the GUI not loading properly, or with hardware issues, or booting it off a disk it shares with ms Windows.  There can still be a few bugs attributable to Ubuntu itself:  however, in the previous LTS release (10.04), I have yet to find any.

(See the entry on “Dual Boot Headache”, below).

But the underlying Linux kernel almost never crashes—* ever *.  And that means no lost data.

Workarounds found by a little web searching (many of which I’ve tried to include here) will get you through.

Windows isn’t perfect either.  Let’s draw a breath, and remember Vista.  Or more-to-the-point Windows Me (Millennium).  And Win 8 looks like it’s shaping-up tp be a rather * difficult * default interface to get along with.

Ubuntu will run with lower system requirements, and has its own (pretty comprehensive) documentation.  And it is * easy *, ** easy ** to install, re-install, un-install, compared to ms Windows.  Once in a while, it won’t load its GUI properly, and you’ll have to do   ctrl + Alt + F1   to accomplish a proper shutdown.  Or try ctrl + Alt + t, and when the Terminal appears, type “sudo halt” (without the quotes), then use your password when prompted.  There will be other (usually minor) annoyances.  This just seems a fact of Ubuntu.  But learning to remember and use a few work-arounds won’t kill you.  It will probably be good for the development of your computer skills, and life-skills as well.

The brand-new (at the time of this writing) Linux Mint build “Linux Mint 13 XFCE” could be another good option:  somewhat higher system requirements, but very stable and easy to use.  XFCE Desktop Environ isn’t MAC OS/X:  and it is not quite as stable as Linux’s good ‘ol standby, the now deprecated GNOME 2.x.  But I’ve had only one problem so far—a freeze-up of the GUI login-manager after I had closed the lid of this netbook/mini-lappy to go use the restroom; I was tired, and mis-typed while logging back in, and it froze GUI.  But that hasn’t happened since.  XFCE is said not to be as stable as GNOME 2.  But it * almost * is.  And it has been around for years, runs GNOME-oriented and KDE stuff equally well, and it is relatively well documented:  there are plenty of XFCE hacks posted online, and XFCE runs well on top of Ubuntu and its variants.  These factors make XFCE a good DE for desktop productivity and work-flow.  XFCE DE is arguably the best Desktop Environment that’s available for Ubuntu and Ubuntu-variants right now, as I type this in February of 2013 (unless maybe you want to go with Cinnamon DE, or learn Enlightenment E-17).  But NO, unfortunately, XFCE is not as good (yet) as good ‘ol GNOME 2.x was.

Linux Mint 13 XFCE build has been assigned a sort of “separate department” by the Linux Mint developer community, which means it will be treated as an equal in the LM family, and not as a “stepchild”.  Linux Mint’s new XFCE build is also based-on Ubuntu as an upstream, and so will likely benefit in terms of Ubuntu’s lauded ability to boot where others can’t.  This “build” of Linux Mint runs very well on my netbook, and seems good for productivity on equipment with a 10.01” screen and larger (diagonally measured).  If you’re wanting to boot desktop Linux on something smaller (like maybe a tablet that’s not much wider than a pop-tart), you probably * will * want * something like Ubuntu’s new Unity interface, or perhaps try PeppermintOS.

18. Entry-18:  Dual Boot and “Dual Boot Headache”:

“Dual-booting” is just having Linux installed to a space (partition) on your harddrive, where you also have Windows installed to it’s own designated space (your Windows partition).
Sometimes, for reasons that can be a bit hard to pin-down, this “dual-boot arrangement” causes difficulty booting Linux after the first time, and then also with booting Windows.  If I had to speculate, I think I’d say that (perhaps) a certain well-known software corporation that shall remain nameless may have put something (or things) in their software stack that causes the two systems to not get along with one another, when installed to the same harddrive.  But I won’t speculate, or accuse anybody of anything.

If you seem to be experiencing the dual-boot headache thing, I guess I’d try booting Linux with ACPI turned off, and Also there’s a setting you can adjust, so that the desktop environment (Unity, XFCE, classic GNOME, or most of the rest) will render the desktop as “2D”, instead of with “3D effects”.  Sometimes this seems to help, for some reason.  You may be able to get this functionality back later-on, after having solved the underlying difficulty with your dual-boot arrangement.  Turning off the Compiz cube also, if it is enabled by default.

And I’d also be sure and visit the site “Linux on Laptops” (at the aforementioned link—or just Google it), if installing to a lappy.

If I had it to do over again, I’d probably still do the hdd install, if for no other reason than I like to do some (very moderate) experimenting.  But there’s an easier alternative.  Desktop Linux to-day can be installed and run from a USB thumb-key.  There are several ways to do this, which I have tried to detail in my blog, in other documents (see “newbie get started and migrate to linux desktop tip by tip”).  Another worthy option is an external, add-on harddrive, or a micro-drive (mini-external-harddrive).  If you have the money to buy an external harddrive, many times this can be a good way to get around the headache some computers have, when it comes to dual-booting Linux from the same drive which contains MS WINDOWS.  But it’s not an absolute cure—it doesn’t always solve the problem—if you have the problem in the first-place, which a lot of machines don’t.  Mine (a Western Digital Passport USB-powered external) only cost about 100 USD, at the time of this writing.  But that can be a lot if you haven’t got it.  It is probly also true that an install to an external harddrive is easier to undo or re-format, than the traditional dual-boot setup.  Provided you are able enough with graphical disk-tools that you don’t accidentally modify the wrong drive or partition—see the tip having to do with installing Linux, further-on in the document I mentioned.  (Or consult entry No. 25 here.)

Note that these type of add-on drives were not invented with the intention that they would be on and running all the time, but were rather intended more for things like storage and backup.  However, technology advancing as it does, many people on long flights now seem to have them up-and-running for many hours without issue.

Remember that an external harddrive has no built-in cooling, so an external way of helping it dissipate heat may be advisable.  (Like maybe sitting it on a laptop cooler or netbook cooler).  As there is no protective chassis and computer case, anything that protects against unexpected vibration might be a good idea too.  Perhaps just operating it while it lays on one of those gel-filled vinyl masks people sometimes wear— across the region of the eyes to soothe and cool the blood vessels—might be an idea.  Such a thing will absorb and dissipate a fair amount of heat, and is good at dampening vibrations, such as those that can sometimes occur in takeoff and landing.  Drugstores often have them to sell.

It is also possible (depending on the hardware) to remove the harddrive from a junk computer and physically “mount” it inside an enclosed or semi-enclosed “chassis”, which you could make yourself.  Some of what the construction trades often call Luan (Loo-awn), a.k.a. “doorskin”, or even odds-and-ends that may be available could be good to use for this.  Be sure it has adequate “vent-holes”, so it can dissipate heat.  A link pertaining (somewhat) to this is:
http://www.justlinux.com/forum/archive/index.php/t-148499.html

And let’s not forget about VirtualBox and VMWare.  These are just two of the better-known programs which allow you to run desktop Linux as a “virtual machine”—an imaginary  ‘other computer’ * inside * the  “imagination” of your own WINDOWS operating system—and this just by downloading and installing either VirtualBox or VMWare to your WINDOWS system.  Ubuntu orients this as the “Wubi method”; Linux Mint orients it as the “Mint4Win” method.  This seems preferred by many people, and is easy to un-do (just un-install it from Windows’ Control Panel, if you don’t like it).  However, some people report problems with this method when they try to get Linux to copy files to another disk (like a USB thumb-key) or to “surf the web” with its browser.

I don’t know much about this “VM” method of using desktop Linux—my own [limited] experience runs toward the other methods I have talked about.  As with * any * alterations to your ms WINDOWS system, you should independently research this method ahead of time, and set a restore-point, if you intend to try it.  This is an old rule.

Start out with live-cds.  Or live USB thumb-keys:  most USB-installers for desktop Linux produce a “live-disk” type install—but it really doesn’t matter if it’s “live” or “full, traditional hdd-type” install:  desktop Linux won’t make any changes to your harddrive unless you authorize it.  Live cds/DVDs and live USB thumb-keys do not make any changes to your harddrive, as they run wholly from the computer’s RAM.  This is what I did.  Boot one of these “live bootable Linux thumb-keys” and play around with it.  Or use the cd/DVD method.  You could just get a 10-pack of cd-R s at your local discount-store—they’re basically cheap nowadays.  Frankly, so are USB thumb-keys (“jump-drives”).  But it is IMPORTANT to spend the extra few bucks, and get the Quality ones, because we’re going to be playin’ around with an * operating system * here.  And if * that’s * poorly copied because of cheap discs or other media, then all bets are off.

19. Entry-19:  You might wanna watch-out for Microsoft/Windows’ * “Secure Boot” *.  MS reportedly already has this thing in the product-pipeline, for bundling with Windows 8.  Its stated purpose is something to do with protecting Windows users from hackers, and unauthorized boots—like from the machines’ BIOS.  At least that is my limited understanding, at the time of this writing. The Linux community seems to be working on prospective solutions to this, in supposition that it * may * be a boot-issue with Linux.  I guess we’ll see.  Some Googling about this can help you, if you suspect it is stopping you from booting other os.

20. Entry-20:  Alt + F2

Holding down Alt, and hitting F2, is the (rough) equivalent of doing Start > Run in Windows XP.  Now, this functionality in the Linux world is usually a function of the GNOME Desktop Environment—not of Ubuntu itself.  But it seems to also work fine in XFCE, or Linux Mint 12’s “GNOME 3 + MGSE” default desktop environment, and many others; and the feature seems to have been “fixed” in Ubuntu 12.04’s Unity Desktop Environ (it was said to be “broken” in Ubuntu 11.10).  Alt + F2 is a program-launcher, and more.  For a list of commands, see my document “l alt plus f2 commands list” , or just do some Googling.

21. Entry-21:  Don’t use the Linux * Computer Janitor * feature—it isn’t perfected yet.

22. Entry-22:  There aren’t any Linux viruses “in the wild”, at the time of this writing.  But it is best to use good networking practices anyway.  Even if you feel pretty safe using desktop Linux, and perhaps get pretty bold, where it comes to opening files and webpages you were afraid to open before, you still don’t wanna become like Typhoid Mary (Google that), and pass-on some strings of infected dos-code you got from a webpage, to your friend’s computer via an e-mail.  It won’t make your own * Linux * install sick; but it could knock your friend’s Windows computer for a loop.  So a good many of us Linux desktop users scan questionable files that we intend to attach to an e-mail with Clam-AV-for-Linux, just to check that file.  The major Windows anti-virus vendors nowadays also have available versions of their product that will run in Linux, to download to the Linux install and use, free-of-charge.  ClamAV—unlike some of the others—does not auto-update itself; but you can solve this by setting a reminder for yourself, or else set a task-scheduler to run “fresh-clam” from a Terminal window:  KAlarm and/or Session Manager are usually good equivalents for Windows’ Task Scheduler.  All it takes is a right-click on any file in-question, and you get the option to scan it with your anti-virus program, just as in Windows.  And just as in Windows, you should not have more than one anti-virus program installed at the same time—this can cause them to interfere with each other.

23. Entry-23:  Linux is easier to run as a live, compressed file-system, than Windows or Apple/MAC.  This has advantages—especially in the area of privacy/security.  And it can nowadays be set-up wholly by means of graphical/GUI apps, which can be run from either ms Windows, or from Linux itself.  If you can become familiar with running Linux as a live/compressed file-system (like from cd/DVD, USB thumb-key, or frugal/P.M.I.-type install), then:
Each reboot of this type of running is the same, effectively, as a reformat/re-install/fresh-install.  Why?  Because instead of loading needed files to the RAM from an uncompressed partition in the harddrive, a live file-system mirrors these files into the RAM via its de-compressor, as they become necessary (Google .sfs for Linux, Unionfs, and aufs for Linux).  As you open one app and close another, the mirrored system-files just go “poof”, out of the RAM, and disappear.  They can be called again, from the fixed, read-only file-system image on the cd/DVD, or frugal/P.M.I. install partition. Settings and changes (like to your * personal * files) have to be remembered in a special “persistent-save folder”, which is best created at the time you are creating the bootable live media—if you need such persistence at all.  This “persistent save folder” is mis-named:  the way most are set-up, they are really a separate, dedicated partition, formatted to something like Linux’s .ext2.  If you want persistence, it is probably best that it be kept to a minimum, as these persistent folders seem to slow the system more and more with each additional mega-byte, so it is often recommended that one first re-master the distro (with something like RemasterSys), and * then * create the bootable arrangement.  That  way the most of the customization as possible gets put into the bootable custom .iso image, and fewer changes are left to be saved in the persistent folder.  Of course, if you can live without the persistent, then your surface-area for attack (black-hat hackers) is even smaller, and is * live *—not a writable file-system. Yes, arguably you could get hacked in the persistent, but it seems as though it would be * hard * to accomplish—certainly * hard * for most malwares, which are usually “bot” (read “robot”) programs, as these almost always expect to meet with a “real” filesystem— * not * a “virtualized” one.  It is also true that such Persistent-Save directories (really a type of partition, as I said) can be * encrypted *.  And of course, if you can live without a persistent-folder, then it will be even more secure.

You also get a new MAC address and IP address each time you re-boot a live, compressed instance of any Linux desktop.  This will defeat a great deal of tracking, and corporate robot-programs keeping track of your web-surfing habits.  As for somebody hacking you in the RAM-disk (where the de-compressor opens files), again, most hacker-bot programs expect a “real” filesystem, rather than a “virtualized” one.  And if you set a timer to remind you to re-boot hourly, most malware will not have had time to work, before being flushed-out of the RAM with the re-boot.  What Unix malware there is in existence can’t attack Ubuntu anyway, because it’s aimed at * servers *, and the Ubuntu desktop has no server stuff running, by default.

It will give you protection from rootkits (a serious type of malware—though this usually targets * servers *—not * desktop * users *.  There are some rootkits that target Linux (servers), and * many * more * of * these * that * target * Windows.  And the impression I get is that the Windows ones aren’t as discriminating, where it comes to only picking on you if you’re running a server.  From the way desktop Ubuntu is set-up, by default—with all software ports closed—it would seem to this writer that it would be * harder * for it to get rootkitted, than the average Windows desktop.  This default setting in desktop Ubuntu does not, however, interfere with normal opereations—you can still get on it and surf the web, and download music, &tc with Ubuntu desktop, just like it is.
It is a protection against identity-theft.
It is a protection against spyware.
It is difficult for black-hat hackers to break-into.
It is a protection against corporations (or others) tracking you.
Due to the fact that this way of using an operating system is * read-only * (as far as the system files themselves), if you are a paranoid/privacy-issues person (like me), then this way of running will save you from long hours spent trying to understand things like firewall configuration, or sandboxing, intrusion-detection, log-file reading, perhaps also encryption (?—depending on your purposes in using a computer), and other security measures systems administrators take, for “full-install” file systems.
It will defeat most software key-loggers (especially if you have the sudo utility set-up in such a distro and invoke it using a salted text-file, if you are using it with persistence configured—see appropriate files in my blog; or, better yet, you are able to live without having a persistent-save folder at all.)  * Hardware * keylog-devices are another matter.
Hourly re-boots give you the maximum protection.  This is because, as the black-hat hacker’s software key-logger has begun to install and snoop on your system (from that script you got from a web-page), it will get flushed-out with the re-boot, because all the system-files in a live distro are read-only—and also they are * ephemeral *:  when an app closes, its virtual coding (and anything that may have attached to it) goes “poof”, into the aether of the universe; it will be “facsimiled” again, from the appropriate track on the live boot-media (cd/DVD, USB thumb-key, or frugal install partition on a harddisk) when the app is called again.

You will probably be able to afford to be “bolder”, in your web-searching habits, even than you could be in a regular Linux install.

Many political dissidents world-wide use this type of software arrangement.

24. Entry-24:  There * is * no single method to install a software in desktop Linux.  But there are easy GUI interfaces to help us.

The easiest way is by using Ubuntu’s Software-Store/Software Center (Software Manager, if you’re in Linux Mint).  This is a lot like Windows’ Control Panel.  But it will * add * a software that you want, as well as un-install a software program that you decide you didn’t want.  In desktop Linux, it is generally best never to un-install a program that was already there when you downloaded the distro.  As an aside, it might be useful to mention that Software Center is one of the more resoursce-intensive graphical utilities installed by default, to a desktop Linux system.  Firing-it-up will pull on your RAM and CPU more than the rest of the stuff in your distro, usually.  But because Ubuntu has such low hardware requirements to begin with, this is seldom an issue—unless you’re wanting to run desktop Linux on an old junker.  Well, desktop Linux has a very solid reputation, where it comes to rehabilitating old junkers, that somebody rescues from the recycle-bin.  To learn more about some special hacks, and related op information in this area, just Google it.  I have also tried to furnish much of this info here, in my data-base.

It is always best to run Update Manager, before installing a new software.  I usually forget, though, and it does not seem to have hurt me, yet.  But try to remember.

Where un-installing software is concerned, it’s de-preferred to un-install something that was already present in your distro, when you downloded it, because it could throw-off your Linux * dependencies * libraries *.  This is similar, conceptually, to .dll files in Windows.  But Linux’s dependencies are touchier than .dll, for reasons I won’t go into right here.  It is true, however, that many of the methods of installing desired software will check, “automagically”, for dependencies-issues, and at least inform you, before you give final approval.  Such as these would be:  Software Center (Software Manager in Linux Mint; GUI), Synaptic Package Manager (GUI), apt-get (Terminal).

Other methods of getting softwares would include (for Debian-based Linux):  using your distro’s Software Center/Software Manager to access NON-default repos (“repos” = “repositories”); you must first do a procedure(s) to enable getting from a NON-default repo, and the practice is depreferred by Cannonical, Ltd., the makers of Ubuntu, and most of the other distro-makers.  The reason is because  these repositories may not have been thoroughly vetted by the distro-makers, or authorized persons who work with them.  And so such programs might contain some un-reviewed mistakes in the coding, that might lead to an issue with Linux dependencies libraries—or–some might say—even some malware (“malicious code”).  Well, I can find but scant evidence of people inserting malicious code into Linux apps—even those from “PPAs”/”non-default” parts of the repository “metaverse”, on the World Wide Web.  It’s worth remembering the whole reason for the “default Repositories” in the first place:  it is to protect us users from dependencies-library issues; but also against malware in Linux.

Next, there are further methods:  Synaptic Package Manager (GUI), apt-get utility (Terminal), .deb files (similar to Windows’ .exe; less secure?), w-get (Terminal), .tar and .targz (a type of compressed file; use Linux Archive Manager—GUI), and compiling-from-source (easier than it sounds).  The program “G-Debbie Installer” is a GUI Linux “wizard” for helping us install .deb files.  Look for it in your system.  If not present, it can be installed.

Can’t find the right Linux app, to do what you need, no matter how much you Google it?  Search in Software Center, and in Synaptic Package Manager.  There’s a built-in apps search engine in each, and these will sometimes yield results that the regular internet does not.  * Then * you can Google it for some more reviews/instructions, if you want—because you’ll know what the exact name of the program is.

A final note I will render here, as to software/apps in Linux:  many desktop Linux distros have good rapport, as I just said, with old machines that are destined for a landfill.  If the system requirements for the distro are too high for the old junker, often the distro’s makers offer a “lighter-weight” version of their os, tweaked to require lower hardware resources.  Sometimes * a * lot * lower *.  (Linux Mint Fluxbox comes to mind:  LMF is ready for download, with no additional manipulation necessary, on the part of the prospective user.)  * BUT * THIS * IS ** HARDWARE ** BACKWARD-COMPATIBILITY *.  Where it comes to backward-compatibility * with * its * own ** SOFTWARE **, Linux is a lot more like ms Windows, with respect to * hardware *.  In Windows, often you have to just buy a whole new computer, when it’s time to upgrade to the next version of ms Windows.  Because the system requirements are usually higher for the next version.  When XP came out, sometimes it would not work on existing computers.  Vista/7 * won’t * run on a lot of equipment that was sold with XP installed.

Linux does not suffer with this problem—with respect to * hardware *.  If you’re running Ubuntu 10.04 successfully on your current laptop, manufactured c. 2005, your chances that the next LTS release of Ubuntu will boot and run on it are way better than would be the case with comparable ms Windows.  Fine.  Wonderful.

But if you want to install an app that you were used to using in Windows XP, to, say, a Windows Vista, 7, or even 8 computer, you often find that it will still run, or that the makers have coded a new version of their app for use with the new version of Windows.  This is also true in Linux—VLC media player, for just one example—does alright, when it comes to keeping-up with the continued advancement of desktop Linux [though this, and several other popular programs often used to “lag” behind the versions of their Windows siblings—often by several builds:  but not so much anymore, as of the time I write this (late 2012)].  The Windows version will still usually be more current, because Windows is so ubiquitous—so this gets taken care of first.  Linux doesn’t suffer from viruses, so an update to an app is not as critical, in this respect.  So far, so good.

But now what if I find I would like to install a newer version of VLC to my old(er) desktop Linux distro, because I have found the default version of VLC available for Ubuntu 9.10 has a bug, and the save-playlist feature won’t work?  I could just upgrade to the next release of Ubuntu (10.04) on this machine—that’d give me the newer version of VLC.  But my roommate, who also uses this computer, doesn’t like the new interface in Ubuntu 10.04.  She wants to keep the rest of Ubuntu 9.10 just the way it is.  So, can one install the newer version of VLC to the older version of Ubuntu?  This should be easy—Linux is open-source, right?  In Windows, you could do it without even breathing hard.  Just go to control panel, and un-install the current version, and then install, say, the latest version of VLC—even if you’re still running Windows XP.  Maybe even if you’re still on Windows 2000.

Well, you * can * do it in Linux, too.  But it’s surprisingly difficult.  Why?  Probably because Linux suffers from a shortage of manpower (by which I mean * both * man- * and * woman- power).  In order to keep-up with the additional demands such backward compatibility would place upon the makers of apps, and of the various distros, a lot of additional man-hours would be needed.  For the makers of something like VLC—which might be cash-strapped from the recession to begin-with—such additional expenditures could be hard to justify, where it comes to a group of operating systems that collectively own only about 1% of world martket-share.

I would say that the easier solution (and one that can be more fun) for a lot of us in Linux is “distro-shopping”.  If you Google it enough, often you can find some (lesser-known?) Linux desktop that you can get to do pretty much * exactly * what you want, with a modicum of operator-customization/configuring (“tweaking”).

This is also a way to burn-up a lot of * time *, a vice of which no small number of us Linux users are often accused.

25. Entry-25:  Swap partition and Hard-disk Layout.  Linux Swap is the equivalent, basically, of Windows’ Virtual Memory.  If you boot Linux on any computer that has once had Linux installed on it, chances are that the old Swap-area is still there, and the booted Linux will be able to use it.  You can configure Linux Swap as either a Partition, or as a file/folder, which is inside an os.  The dedicated partition is the preferred way.  This is easy to do with G-parted, run from a live boot—like a cd or USB thumb-key that boots Ubuntu.  In Ubuntu 11.04 & earlier, go to System > Administration > G-Parted.  Make the Swap partition about the same size as your installed amount of RAM.  If your machine’s RAM is 512 Mb., make the Swap about 512 Mb.  It seems to be usually okay to make it somewhat larger, too.  But I wouldn’t get carried-away.  Of course, if you’re just going to use desktop Linux’s graphical installer (the Ubiquity program in Ubuntu—though it’s just labeled “Installer”), to install one instance of desktop Linux to a machine for a dual-boot arrangement with ms Windows, then it’ll ask you a few questions here-and-there, and figure-out most of the rest * for * you *.

The way I usually lay-out a harddrive, is to use Windows’  Disk Manager to re-size downward (shrink) my Windows install, and to about as small a size as I think I can get away with, but also take into account that I should leave room for Windows updates, for the future.  Knowing which partition is your actual Windows operating system, and what size at which to arrive, takes some research, and some going to the webpages of your machine’s manufacturer, and probably some guess-work.  This is another reason why it is so important to backup everything before you begin—your personal files—but also your Windows itself—as I have emphasized in other entries here.

If, for some reason, you decide you are going to try to do some partitioning with non-graphical tools, such as the program(s) known as “fdisk”, you should know:  A) fdisk, while not exactly intended for a “n00b”, is easier than you’d think:  what is really more at-issue, is * YOU *—in terms of your understanding of * harddisk * ** geometry **, and * harddrive * * partitioning *, generally.  B), there are * TWO * programs, and * both * are called “fdisk”; there is fdisk of Windows/DOS platform, and there is fdisk of UNIX/Linux/*nix platform.  Both have very similar commands, and function very similarly.  BUT THEY ARE NOT THE SAME PROGRAM.  Both will make changes to the other platform’s partitions, BUT YOU SHOULD NOT USE THEM THAT WAY.  Often, it does not cause any harm.  BUT YOU SHOULD NOT USE WINDOWS fdisk TO CHANGE OR WORK-ON A PARTITION INTENDED FOR UNIX/Linux, AND YOU SHOULD NOT USE Linux fdisk TO WORK-ON OR CHANGE A PARTITION THAT RUNS MICROSOFT STUFF.  A possible exception to this is FreeDOS, which can probly be set-up with a partition from Linux’s fdisk.

Some (mostly older) computers will have just one big NTFS partition.  So there’s not much to figure-out there.  But newer machines sometimes have four (4) primary-type partitions on them, when they come from the factory.  Figuring-out which one is actually your active, booting Windows operating system, so that you can shrink * that *, can be tricky.  (Especially if you’re a * dumbass * like * me *.)  And if your machine has * four * Primaries, then you’ll have to either * destroy * one of them, in order to create an Extended-type partition (and get the advantages of that); or else be content with just over-formatting one of those Primaries with the Linux you want, or else scrapping the whole notion of messing with the harddrive, and booting Linux some other way (of which there * are * nowadays several—but * I * like to install to hdd).

Further, I want to say that when you get a new (or merely “different”) computer, and run Windows Disk Manager on it, or a live Linux’s G-Parted, to see what partitions are on there, where the thing has four (4) Primaries, what you often find is that 3 of them will actually be Windows—one being the Windows operating system itself, and the other two being Windows * recovery * jazz, which you should strive to preserve.  The fourth will often be some crapware from the machine’s manufacturer:  but you’d better do what you feel you need to, to * feel * confident * of what it is.  Your own confidence, Bubba, is all that stands between you and a wrecking of your ms Windows install, perhaps requiring a re-format (perhaps from “bare-metal”) to get Windows back.  And Windows—unlike Linux—can be a Royal Pain to have to re-format with.  And you can’t always tell ahead of time, if a Windows re-format is gonna be a walk in the park, or if it’s gonna be nasty.  It seems to depend somewhat on what “mood” ms Windows happens to be in.  Linux does not come without any risks.  The least risky way, as I’ve tried to lay-out in this paper, is simply not to install it to the harddrive at all, but rather to run it some other way.  Every way of using it has some pros, and some cons.

When you see partitions on a harddisk identified as things like “PQ Service”, and “RECOVERY”— these are often Windows recovery stuff, which a tech could use to recover ms Windows, if it got seriously broken.  We do not want to delete those, or even re-size them—or bother them at all, if we can possibly help it.  Some research, as I’ve said, is your * friend *.

You ought to update Windows first, by running Windows’ Updater, and install any updates that may be waiting.  Re-boot, to be sure that they are applied by the system.  Then do this again.  Do it as many times as it takes, to be sure that there are no waiting updates.  De-frag Windows, by running its de-fragger.  Setting a restore-point might not be a bad idea, either.  Then be sure to have backed-up * all * personal files you don’t wanna take a chance of losing.  Everybody with a computer should be doin’ this at least occasionally anyway, so now is a good time to start.  Then you gotta back-up your Windows operating system itself.  Go to Control Panel > System (& Security if on Vista/7), > Backup & Restore Center.  Follow the directions carefully.  The disc / discs you create may be your only hope of getting ms Windows back, should something go horribly wrong.  I’ll add right here that I am * far * from being a “computer whiz”, yet I have been able to install desktop Linux at least 20 times now—and to various partitions on 4 different machines which I own, plus an external (WD Passport) harddrive which I also have.  And I have not had a tragedy.

DO NOT SHRINK AN INSTALL OF MS WINDOWS TO TOO SMALL A SIZE.  I’ve covered this in entry No. 6.  Please pay attention to what I’ve said as to this.

Now open Windows’ Disk Manager [in Vista/7, Start > type “partition” into the search field.  In older versions of Windows, you may have to search the system, or use a third-party software, or Windows’ own FDISK program (non-graphical:  works from an ms dos-like environment only), or perhaps a dos-type boot-disk, such as UBCD-4-WIN].  It’s okay to re-size ms Windows * while * it is running—re-sizing it from itself.  This is one thing that Linux can’t do—re-size its own partition, * while * it * is * running * from * that * same * partition *.  But Linux can do many other things, which ms Windows cannot—at least not by default, or without the installation of a third-party tool—which those of us with some experience know can be a problem, in more than one way.

I frankly prefer to use a dos-based utility to re-size ms Windows, like those I have described, as it is said to be “safer” (i.e. less chance that something could go awry—like maybe with the partition “re-tabling”, or the way in which that particular harddrive receives it, or just the occurrence of a “bad day”).  But I’ve let Ubuntu’s Ubiquity installer program do it during the install process (where I did not use the “Advanced” / “Do Something Else” option), and all went fine.  Ubiquity uses the tool * G-Parted * (G-Parted = “Gnome PARTition Editor) to accomplish this.  But G-Parted is a tried-and-true Linux tool, which can be run fine from a live Linux environ (a cd, DVD, or USB thumb-key), and I’ve done it this way too, on several machines.  Worked fine every time.  Even to re-size ms Windows.  Or you could just go with the defaults in your Linux distro’s installer, and let Ubiquity (or Anaconda in the case of many RPM-oriented distros) do it “automagically”.  I guess it’s up to YOU.

But back to the (manual) way * I * layout a disk……….

When using these graphical programs, I also prefer to re-size a partition from the RIGHT end, in the re-size dialog window.  I guess my mind figures that the Left end would be toward the hub, in the platters of the harddrive.  If this is so, then re-sizing from the RIGHT end would not move the cylinder head, where the partition begins.  And I guess I figure that this will be less likely to throw-off the bootability of the os, ’cause it would be one less thing the bootloader would have to adjust to.  Anyhow, I’ve always done it this way, unless I could not help it.  And my results have always turned out fine.  But it’s probably just the way my psychology is working:  there may not be any truth to this principle.

Once Windows has been shrunk, boot Windows as soon as you can, and let it run for a little while.  Perhaps install some waiting updates.  THIS STEP IS IMPORTANT, AND YOU’D BE ADVISED NOT TO SKIP IT.

Now fire-up a Linux live disk, and use G-Parted to create what they call an “Extended” type partition.  [In Ubuntu 11.04 & earlier, go to System > Administration > G-Parted.]  This is going to hold our Linux stuff, plus a big FAT 32 (or NTFS) “logical-drive” partiton—which will act as a great big internal thumb-drive, and allow the contents (our movies, music, wedding pictures, perhaps our documents as well) to be read/written-to from * both * systems—Windows * and * Linux.  It can have a lot of space to store these personal files, compared to the usual USB thumb-key of 4 or 8 Gb.  And Windows * likes * having this “vFAT partition”.  And because all operating systems on planet earth can read/write to FAT 32, your data will always be accessible.  It will also be easier to off-load your personal data, should Windows get badly virused.  FAT32 is the historic, files-holding native Microsoft/Windows format, which has been in use since the 1980s.  Many persons who * only * run ms Windows create * just * such a partition to hold personal files, just in case Windows gets badly virused.  A big FAT32 partition like this is also nice in case you have to re-install your Linux distro (or even Windows).  Though I have never had to re-install a Linux desktop distro because something “got broken”, or for any other reason than that I frequently do software “experiments”, and sometimes run out of space to hold new distros.  I have had very good luck with desktop Linux—especially the LTS versions of distros like Ubuntu, and Linux Mint.  As an aside, I’ll mention that Linux is easier to do a re-install with—to re-install it to its own partition—than is the case with ms Windows.  That’s because Linux is given away free, whereas Microsoft is always worried about people pirating Windows, and so there are embedded “security devices” in your Windows system, which sometimes makes it tricky to re-format with.

Having this nice big FAT 32 partition is a nice addition to any computer—no matter what kind it is.  And when Windows “sees” it, it will usually call it “E:\”-drive.  But this, again, depends on how many and what kind of drives Windows may have already labeled, and whether or not you’re running on a netbook with no cd-tray (in which case “D:\” will not be already taken), and similar factors.  So try to notice, when you boot Windows again, what label it has put to the new vFAT partition.  BE AWARE, ALSO, THAT SOMETIMES WINDOWS MAY CHANGE THE DRIVE LETTERS IT HAS ASSIGNED.  This was more of an XP phenomenon; it doesn’t seem to happen in Windows 7.  But I am giving you the caveat right here, that you should “keep an eye on it”.  Because this is just kind of a “glitch” in ms Windows.

Under Linux, the vFAT partition will be shown with its UUID displayed (looking at it from G-Parted, anyway, and maybe the files-manager too), and other info about it—like its size, and type of volume-format.  Linux’s Nautilus files-manager will also show it, in its Left view-pane.  And you can also give it a “volume label” using G-Parted, which I encourage you to do.  Giving the partition (“volume”) a name (like “franks1”, or “ubuntu1204”, or “swappy”) using a Linux program like G-Parted, will help you keep things straight.  When viewed from ms Windows’ files-manager (“Explorer”), or even “My Computer”, or Windows Disk Manager, your Linux partitions are gonna show-up as “unknown file-system(s)”—because ms Windows cannot read/recognize UNIX-type/*nix-type disk formats.  Not without some third-party software, anyway.

I like to create these partitions in the order of:  “what I’d probly LAST want to delete, I will create FIRST”, and so on from there.  Because once you have created some partitions on any disk (a harddrive or a USB thumb-drive, et al), * IF * YOU * DELETE * ANY * IN * FUTURE *, YOU * ARE SUPPOSED * TO * HAVE * TO * DELETE * IN * THE ORDER * OF * THE * ** LAST ** * CREATED * ** FIRST ** *, and then so-on.  This is another thing to keep in-mind, if laying out a disk manually.

First, make a nice, big, FAT 32 partition out of the void of “unallocated” space we now have.  Or maybe better yet, an NTFS partition—you can use NTFS for this nowadays, and it’ll work just as well for this “personal data partition”.

Note that Linux now supports all your Windows formats—it’ll read and write to * everything —FAT 32, NTFS, .doc, .docx— everything *.   [You * can * also read a Linux-formatted partition or drive from ms Windows, by the installation of certain third-party apps (often free).  Research this, and create a Windows’ Restore-Point beforehand, if you decide to try it. ]  Most of the documents-apps for Windows [Word, Word-Pad, MS Works, &tc.] that * come * installed * to recent (after, say, 2009) versions of Microsoft Windows, will open and read/write to .odt document format.  Which is a nice international standard from the ISO people, and the documents-format many Linux desktop users prefer.  Modern desktop Linux will also let you save to .doc, or even * .docx *—or any “Windows” format.

Be sure you have clicked-on the big band of “unallocated space”, so that the G-Parted tool is “pointed-at” the correct disk-space:  DO NOT ACCIDENTALLY OVER-FORMAT YOUR WINDOWS STUFF, BY MISTAKE!!  In G-Parted, just go to “Partition” in the top-ribbon/toolbar; then select “New”, then use the slider to set the size, and use the drop-down menu to select the format for the new partition—in this case, select “Linux-swap”.  After you click “ok”, G-Parted will “load” this, the command you have “set-up”, to its queue.  You’ll see that along the bottom of the interface.  In order to execute it, you’ll have to click on “Edit, then then you’ll be asked to confirm.  You may be asked to confirm more than once:  this is just to get you to make sure that you really intend to carry-out the step(s) you’ve just told it to do.  BACK-OUT IF YOU’RE NOT SURE (by just closing G-Parted).  Once you have finally confirmed a command in G-Parted, and it has begun a process, it is unwise to abort it:  doing-so could render the target partition unusable for anything in the future—or it could take a person knowledgeable in disk partitioning-and-manipulation to put it right—if it * can * be.  Better to let it go ahead and execute you set-up command, even if you suddenly realize you’ve made a mistake.  G-Parted, or other means, can usually put it back, later-on.
When that’s done, we can make a small Linux Swap partition, about the same size as your RAM.  The dedicated partition is the preferred way, as opposed to other ways of creating Linux swap (such as swap * file *).  But you can use the “swap-file” method, if you prefer.  The swap-partition method is easy to do with G-parted, run from a live boot—like a cd or USB thumb-key that boots Ubuntu.  Make the Swap partition about the same size as your installed amount of RAM.  If your machine’s RAM is 512 Mb., make the Swap about 512 Mb.  It seems to be usually okay to make it somewhat larger, too.  But I wouldn’t get carried-away.

Next, create a somewhat larger Linux partition, to hold the desktop Linux distro I’m going to install.  The minimum size for this is probably about 6 Gb + for Ubuntu, more for most others.  12 Gb. is a good number.  You wanna take account of room for future updates, and programs-suites you might want to install.

I prefer to use any remaining space to create yet another partition / partitions, with which I can experiment with Linux installs I may want to try, but perhaps not be ready to use on a daily-basis.  Or maybe to hold an older, very stable Linux distro, which no longer updates, and whose own system-files don’t take up much room, of themselves:  something like Xubuntu 11.04; this is good to use as a * Linux/Unix * formatted * partition, for the “double-backup” of personal files.  It gives the advantage of the .ext4 superior Unix-type filesystem format (or use a different Linux/*nix format for this one if you want), which is more stable—especially long-term—than FAT32 or even NTFS.  This is advisable, because everybody who owns a computer should backup personal files to ** more ** than ** one ** location—even if on the same disk (though I hasten to add that you are also supposed to backup to * another * disk/discs—like removable media, such as cds/DVDs.  And using the free space many cloud-services offer doesn’t hurt, either.  That will give us 4 (four) layers of redundancy, split-up among different disks—which is what IT professionals recommend.

You do the math, according to how big your disk is to begin-with.

Just remember that * if * at * some * point * in * the * future * you * decide * you * want * to * delete * some * partition *, and get the space back, you are supposed to delete partitions in the reverse-order in which you created them.  This means “last created (the one with the highest number), first to be deleted”.  Doing otherwise can mess-up the file-system tabling of a disk, and necessitate a lot of wrangling, to straighten it out.  Or you may just luck-out, and come out smelling like a rose.  But you can’t know that ahead of time, so the prudent course is probly the way I’ve described.

I want to mention, too, that having a FAT 32 / NTFS “personal” partition has another advantage (well, several, in fact, but I’ll mention this one):  it will also let you save a desktop Linux as a “frugal” install, and boot it off your harddrive as a live, compressed file-system.  Even if you’ve remastered it in your full-install partition, so that the live-image has all your favorite apps.  You use the same FAT 32 partiton for both—saving personal files, to be accessed from both Windows and Linux, and also as a way to “frugally-boot” a re-mastered copy of your desktop Linux distro, which has some serious security advantages.  (See entry No. 23.)

I guess I will add right here, as a kind of special note, that since * at least * version 9.10, Ubuntu’s Installer program (which you can activate from the live-cd, by DOUBLE-clicking on it) seems to have been much improved.  If you happen to get to reading an old thread on the internet, and people are talkin’ about how they had to partition their harddrives themselves, and what a pain it was, and how they then had to copy the system files to the harddisk, &tc &tc, be sure and check the * dates * of these posts, as to * when * the authors posted them.  AND BE AWARE NOW, that Ubuntu’s Installer nowadays does all this * for * you (assuming it’s working right on your hardware), if you choose to install it to your system’s internal harddrive.  (Unless perhaps you choose “custom install” from the menu for some reason).  So you probly won’t have to run G-Parted yourself, or any of that crap, if you don’t wanna.

I have installed Ubuntu 9.10 to the harddrive of two machines, and both went-off without a hitch, and each install took about 90 minutes of my time.  I currently use Linux Mint 13 XFCE edition.  And I have many other installs, split up among 4 computers and 10 different drives.  I’ve installed to some other machines, too.  But it is not enough to make me an authority.  Like I said in the opening paragraphs, I only write all this because I like Linux, and in the Hope it may help others.  Be prepared, though, if you are intent upon installing any Linux distro to a machine’s internal harddrive:  read the manuals, and the rest of what I (and others) have to say here, and make sure your battery is full if you’re gonna try this on a laptop.  And be sensible, as to when you plan to undertake it, if you are gonna mess with the harddrive-method:  don’t plan it for that morning when you know it’s gonna be your turn to pick-up the kids from soccer practice.  Because it might take you longer than 90 minutes, Dude.  And for goddsakes be sure you have backed-up all your data first—like to removable media such as cds—and defrag WINDOWS, then open Windows Updater and download and apply any updates that it can find.   And then create a resore-point, just for the hell of it.  These are easy things to do—especially in WINDOWS 7 or WINDOWS 7 Starter-build.  Especially if you are gonna install a Linux distro to the machine’s INTERNAL HARDDRIVE.  [The equivalents of these procedures, * where applicable *, are just as easy in most modern Linux, BTW.  You don’t have to defrag in Linux, for instance, because Linux has a better file-allocation and write-to-disk modus. ]

26. Entry-26:  If you can’t live with the default Desktop Environment (the “DE”) in your new Linux system, install another one.  There are many from which to choose.  You’ll be offered the new option the next time you log-in.  But the opportunity to log-in to what you already had, as your existing Desktop Environment, will still be there.

Microsoft Windows is a “do your homework” operating system.  They all are, even MAC OS/X.

But Linux requires a little more homework, because it’s structurally different.  It is these structural differences, however—both in terms of the “business-model”, and in how the system itself is organized—that make Linux a more versatile and secure operating environment.  Google is your friend.

What’s a Desktop Environment?  A DE is basically the GUI—Graphical User Interface that allows you to do stuff “visually”—a’ la’ with your mouse, by point-and-click.  Instead of the CLI / pre-graphics way of personal computing, where we all had to use something like DOS (or UNIX), and do all tasks by carefully typing a whole bunch of “commands” into a plain black-and-white screen, and be periodically met with the response “syntax error”, because we mis-typed.

GUI / Desktop Environment removes this inconvenience (most of the time), and allows us to compute “like human beings”, just by using our mouse.  Without a Desktop Environment, you’d be typing into a DOS-window (or DOS emulator) in ms Windows XP or 7.

The default DE in Windows XP is Microsoft LUNA, and this has been in use for some time.  LUNA is the GUI that runs on top of Microsoft’s NT kernel, which replaced the old MS-DOS kernel somewhere around the time of Windows 98.  I say “somewhere around the time of”, because Microsoft has had the habit of not being uniform in its product changes.  Not all editions of Windows XP were shipped with the exact same set of features, for example—even though we’re talking here of the same grade of XP (“regular”, and not XP Pro, etc.).  And this was true even in respect of some installs of XP to machines that shipped * during * the * same * time *window *.  Also, XP sometimes varied this way with the hardware vendor:  sometimes a certain feature or update would be included in installs to various models of, say, HP, but Dells wouldn’t have these in their version of the os until later—or the other way around.  Nor would users always be informed that they could just go to, say, Microsoft.com, and just d/l the new feature—often free of charge.  Sometimes people would be notified, sometimes it would take an “uncharacteristically” long period of hours or days, for either MS or the hardware vendor—whoever was designated as the “responsible party” in the particular case—to notify users.  So some users of XP got SP2 much sooner than others, and others got it “later”.

Now, I want to * emphasize * that I am not pointing a finger of blame at MS for this behavior, or impugning them in some way.  Yes, I * do * have * other * problems with Microsoft, and I ** do ** and ** have ** occasionally made these known in my online activities.  But I guess I just am one who does not much blame them for the above; it is the way things often work in the “real world”, which desktop Linux doesn’t have to live in (as much).  Because most of desktop Linux (and the Linux metaverse generally) is not-for-profit, and holds a tiny market-share.  This latter gives Linux a “breathing-space”, in that it isn’t (perhaps) recognized as much of a “threat”, by the corporate producers of the other two major pc platforms.  And the (early) XP era was quite some time ago—especially when measured in computer-years.  Things were actually sort of “experimental” then.  Or they had been not long before.  A for-profit corporation simply is not as free to act in certain ways, compared to “ngo s”, like the Linux Foundation, the Debian Foundation, the GNOME Foundation, et al.  Or even Cannonical, Ltd., which seems to be part for-profit, and part not-for-profit, as best as I can understand it.  So I am not one of those who finds this practice (in earlier XP) particularly annoying.  But some other folks do.  Depends, I guess, on what is the nature of your involvement in the IT industry.

Anyway, earlier versions of ms Windows (like Windows 3.1) used to use GEM for a “Desktop Environment”—though GEM is really more of just a “window-manager”.

LUNA was probly introduced somewhere around Windows 95, and lasted, more-or-less (and with alterations), until Windows Vista.  Vista introduced Microsoft Aero, and this is the current DE in Windows 7 (although it is possible to d/l and install third-party Desktop Environ to W7—or even earlier versions of ms Windows—though most users seem not to know this).  Windows 8, as you may have heard, comes with a new and “touch-screen-oriented” GUI known as Windows Power Shell.  The new Windows Power Shell interface has no Start button in the lower left:  instead, users are supposed to launch applications by means of the “squares” that appear on the screen, or the Windows key, which if I understand correctly brings up a search-field.

One might characterize this as the “ANDROID way” of using the machine.  And for Windows Phone, or more-to-the-point Windows 8 running on a * tablet * device, this is probably called-for.

But in the user-realm of tower and laptop computing, so many people have complained about this, that Microsoft has announced that they are creating a “patch” for Windows 8, that will restore the old Start-menu functionality.  This patch is slated to be ready in June, 2013.  The Power Shell interface will remain available on the system, should the user wish to access it.

Where it comes to Linux-based desktop operating systems (“desktop Linux distros”), Linux desktop operating systems have had their own issues with what we might in this context call “product consistency”.  The suite of apps you get in one d/l of Fedora 15 is almost always the exact same as the set of apps installed to anybody else’s download * of * the * same * release *.  Ubuntu, as I may have mentioned, has the (somewhat) annoying habit of “shuffling” the suite of default apps, from one release to another.  So we might get a very different photo-editor in Ubuntu 10.04, than we were used-to in 9.04 and 9.10.  But the changes will usually be the same, however, for all downloads of that release.  It is often possible, BTW, to just install the app that we had, with just a few mouse-clicks.  The Linux desktop has other quirks in this vane, some of which I have tried to make mention.  Really, both platforms (ms Windows * and * desktop Linux) have annoyances.  Depends, I guess, on which you prefer.

Myself, I prefer desktop Linux for its possibilities in privacy and security.  But there are other virtues, too.

Where it comes to which Desktop Environment to use in Linux, frankly, I like XFCE Desktop Environment for Linux, and also IceWM.  Both these run real well on-top of all releases of Ubuntu, including 11.10.  (Though I am currently using Linux Mint 13 XFCE Edition as my daily desktop, BTW.)  XFCE and IceWM seem pretty good for work-flow, on screens 10” and up.  However:  where you’re a * social user *, trying to run a Linux distro on a tablet computer not much wider than a pop-tart, you may well * prefer * a Desktop Environ such as Ubuntu’s Unity, or Gnome3.

XFCE is a historic Linux Desktop Environment—meaning that it’s been around for quite a few years now, and is used by a considerable number of people.  XFCE is technically classed as one of the many “Lightweight DE s” available to Linux desktop users, but it has “grown up” considerably over the years, and now XFCE 4.8 is practically as large and full-featured as classic Gnome (Gnome 2x).  For a long time, XFCE was half-jokingly known as “GNOME-Lite”.  XFCE is even used by many MAC-book owners!

XFCE is even used by many MAC-book owners!
–>I suppose some could still apply that appellation; but really, the current build of XFCE (the 4.8 build, which is what I got with Linux Mint 13 XFCE Ed.) is significantly improved over all earlier versions of XFCE. From its origins as a lowly window-manager, XFCE has grown-up. XFCE releases like 4.8 and + up are now full-blown Desktop Environments, the practical equivalent of GNOME 2.x and/or KDE 3.

XFCE uses similar backend software (such as GTK), and has an overall look and feel similar to “classic” GNOME (the Gnome 2.x series).  It’s generally regarded as lighter and faster than Gnome 2.x:  however, most of those comparisons were made prior to Gnome 3.  Now the two desktops take such vastly different approaches, and such comparisons would have less meaning.

XFCE doesn’t try to reinvent the desktop like GNOME 3 and Unity do, but it does provide an environment in which long-time GNOME 2 and Ubuntu users will feel at home.  XFCE was once GNOME’s less resource-intensive and more minimal cousin, but GNOME 3’s shift has made XFCE a distinct environment.  XFCE is not really either GNOME-oriented, nor KDE-oriented, however; therefore, XFCE can usually run both KDE-oriented and GNOME-oriented packages.  Really, most Linux DE s nowadays are able to run either sort of packages—either those originally built for KDE, or those originally intended for GNOME.  And it has been this way for some time.  So which DE you decide to use has much more to do with things such as what you’re using the computer for in the first place, so-called work-flow, and etc.  And what size your screen is.  As I said above, where you’re a * social user *, trying to run a Linux distro on a tablet computer not much wider than a pop-tart, you may well * prefer * a Desktop Environ such as Ubuntu’s Unity, or Gnome3.  And maybe even if you’re * some * other * kind of computer user.

You may wish to check-out the add-on packages “XFCE-4 Goodies”, and “XFCE-4 Utils”, as well as many others that are available to enhance the XFCE desktop experience—especially where you can view them from Synaptic Package Manager.  XFCE-Gnome-support is a package to check-out.  No less than Linus Torvalds himself—the creator and namesake of Linux (Linus + Unix = “Linux”), has recently said publicly (relative to the time of this writing) that he’s switched to XFCE on his personal laptop, because he has found it better for productivity than GNOME3.

IceWM is technically classified as “only a window-manager”, and not a full-blown DE.  But it supports enough features that it’s about as good as a “full-blown” DE, * in * practice *.  IceWM may not be as full-featured, but it is very light-weight, and so it will be fast, and will probably work well on lower-end hardware.  IceWM is built to have the look and feel of microsoft Windows 95/98.  But IceWM will support additional user-accounts, provided that the underlying operating system (distro) supports them.  Most desktop Linux distros do, but some versions of Puppy Linux run only from Single User/Root.  There is a downloadable version of IceWM called “IceWM Gnome-Support”.  This has some added stuff so that it will be even better at running some traditional Gnome packages you might want, but which might not be included in your distro by default.  Things like Gnome Session Manager, or G-Parted.  Look in your distros default repos (Software Center) first, if you become interested in IceWM.

If you are gonna try IceWM, it seems it is prudent to also install RoxFiler with it, even if you are not going to ultimately set this as your default files-manager, but use something like Nautilus instead (which you may already have installed).  You can control which files-manager is the default, should you have more than one installed to your system, by using a system-setting.  IceWM’s legacy (third-party) graphical tweaking utilities are, unfortunately, no longer maintained.  But it is not so hard at all to adjust it, by just opening the appropriate text-file in a terminal text-editor, and removing the comment-symbol (the # symbol, in this context) from the appropriate line.  There is good documentation as to how to tweak iceWM this way, online.  The powerful (and highly technical) Linux distro * Scientific * Linux * actually offers IceWM as a desktop option.  See also entries No. 17, 20,46.

I should probably add here, that Enlightenment DE for Linux now seems to have been perfected, though I have not had a chance to try it out yet, at the time of this writing.  I’ll try to update this article, when I’ve had a chance to try it.  I haven’t heard much that’s negative, where the current version (E-17) is concerned.  Enlightenment E-17 is a full-blown DE for Linux, and seems to run equally well on all the major distros.  Enlightenment is full-featured, but works rather differently than desktops you may be used to.  None-the-less, it seems like it’d be good for those of us who wish to do serious work on a desktop computer.  So it seems like a viable alternative to good ‘ol GNOME 2x.  But * expect * somewhat * of * a * learning-curve *, if trying-out Enlightenment E-17 Desktop Environment, whether on Ubuntu, or Fedora, or some other Linux distro.

I guess I can say a word about LXDE as well.  I really love it in Knoppix 6x.  Nice.  Lightweight, yet full-featured.  I’d think this would be the future of the Linux DE.  Unfortunately, there has been somewhat of a bug, where it comes to using it on top of Ubuntu, and therefore also with instances of it in Ubuntu variants.  This may have been fixed.  I haven’t had time to play with an install of Lubuntu (LXDE + Ubuntu = “Lubuntu”—which can be downloaded as its own .iso, ready to go).  So I really don’t know if it has been tracked-down or not.  If they have really solved it, then I’d tend to think the LXDE Desktop Environ would be good for work-flow, and productivity on a Linux desktop install.

Again, apologies-in-advance, to those whose favorite Linux DE I have not mentioned.

27. Entry-27:  You install a software from control-panel in Linux (“Software Center”), as well as un-install. Linux’s “control-panel” is a * two * way * street.  This is the easiest way to install a software.

You should probably NOT un-install / otherwise remove a software package that was already present, when you downloaded your distro.  A good rule-of-thumb is, “if you did not install it, then don’t remove it”.  If you want to build a really custom desktop, it might be best to just start with something like Bodhi Linux—which is stripped-down, by default—and then “build-upward”.  This is the way many knowledgeable Debian-people seem to recommend:  “instead of deleting programs, start small, and then build-upward”.

Another good distro to use as a “base-from-which-to-build-upward”, is probably Xubuntu—this one is actually * made * by the makers of Ubuntu, and so is actually an “in-house” build—not a third-party variant—if knowing this gives some of you a psychological “security-blanket”:  as for myself, I have never been able to tell much difference, as to whether it is “in-house”, or a “community-based” variant-build; both have their pros-and-cons.  Xubuntu comes with XFCE as desktop environment—not Unity, like in Ubuntu-proper.  XFCE for Linux has been around for quite a while, and is well documented, and pretty easy to use—indeed, I find it is rather like an “upside-down” version of Windows  XP’s desktop environ (Microsoft LUNA); the “Start” button is in the * UPPER * left, not the lower-left of the screen (except in certain distros like Linux Mint); and the “Task-bar” (called the Tint2 Panel in Linux) is a band along the * TOP *, not the bottom (again, Linux Mint excepted).  XFCE is pretty easy to use, for a person coming from Windows.  Once you get used-to XFCE’s few quirks/glitches, it is serviceable, for productivity.  At least until such time as DE s like MATE, Cinnamon, and/or LXDE get perfected.  These three DE s have only been in development for a little while, at the time of this writing; so let’s try to be positive, and cheer-on the benevolent geeks working on them.

28. Entry-28:  There are at least a few ways to repair your Windows install, if it starts to act hay-wire.  But not many of us seem to know how to boot and use the recovery-console in Linux.  Nor do we much bother.  Because Linux is free, it is often easier to just re-install a backup you’ve made of your system, using a tool like RemasterSys.  Or try a different distro.  Once you get desktop Linux really working well on your hardware, it is often * very * stable.  Often more-so than Windows.

There is no Safe Mode, or System-Restore feature in Linux.  No OEM product-key either.  These are not needed.  You do not have to pay for Linux.  The original live-cd from which you installed your current version serves as your backup disc.  [If you really wanna backup your install—like after you have it all tweaked and set-up like you want it—RemasterSys seems to be a real good way to do this in Ubuntu and its variants.  Works for me.  There are other ways to re-master Debian-based Linux.  Puppy and Fedora, by contrast, have their own re-mastering tools, pre-installed.  And there are plenty of harddrive-partition cloning/”ghosting” tools available for use with Linux—most of ‘em free.  There’s the P.I.N.G disc, which you can download, and there are even tools that you can pay-for, if you want.]  You backup data added to the system with removable media, and with cloud-services.  Anti-virus/anti-malware programs are useful to run in Linux, mostly to scan a file you intend to pass-on to someone with a Windows computer.  The free versions of these are perfectly adequate—you don’t have to pay for anything.  Some people who run Linux as a * server * (and not as a desktop) use Acronis True Image when “ghosting” the harddrive; but this is a pay-ware tool.

If Linux does have some major problem (rare, if you have followed the advice above), it can be re-installed, or another distro can be installed, or it can simply be wiped from the hdd. (Research how this is done—especially if you intend to do the latter).

If you want to make a backup of an install of Linux, you can create a ghost-image with
P.I.N.G. , which you can download for free; or, you can use any one of several utilities/discs that you can d/l free, or that you can buy.  I tend to like RemasterSys for Debian-based Linux. Yes, it is not available from the default repos, and so you have to edit the tail of a system text file, in order to enable its repository, so that it will download and install (see entry No. 46). Well, I have never had a problem with it.  It is a nice GUI.  And when you execute it, it will show you itself running, in its own “terminal”.  RemasterSys for Debian/Ubuntu-based Linux lets you create a partial or a “full” backup of your desktop Linux install, or create your own custom distro, complete with its own portable .iso file, and a checksum hash for that file!  It is pretty easy to use, even for us noobs (once you have successfully installed it).  RemasterSys comes already installed in some Ubuntu variants, like Vinux and Backbox Linux.  There are programs like RemasterSys available for other, non-Debian Linux desktop systems, such as Slackware; and Fedora and Puppy Linux, as I’ve indicated, come with their own re-master utilities.

For an equivalent program to back-up personal files—like your documents, wedding pictures, and videos, there’s DejaDup (pronounced “day-zhah-doop”) for Linux.  This comes pre-installed in Linux Mint 12.  It is a pretty good equivalent for things like Windows’ SyncToy.  Indeed, there are a whole * host * of backup tools for modern desktop Linux.

If you’re hell-bent on being able to set Windows-like “restore-points” in desktop Linux, I think I’d check-out Back-In-Time for Linux; this may be available from your default repos, depending on what distro you’re using.

29. Entry-29:  Even to-day—at the time this is being written—modern desktop Linux * still * has very great difficulty supporting certain hardware.  Certain sound-cards just will not work.  Broadcom wireless networking (wifi) cards did not have support OOTB (“Out Of The Box”, as they say) until about this year—2012—or somewhat earlier in certain distros, depending on the wireless-card model.  Instead, you had to download the reverse-engineered driver and something called NDIS-wrapper, and configure it yourself, in Ubuntu (though Linux Mint and some of the other distros that use Ubuntu as a basis, * did * include these drivers in their installed software).  Even then, it was sometimes unstable, and would periodically drop your internet connection.  But now it seems to have been fixed pretty dern well, because Broadcom and Linus Torvalds, the creator/maintainer/namesake of the Linux kernel, seem to have come to an “accomodation”.

There can be trouble with certain GPUs/graphics-cards.  If you’re planning to buy a new machine (or a pre-owned one) on which you intend to run Linux, do some research before you pay for it, as to the graphics-card, sound-card, wifi-card, Ethernet-controller, track-pad/mouse-pad (the kind where you drag your fingertip around on it)—if that last is important to you.  Try and find out about the BIOS, too.  Very new equipment may contain Microsoft Secure Boot in its firmware, and thus may have serious difficulty allowing any non-Microsoft operating system to boot and run on that computer.

30. Entry-30:  Linux has considerable difficulty in supporting certain peripheral devices.  Linux’s track-record with HP printers and Epsons is pretty good.  Lexmarks can still be almost hopeless.  But some models of Lexmark printer are supported.  Cannons can be difficult—or impossible, to get working with Linux.  But sometimes they work like a charm.  Depends, again, on the series and exact model.  I have yet to see desktop Linux (Ubuntu, Mint, Xubuntu, Vinux) fail to support a USB-mouse.  Now, when my neighbor and I booted Ubuntu (9.10) on his (custom-built) tower computer some time ago, Ubuntu 9.10 did not recognize the wireless keyboard (I think it was a * d-link * brand, but I’m not sure I remember right).  It had no trouble, though, with the wireless mouse (also a “d-link”).  Logitech equipment seems to do well with Linux.  GE also. And GE stuff is * cheap *.  Watch out for GearHead; this one * can * be good with desktop Linux, or * not-so-good *.    Depends on the exact model, device in question, and which Linux distro you’re trying to use with a GearHead peripheral component.  NetGear equipment, on the other hand, seems to vary, according to the exact model/series.  Same for Belkin stuff.

HP stuff seems to almost always work.  * Almost * always.  Sorry I can’t be more “brand-specific”.  Things are, well, just “the way they are”, where it comes to this issue of peripheral-device support.

31. Entry-31:  For most of its existence, desktop Linux has lacked equivalent programs for a few of the lesser-known [“obscure” (?)] OFFICE software programs.  This situation seems to be rapidly changing, however..  But this is another reason not to “dump” your ms Windows install, and format the whole disk with Linux.  Once in a great while, you may need some relatively obscure office software, that may not have a native Linux version.

32. Entry-32:  There are many * very nice * audio / music editors available for modern desktop Linux—most of  ’em * free *—in keeping with the Linux/open-source philosophy.  But none (so far) as really * insanely *, * Kick-A%#&! * * Great * as in the other two major platforms.

33. Entry-33:  There are some * nice * games in Linux—* BUT *.  Some Windows-oriented games now have a Linux version, or have actually been * ported * to Linux.  BUT MOST WINDOWS-ORIENTED GAMES—ESPECIALLY THE NEWER ONES—WILL NOT WORK IN LINUX, OFTEN TIMES EVEN IF YOU TRY TO USE A COMPATIBILITY-LAYER, SUCH AS THE WINE PROGRAM, OR RIL (Run In Linux).  This is a fact that the Linux developer community has been working on for years.  But so far, no dice.  (The same is broadly true for Apple/MAC computers, by the way—most games that run on Windows are not gonna function on MAC.  They just won’t fully work.)  So if you’re a serious gamer, it’s arguably best to just boot into your ms Windows partition, to play Minesweeper, or what-have-you.  Or else perhaps invest in an X-Box.  Linux does most other desktop things very well, though.

34. Entry-34:  If you un-install desktop Linux, you will not be able to boot Windows again, until you have either re-installed a Linux distro, or else booted a Windows recovery disc, and run “fix MBR” [a disc like those you should make ahead of time, with Windows’ own * backup-and-restore-center—assuming this utility is installed to your version of XP—it is not available in all builds of XP, by default; or, you may be able to use UBCD4Win, a bootable dos-based live disc, which you can download:  note that the latter (and perhaps even the former) should probably only be used by a person with some experience in this area]. This is because when you install desktop Linux, you also get the bootloader, GNU Grub.  This is necessary because it is hard to boot desktop Linux from Windows’ own bootloader.  GNU Grub takes-over the booting of all operating systems on the computer.  GNU Grub is a most excellent product (and free, too!).  It is very reliable, and will boot what you want (including your Windows install), usually without any trouble at all.

Having ms Windows on a computer’s harddrive and also another operating system (like Linux) is called a “Dual-Boot” arrangement (sometimes “multi-boot arrangement”).  But you should know that GNU Grub comes in two (2) parts:  the “MBR” piece(s), and the “OS piece”. The MBR piece(s) will be installed to the harddrive’s MBR.  These would be GNU Grub Stage 1”, and Stage 1_5 (if present in your version of Grub; 1_5 is an optional bit of software).  Grub Stage 2 is the “meat-and-potatoes” part of the bootloader, where most of the action takes place.  BUT STAGE 2 IS INSTALLED TO A FOLDER, * INSIDE * YOUR INSTALL OF DESKTOP LINUX.  If you un-install desktop Linux, you will have also un-installed Grub Stage 2.  This is okay, but until you either A) re-install Linux, or B) use something like UBCD (or your Windows recovery disc) to restore your machine’s MBR to “Windows-only” condition, you will not be able to boot Windows, or anything else.

35. Entry-35:  When do you have to de-frag in Linux?  Well, there * are * actually ways to do this in Linux, * but * you * do * not * need * to *.  Because Linux uses the underlying Unix file-system structure, and writes-to-disk in a much better way than Windows, de-fragging a Linux disk is never needed.  Furthermore, your files will not decay all by themselves, just because of the passage of time.  This is an advantage Linux has, at a more basic, lower-level, which it has inherited from Unix, which always * was * better at write-to-disk (at least in * this * area).

36. Entry-36;  CaSe sENsitVITy in Linux:  Unlike the old Windows-DOS non-graphical (“CLI”) environment that you may be used-to from the “stone age” of personal computing, the CLI/Terminal-emulator interface in Linux is * case sensitive *:  Linux cares whether you are typing a lower-case “r”, or a capital “R”—* when * you * are * in * a * Terminal * environment *.  When in GUI, though, most of the modern desktop distros (like especially those I’ve recommended for us newbies) do not care about UPPER-CASE vs. lower—especially in search-fields.  But security stuff like login-managers care.

37. Entry-37:  Linux counts from zero ( 0 ), NOT from one.  Try to remember that Linux counts things (like interfaces) from 0, if it can.  This is true of most Unix-type systems, such as Linux, BSD, Solaris, Unix 5, and probably also MAC OS/X and Google’s Android.  That’s why Eth0 is Eth0—and not “Eth1”.  Eth0 is “Eth” (ethernet) + 0 (a zero).  The zero (0) is thinner than the capital letter “O”.  That’s the principal way to tell the difference between “0” and “O”.  Some computers are hooked-up to more than one Ethernet connection, and can switch back-and-forth; in these cases, Linux will display both Eth0, *and* Eth1.  (And perhaps also Eth2, Eth3.)  WiFi is different, but also similar.  Look elsewhere in this data-base for information pertaining to wireless connectivity.  This can be confusing, because some of the programs we want to use to set-up desktop Linux in the first place, * do * count from one—or else part of the program that * used * to count only from zero has been updated, so that it * now * counts from one—but some other part of it still counts from zero (as in 0123456789 10 11 etc.).  This is not as bad as I make it sound, and (usually) is not something you’ll find yourself having to wrangle-with.  But I’m gonna try and lay-it-out right here, anyway, just because it can be useful information.

38. Entry-38:  You * can * try dragging your cherished Windows apps with you, over to your Linux install, and running them with a compatibility-layer, like the Linux WINE program.  But really, it is almost always better to try a native Linux app first.  You will find most are already installed—inside your desktop distro—just waiting for you to discover them.  Modern desktop Linux offers you about 30,000 additional softwares to choose-from, through its package-management sub-systems (Ubuntu Software Center, Synaptic Package Manager, et al).  These are * free * downloads, most under the GPL license, or similar licensing.  EULAs are available, generally, for anyone’s inspection.

39. Entry-39:  “RIVER” CAN BE BETTER THAN “POD”:      If you’re looking to get away from WINDOWS, getting an iPod is probably not your best bet. While there are many Linux programs out there that interface well with the iPod (AmaroK, GtkPod, etc.), iPods aren’t ideal for Linux, and you’re probably better off getting an iRiver or a Sandisk player. They tend to work well with Linux (without helper applications) and support drag-n-drop. iRivers, too, supposedly support the Ogg format (not just MP3).  (Although I personally tend to think the .ogg has been a project that didn’t turn-out so well, so I use MP3.)

40. Entry-40:  Can’t find an instruction on how to do something from the Terminal, even though you’ve Googled it repeatedly?  Try Googling the same question, but address it as a * UNIX * question, instead of a question about Linux.  Linux, Dude, is really just a cut-down version of the old AT&T UNIX, which they built for their “big iron” (mainframe computers), back in the 1970s.  Linux is cut-down, so it’ll run on a small computer:  but Linux is also up-to-date.  Just about any Terminal command that works in UNIX will also work in Linux.  You’d be amazed at how many UNIX text books are available free, online.  Also forums, message-boards, other UNIX resources.  And most commands that work in UNIX will work fine in Linux.

Just don’t let some jerk con you, as a new user, into running rm -r, which will delete all your files.  In Linux/UNIX shells, “rm” means “remove”; and most systems will also be able to use the command “remove”—as in “remove firefox”—which will delete Firefox browser if it is installed (but still leave its settings and customizations folders).  The “remove” command does more-or-less the same “deleting” as the “rm” command ; but you should know that there may be subtle differences.

The command “purge”, as in “purge thenameofafolder -r” will NUKE all the files in the folder “thenameofafolder”, and do so “recursively”:  that’s what the “-r” means.  The “-r” is a ** command ** modifier **.  Linux/UNIX people call these “command modifiers” “command arguments”.  Also “commandline switches”, “commandline options”.  “Recursive” means “all folders deep, all the way to the bottom of the dang stack that’s under the folder that was just named in the command”.  So if you run “purge /usr/bin”, you can kiss-off a lot of your Linux distro’s system files, and your install of Ubuntu or Fedora will almost certainly be wrecked.  In Linux lingo, they call this “borking the install”.

NOTE that you can also usually check/verify a command before you run it, by opening a Terminal, and typing “man, space, <the command in question>”.  Do this without the quotation marks and commas that are in this pretend example.

For example, if curious about “rm -r”, I could go to Terminal, and type:

man rm

and then hit Enter.

I will add quickly here that I have yet to see a varifiable instance of some novice being “tricked” into “borking” his or her install of desktop Linux in such a way.  I find that the people on forums and in chat-rooms (such as #ubuntu IRC) really are there to try to ** help ** us, and I perceive very few shenanigans.  This is probly because Linux (desktop), even moreso than other end-user systems (a’ la’ Winows and MAC), is, well, * harder *—at least until you get used-to it.  And even then. Desktop Linux, dude, is going to make more demands on its user-base (which means YOU and ME), than the other two major platforms.  But it offers greater (potential) rewards.

The main reasons the Linux desktop is a little more demanding, is that A) it * is * not * bundled hardware—meaning that desktop Linux—like, say, Ubuntu—does not usually come pre-installed with a computer you buy new (or even used) from a store.  Main-line hardware “vendors” like Dell and HP either do not offer such machines, ready to take home and use, or else they just do not make much fanfare of it when they do.  Which makes * pre-installed * desktop * Linux * enough of a rarity, that little bootability and related hardware glitches never get enough attention from the in-house engineers.

At least not to the degree that MS products receive. Windows comes pre-installed (“bundled”) with most computers you buy today.  Windows dominates the market in end-user operating systems, even at the time of this writing (late 2012).  And so you can bet your sweet bippy that the engineering staff at the major manufacturers spends considerable time making sure that any small glitches—in re-booting or using ms Windows—on their models about to enter the product-stream, are solved * before * you’re handing your Visa card to the salesperson at the store.  Microsoft “dines out” on this “free lunch”, bought by its market-share dominance.

And B), desktop Linux—with the exception of a few distros—is * community-driven * * FOSS-ware *.   Free Open-Source Software developed by “volunteer” communities, most of which, arguably, do not actually meet, but exist only in “cyber-space”.  Which works remarkably well.  Wikipedia uses essentially the same model.  The difference it makes, though, is that en.wikipedia.org * is * not * coding * an *  ** operating ** system **.  They’re compiling an * encyclopedia *.  Which is hard enough.  But you know, if you erroneously “learn” that, say, Owensville, Missouri was the first city west of the Mississippi to legalize gay marriage—and then later find out it was a hoax—it won’t freeze-up the GUI of your computer, causing you to have to use ctrl + Alt + F1 to log-in at an x-window, and then type “sudo halt” (and again authenticate with your password), in order to shutdown and boot-back-up.

I * love * the desktop Linux community, and I * love * its values.  Just as I love Wikipedia.  But it’s like I said (if I can refrain from pushing the analogy too far); a good encyclopedia is one thing, BUT an * operating * system * is another.

Add to this that * some * of the hardware makers (of both computers * and * peripheral devices—like USB web-cams and printers) just do not have time to cooperate with a Linux developer who’s calling for help in writing a driver for * ~ * whatever, and you find that desktop Linux * has * some * usability * headaches * that Windows doesn’t.  Having said this, it is almost * astounding * how many of the hardware makers * DO * seem to have time to help Linux devs write drivers.  Maybe they’re trying to tell us something (?).

One must add quickly that Windows * does * have some usability headaches—especially if you run it “by-the-book”; I can spend (and * have * spent) all day (and part of the next) trying to find a certain good free app to do * ~ * whatever task on the Windows platform, that has * good * reviews, and is free of malware.  And there are other examples.  MAC is not left-out, either.  Try taking your MACBOOK Aire to the Public Library in some don’t-blink-or-you’ll-miss-it town, and printing something-out on their old Lexmark.  See how far you get.

Desktop Linux has * particular * issues *, however.  I hate to keep harping on it, but it boils-down, really, to A) market-share prevalence / the “bundled-hardware” issue, and B) the fact that the distributed desktop Linux developer community is * free *—like people are free in a * Democracy *.   Which basically means that Linux is designed “by committee”.   I do not want to compare Ubuntu with, say, the American House of Representatives’ Ways-and-Means Committee.  Ubuntu is about a million times more efficient.  But it still possesses at least one feature of such “democratic” institutions:  people sometimes fall-out with one-another, and their cooperation is afterwards diminished.

And sometimes two persons just have difficulty communicating.  Yes, this could be a problem at a large * company * that is in the operating system business (like for example Microsoft).  But having had some experience—first-hand—with corporate hierarchies, I can attest that this “human factor” is * tempered *, in the corporate world, with things like, for example * fear *.  Like for example the fear of being * fired *, for one.  And let’s not forget that a * corporation * is unlike a “community”:  where it comes to an Inc., or a Ltd., or even an LLC, there is at least one feature that is distinct, from some “community-driven” organization.  Namely, it is that somebody is “God” of the company; and he (or she ) can strike a much more serious financial blow upon any person involved in the project.  This often has a sobering effect upon personalities that might otherwise find it difficult to get on the same page.

Desktop Linux, for all its being handicapped by its “democratic tendencies” and “resistance to being evil”, does remarkably well.  But there will always be little things wrong with it.  * Always *.  Even were it to become the dominant platform.

Do a little research, on any Linux command you intend to run.  This is just common sense.

41. Entry-41:  If you’re stuck, and really having difficulty getting things to work, Googling around for “Post-Install Checklists” for your distro (i.e. “Ubuntu 10.04 post install checklist”) can help shed some light.  I do not endorse these checklists—not any particular one; but the information can be helpful.  You might also try Googling “thenameofyourdistro ~whatever it is + “perfect desktop”.  As with all programs—Linux and otherwise—I exhort you to * research * any program you intend to install, at least somewhat, before installing * anything *.  Especially if it is not available from your default repos.  And I prefer not to install Ubuntu Tweak.  I prefer to learn my way around my os well enough to do my own tweaking.

42. Entry-42:  Copying and pasting into and out-of a Terminal window:

First, let’s note that * some * distros don’t want to let you execute certain commands/other stuff, * if * you * pasted * it * into * the * Linux * Terminal * emulator.  Why?  This is apparently a safety setting, to guard against unauthorized persons (blackhat-hackers) running scripts surreptitiously on your machine.  Try typing the same thing manually, proof-read it, and * then * hit that Enter key.  This behavior will be true for certain distros, and sometimes just for certain types of commands in a distro.  Consult your distro’s documentation.

Copying and pasting into and out-of a terminal window does not work using the traditional keyboard commands [read on].  Commands or other data should never be pasted into the Terminal from an ordinary document or word-processing shell anyway.  Us newbies should always use a GUI text-editor for this.  Just look for a desk-top note-taker app that knowledgeable Linux ppl tell you is acceptable for this purpose.  Often you will then be able to find it in your menus, because most builds come with one.  G-Edit and Leafpad are two popular ones.  You do not want something that puts hidden formatings that you don’t see with your eye into the Linux Terminal, because this can have a way of “confusing” the Terminal-programming—and that is what normal documents-creator programs like Word, Works, Abiword, or OpenOffice Writer will put-in.  Two common Linux apps that are acceptable to use as GUI text-editors for pasting into the terminal are 1) G-edit, and 2) Geany.  There are more.

In order to paste a line of text or code into the Linux terminal, highlight it, ctrl + c to place it on the clipboard; then click the mouse in the terminal, watch where the cursor-bar is flashing—that’s where the paste will start-from; then ctrl + SHIFT + v.  This usually works.  There are other methods detailed online, or in this data-base.  Just using a “wheel click” / “middle-click” may work:  consult online documentation appropriate to your distro.

To copy from the terminal, drag the mouse in the Terminal to highlight what you want; then ctrl + SHIFT + c.  Or I think you can try clicking the mouse-wheel.  The wheel of your mouse—if you push-down on it hard, and then immediately release it, will produce a special, “middle-click”.

43. Entry-43:  Sometimes the “paste-special”, or “paste as unformatted text” feature in Open Office Writer (or even Libre Office) remains greyed-out, even though you have an amount of information copied to the computer’s clipboard.  A workaround is to simply ctrl + v, pasting the text as formatted; then, highlight it with the mouse, ctrl + x, and then try again to paste it with the ribbon-icon.

Where I have a * large * amount of data I want to paste, and the “paste as unformatted” feature is greyed-out (usually because it’s the first paste to that document in today’s use-session), I find I will just leave the content  I wanted to capture highlighted, then switch to the document I’m having trouble with, highlight some brief, random line of plain text in it, ctrl + c, then paste * that * to neutral space somewhere, then highlight it, then ctrl + x to cut it to clipboard, then go to Libre Office Writer’s ribbon icon for clipboard, and find that “paste as unformatted” is now available.  I ctrl + v to paste the brief line, then use Backspace key to remove that line.  This has caused the “paste as unformatted” feature to “get going”.  It will usually continue to work from now on, in * this * file, for the rest of my session.

Is this an annoying bug?  Maybe.  But I will say that my own cloud-and-database contains more than 1.100 files—mostly .odt and .doc.  And I don’t notice this bug much anymore.  I’ve gotten used-to the workaround.  I do it as second-nature.

44.  Entry-44:  As to upgrading a Linux distro that you  have installed:

Don’t use the “Upgrade” button, as in “Upgrade to the next release of your distro”.  This is often found in the Updates-Manager.  I don’t think this really works very well.

If your distro publishes an “alternate install cd”, boot-up from cold with that, and then use the installer.   You should always * test * the new release you intend to upgrade to, though—test its regular build that you intend to be the final result—by running it as a * live boot *—just as you did before you installed Linux for the first time.  Try its browser, and make sure that all the features you intend to use will work with your system hardware.  You should probably never assume that the next release of the distro will have just as good hardware recognition as the one that you are currently using.  Sometimes, a driver will have been changed, or some related “interfacing software”.

The system should be able to download and install any other needed files from a broadband connection.  Sometimes it calls this “completing the install”.  It should successfully over-write your existing install of Ubuntu, yet preserve personal files.  But we cannot rely upon that to be “perfect”.  SO IT IS ENCUMBENT UPON YOU TO HAVE * BACKED-UP * ALL YOUR PERSONAL STUFF.  Be sure that you checksummed both discs first.  Don’t try to “leap-frog” over the next release to an even newer one [unless you’re gonna do it the “Linux Mint way”, and completely uninstall your existing Linux, first having backed-up all personal data, then wiping the partition with something like G-Parted].    Never try to upgrade your desktop system * with * a desktop-version of the new release of the distro’s cd—use the alt-install:  unless you wipe the partition with G-parted first, and then install into the empty space.

If you use this latter method, you can “leap-frog” to a release another (or several) versions higher.  At least in theory.  Remember that in personal computing—as with many other spheres of our existence—theory does not always work itself out so great in practice.  Remember also to leave the cd tray open when you shut-down to change discs.  If you wipe Linux and then do not install the new one (or re-install the old) right away, your pc may not boot Windows again (or much else), without some repairs (like perhaps having to use “Restore MBR” from a repair disc).

45. Entry-45:  As to naming files, for compatibility on Linux and dos-based systems:

NO FILE that is going to be used in Linux should begin with the name of an Application.  A file pertaining to how a user should operate Firefox (I mean some directions) should probably not be called “Firefox ~”.  You might try something like “f-firefox ~”.  I prefer to use all lower-case for file-names, with a space between words, and without underscores.  Truncating/abbreviating long words is often somehting I do, but is probably not necessary.

46. Entry-46:  Need to edit a system text-file?

Well, unless you’re a full-fledged developer, I think I’d stay away from the VI terminal text-editor for Linux, if I wuz you.  Nano or Pico are often installed for this purpose, by default.  These two are easier, and have some safety-devices in place, to help a lesser-experienced person avoid a mistake.  Also, there are usually ways to make it happen with g-edit, which is not a terminal-based text-editor, but is its own GUI editor.  It also doubles as a note-taker, and so is just like Windows’ Notepad, which does the same two tasks in ms Windows—though most Windows users have never needed to edit this type of text-file.  You’re becoming a * Linux * user:  welcome to the elite!  But it is really not that hard.

47. Entry-47:  DOCUMENTS IN LINUX:

Open Office is a Free and Open Source Software (FOSS) replacement for programs like ms Office, Corel Office.  The mainline edition comes with a documents program (one of the best—pay or free, at least as far as the 3.3 version), spreadsheet, and all the related stuff.  This office suite is installed by default with many Linux distros.  If it does not come with your distro of choice, you can usually download and install it without much fuss.

Open Office seems to open all formats now–.doc, .docx, .odt, .txt, &tc, &tc.  Or there is a Linux app that will.  MS Office, equally—whether v. 2007, 2003, or 2010—seems to be able to open anything that was created on a Linux machine.  I will mention, too, that I tried to open a .docx file on my friend Jim’s XP machine, and his Word  2000 would not open it.  I also tried to open the same document (and a couple others) that were .docx, using an old lappy running Karmic Koala (nickname of Ubuntu 9.10), using Open Office 3.1.  And it could not open them.  But Open Office 3.2 seems to have no problem opening and working in .docx (at least not * now *, as of late 2011).  And Libre Office (the new iteration of Open Office) I have found will open and work in any of these formats.

At the time of this writing, Linux comes with NTFS support.  Or this could be downloaded, if for some odd reason your distro does not include it.

If you are familiar with the function, you can use a Macro in Word 2007 to convert documents from .docx to .doc (or .odt) as a “batch process”—perhaps letting it run overnight.  There are other Batch-files methods available to convert a whole bunch of them in one fell swoop.  Just look around on the web.

I find that Open Office Writer 3.2—which is what you use in Lucid (Ubuntu 10.04)—seems like  the equivalent of something from 1998, though I hasten to add that this can actually be * better * in several ways than some of the later word-processors—these being somewhat overdone, from the standpoint of an ordinary home-user.  I really should  try to install Libre Office—I wonder if there’s a version of it that runs nice and stable on Lucid?  Open Office 3.2 is nice.  But it has a harder time opening and closing big docs with a lot of formatting (.doc, I emphasize:  I’ve switched to .odt/.odf, since I first wrote this, and the issue essentially went-away in Linux—but now I have it in Windows 7’s MS Office Word; Word 2007 stumbles-and-chokes, taking noticeably longer to open my documents now that they’re formatted to .odt).

My solution was to download and install Open Office 3.3 to Windows 7.  Now it works smooth.  But if you’re going to have to work part of the time on a system that won’t let you install Open Office (because the person in charge of the equipment won’t allow it), then maybe you’d better think twice about migrating your documents to .odt.  I’ll add only that .odt takes-up * way * less space on your harddrive (or any location) than .doc, and is apparently much more secure (Google it).

UPDATE:  I had to delete my whole extended partition recently, in order to re-size my Windows 7 Starter, as it was running out of room for updates.  I wasn’t completely happy with Linux Mint 12 GNOME-build—though I emphasize that my qualms were not related to the Desktop Environment—the LM team did an excellent job of creating a very usable (and versatile) DE from the GNOME3 “mess”.  It was some other little bugs that got me down.  When the LM 12 live DVD wouldn’t connect to the wifi at my friend’s place of business—where I found myself re-allocating my disk—that was the last straw.  Chances are probly about even that a re-boot would have solved it.  But I remembered that LM 13 (“Maya”) had just reached final release, and that I had downloaded and md5-checked it (the XFCE build) sometime last week.  I went for the XFCE as it is now Ubuntu-based, instead of based directly on Debian (I have trouble with Deb-based booting on my equip.).  And reviews of the other -13 releases poo-pooed the other desktop mixes available in Linux Mint.  LM 13 XFCE comes with Libre Office 3.  Well, I now find that my Linux desktop will open and deal with .doc * on-par * with Windows’ Word 2010 Starter!  What a pleasant surprise!  Thanks, Linux!

Good ‘ol Linux Mint 13 XFCE even found the FireFox bookmarks I had forgotten to back-up in my LM 12 GNOME i386 install, and imported them!  (And some system settings, too!)  What a feat, considering I had completely wiped the disk (except for Windows 7 Starter)—having completely deleted the logical-drives I had been using—and the over-riding Extended partition in which they had been housed.  So not only had I “wiped the partition”:  it is more to the point that * the * partition * ** no ** longer ** *** existed ***.  And on top of that, I had created a * new * extended partition, and had installed Linux swap, and Ubuntu 10.04 * over * part of where Linux Mint 12 Gnome had been.  But Linux Mint 13 XFCE was still able to recover those Firefox bookmarks.  Awesome!  Now, ** that’s ** some Linux ** power ** for ya!

Features are a little more cumbersome (to get used-to) in OO Writer 3.2, and some are lacking [where is “toggle case?; “continue numbering” (where you’ve left-off previously) is more involved than in ms Office Word 2007].  It is also true that I had to fork-out about 130 USD to Microsoft for its Office suite, and it jammed-up one of my big, highly-formatted files in .docx, and I could only recover it by closing it with WINDOWS Task Manager, and then re-opening it (the next day) with ms Works.  Then I had to re-create it by successive copy and paste back into Word ’07.  And even then I could not recover all the images, so those were just lost (note also that this was when I was still having to use MS’s Vista operating system, as I had not picked up my Win 7 machine from the store yet).  So perhaps it should be said that neither system is perfect—neither Linux nor * Windows * .

In either case, if I had only been using Dropbox at the time I was creating this file in the first place, I could’ve just gone to my cloud and retrieved a slightly older version, and saved most of the lost images.  I must say I find Dropbox works every bit as well in Ubuntu and Mint as it does in Windows 7.  And I will add that I am a user who has more than 800 + documents of all sorts (my co-major was Divinity Studies)—several running into more than 150 pages, with images and screenshots.

Nothing against .odt—really, the whole world should be using this standard.  But as I have said, some Windows users (particularly those with older versions of ms Office—and there are plenty of them out there, at the time of this writing—2011), well, some of these persons may not be able to open .odt without installing some update or plug-in, or they may just object to having to deal-with this type of “weird” format on a regular basis.  Many ppl are very conservative, where it comes to their Windoze install—perhaps to a large-measure because they have just learned to fear “viruses”—to live in fear of “viruses”—whatever that means to  them, for whatever reason.  F.U.D. (Fear, Uncertainty, and Doubt) is a factor in computing—even in the Linux meta-verse; but it is more pronounced in the Windoze meta-verse, and this for cultural reasons, as much as because of the software itself.

In any case, what I find * NOW * is that Libre Office 3 (in my Linux Mint 13 XFCE Ed.) handles my .doc files just as well as those formatted to .odt; really, it handles * BOTH * equally well.  And I find all the features are enabled, by default.  Navigation / Search in the “elevator-bar” track on the right.  Everything works, and it’s a breeze!

48. Entry-48:  SAFE REMOVAL OF EXTERNAL USB HARDDRIVES and THUMB-KEYS:

Remember that there’s * hardware * mounting (physically plugging the device into a hardware port somewhere on the computer’s case), and there’s * software * mounting—which basically means establishing some kind of software connection (“interface”) between your operating system and the software inside the device.

Where one is using/is currently in a Linux Desktop Environment (a.k.a. the GUI user-interface) * OTHER * THAN * GNOME *, many drives will not ‘disconnect’ (I mean where the light goes out and drive becomes inactive) in an orderly fashion the way they do under windows.  [Quite frankly, my WD Passport doesn’t seem to have it’s light go out when I try to unmount and have it go out in my Windows 7 Starter install.  Maybe I need to go through and “safely remove” every partition on it, now that I have it divided into several partitions. In GNOME 2.x, I notice that if I just “unmount” one of its partitions that I have formatted to FAT32 (with a Right-click), this unmounts the whole device, ** and ** it ** also ** “powers it down” (the little “powered-on” light on the thing goes out).]

Just pulling the USB cable usually seems okay—especially with WD Passports:  BUT FIRST it is IMPORTANT to make sure no data is still being transferred; RIGHT-click the  icon of any partition(s) on the drive that you think might still be busy, and select “unmount” or “eject”.  If the volume is still being written-to/is still “busy”, Mint will usually tell you, and you can come back to it later.

The “umount” command does not cut the power:  GNOME does.

There are also the Eject command, and hdparm command—as in
“hdparm -y /dev/sda1”

In many releases of Ubuntu, one might have the Eject accessible in the context menu (the RIGHT-click menu).  The “eject” command can also be run from Terminal.  These two are not necessarily the exact same command:  one is GUI, the other, CLI:  their underlying structure may vary.
If you want to be extra safe, you could run ‘sync’ before unmounting or ejecting. Both the Unmount and Eject mouse commands should do this, but it doesn’t hurt anything to do it manually.  If you have ever noticed that it takes a while for Unmount to finish, it is because it is running sync before unmounting.  And, Eject (from context menu) runs the umount command before doing an ‘eject’.  If you have run sync and allowed it to finish flushing the write buffer, you should be able to simply unplug the device without harming it -but I wouldn’t do that on purpose.  The device would still show to be mounted even when physically removed.

The difference in the way different drives react to the commands has to do with the nature of the device.  Real hard drives may not register as removable devices, whereas USB-flash devices usually will.  Also, if you are using a card-reader, insert and remove evidently are not sent to the driver and the same may happen if using an extra hub between the device and the computers’ USB port.

When Windows shuts down a USB drive,
it turns off the power to that drive…which turns off that steady,
orange “power on” light.. [I need to add quickly here, however, that in my manufacturer’s install of Windows 7 Starter on my AAO netbook, Windows DOES NOT turn off the light EITHER.  HARRRUMPH! ]

When Linux un-mounts the drive it (no surprise) un-mounts the drive, only…

Since mount knows how to mount and unmount, but not turn power on or
off, it does nothing with the power flowing to the enclosure…so, the
light stays on..
but, the disk IS unmounted, and there is NO activity *if* you can’t
see it as mounted, and you’ve checked by means of RIGHT-click + “Unmount”, to see if there’s still any data-transfer—even if only at buffer-level.

Some of us are a little bit worried about possibility of bad sectors developing in the HD, since one of the main causes of developing bad sectors in a Hard Disk is voltage fluctuation; and an improper spin-down may lead to voltage fluctuation & (maybe) bad sectors.

HERE’S SOME ADVICE:

http://www.linuxquestions.org/questions/slackware-14/safely-remove-usb-external-hard-drive-703326/

On the subject of quietly disconnecting my WD USB drive, I found the most ‘comforting’ and quiet way is to first ‘unmount’ in KDE then run the following bash script:

Code:

#!/bin/bash
echo “Sync and Flush cache”
hdparm -f /dev/sda
sleep 2
echo “”
echo “Now Drive to Stand-by”
hdparm -y /dev/sda

This should cause the external hdd to “spin-down” and shut-off (it is to be hoped), much as in Windows.
-Bill.
http://sekharpadikkal.wordpress.com/2008/03/01/safely-remove-an-external-hard-disk-in-linuxtested-in-ubuntu/
What Ubuntu doesn’t seem to want to do is powering down the [external] hard drive. You will notice this when you unplug the USB cable, because of the nasty sound coming from the hard disk, similar to the one you would hear if the power supply got suddenly cut off.
Thanks to this post on Ubuntu Forums, I’ve figured out the solution.

First of all, you will need to install sdparm:
sudo apt-get install sdparm

At this point, unmount the drive and then issue the command:
sudo sdparm –command=stop [device]

Replace [device] with whatever you need, and that’s it: your hard disk will spin down and rest, in the same way it would do in Windows after a safe removal. Now you may safely unplug your hard disk.
Notes

This how-to is based on my experience with a Western Digital WD Passport hard disk.
NB: it looks like this how-to won’t work with some Lacie external hard disks.
HERE WE HAVE AN ACTUAL PROGRAM, WRITTEN IN BASH:
http://forums.debian.net/viewtopic.php?f=5&t=52627
The author has written and linked to a script to suspend usb devices so that they comply to the manufacturer’s conditions for safe removal.
This is a pretty nice program, but, apparently, USB suspend is no longer possible under 2.6.34 kernel.  So this program may not be effective on systems booted on the 2.6.34 kernel and + up.  😦

Code:
#!/bin/bash
#
#  suspend-usb-device: an easy-to-use script to properly put an USB
#  device into suspend mode that can then be unplugged safely
#
#  Copyright (C) 2009, Yan Li <elliot.li.tech@gmail.com>
#
#  This program is free software: you can redistribute it and/or modify
#  it under the terms of the GNU General Public License as published by
#  the Free Software Foundation, either version 3 of the License, or
#  (at your option) any later version.
#
#  This program is distributed in the hope that it will be useful,
#  but WITHOUT ANY WARRANTY; without even the implied warranty of
#  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#  GNU General Public License for more details.
#
#  You should have received a copy of the GNU General Public License
#  along with this program.  If not, see <http://www.gnu.org/licenses/&gt;.
#
#  To reach the auther, please write an email to the address as stated
#  above.

#  ACKNOWLEDGEMENTS:
#      Christian Schmitt <chris@ilovelinux.de> for firewire supporting
#      David <d.tonhofer@m-plify.com> for improving parent device
#      search and verbose output message

usage()
{
cat<<EOF
suspend-usb-device  Copyright (C) 2009  Yan Li <elliot.li.tech@gmail.com>

This script is designed to properly put an USB device into suspend
mode that can then be unplugged safely. It sends a SYNCHRONIZE CACHE
command followed by a START-STOP command (if the device supports it),
unbinds the device from the driver and then suspends the USB
port. After that you can disconnect your USB device safely.

usage:
$0 [options] dev

sample:
$0 /dev/sde

options:
-l     show the device and USB bus ID only
-h     print this usage
-v     verbose

This program comes with ABSOLUTELY NO WARRANTY.  This is free
software, and you are welcome to redistribute it under certain
conditions; for details please read the licese at the beginning of the
source code file.
EOF
}

set -e -u

SHOW_DEVICE_ONLY=0
VERBOSE=0
while getopts “vlh” opt; do
case “$opt” in
h)
usage
exit 2
;;
l)
SHOW_DEVICE_ONLY=1
;;
v)
VERBOSE=1
;;
?)
echo
usage
exit 2
;;
esac
done
DEV_NAME=${!OPTIND:-}

if [ -z ${DEV_NAME} ]; then
usage
exit 2
fi

# mount checking
if mount | grep “^${DEV_NAME}[[:digit:]]* “; then
1>&2 echo
1>&2 echo “the above disk or partition is still mounted, can’t suspend device”
1>&2 echo “unmount it first using umount”
exit 1
fi

# looking for the parent of the device with type “usb-storage:usb”, it
# is the grand-parent device of the SCSI host, and it’s devpath is
# like
# /devices/pci0000:00/0000:00:1d.7/usb5/5-8 (or /fw5/fw5-8 for firewire devices)

# without an USB hub, the device path looks like:
# /devices/pci0000:00/0000:00:1d.7/usb2/2-1/2-1:1.0/host5/target5:0:0/5:0:0:0
# here the grand-parent of host5 is 2-1

# when there’s a USB HUB, the device path is like:
# /devices/pci0000:00/0000:00:1d.0/usb5/5-2/5-2.2/5-2.2:1.0/host4/target4:0:0/4:0:0:0
# and the grand-parent of host4 is 5-2.2

DEVICE=$(udevadm info –query=path –name=${DEV_NAME} –attribute-walk | \
egrep “looking at parent device” | head -1 | \
sed -e “s/.*looking at parent device ‘\(\/devices\/.*\)\/.*\/host.*/\1/g”)

if [ -z $DEVICE ]; then
1>&2 echo “cannot find appropriate parent USB/Firewire device, ”
1>&2 echo “perhaps ${DEV_NAME} is not an USB/Firewire device?”
exit 1
fi

# the trailing basename of ${DEVICE} is DEV_BUS_ID (“5-8” in the
# sample above)
DEV_BUS_ID=${DEVICE##*/}

[[ $VERBOSE == 1 ]] && echo “Found device $DEVICE associated to $DEV_NAME; USB bus id is $DEV_BUS_ID”

if [ ${SHOW_DEVICE_ONLY} -eq 1 ]; then
echo Device: ${DEVICE}
echo Bus ID: ${DEV_BUS_ID}
exit 0
fi

# flush all buffers
sync

# root check
if [ `id -u` -ne 0 ]; then
1>&2 echo error, must be run as root, exiting…
exit 1
fi

# send SCSI sync command, some devices don’t support this so we just
# ignore errors with “|| true”
[[ $VERBOSE == 1 ]] && echo “Syncing device $DEV_NAME”
sdparm –command=sync “$DEV_NAME” >/dev/null || true
# send SCSI stop command
[[ $VERBOSE == 1 ]] && echo “Stopping device $DEV_NAME”
sdparm –command=stop “$DEV_NAME” >/dev/null

# unbind it; if this yields “no such device”, we are trying to unbind the wrong device
[[ $VERBOSE == 1 ]] && echo “Unbinding device $DEV_BUS_ID”
if [[ “${DEV_BUS_ID}” == fw* ]]
then
echo -n “${DEV_BUS_ID}” > /sys/bus/firewire/drivers/sbp2/unbind
else
echo -n “${DEV_BUS_ID}” > /sys/bus/usb/drivers/usb/unbind

# suspend it if it’s an USB device (we have no way to suspend a
# firewire device yet)

# check if CONFIG_USB_SUSPEND is enabled
[[ $VERBOSE == 1 ]] && echo “Checking whether $DEVICE can be suspended”
POWER_LEVEL_FILE=/sys${DEVICE}/power/level
if [ ! -f “$POWER_LEVEL_FILE” ]; then
1>&2 cat<<EOF
It’s safe to remove the USB device now but better can be done. The
power level control file $POWER_LEVEL_FILE
doesn’t exist on the system so I have no way to put the USB device
into suspend mode, perhaps you don’t have CONFIG_USB_SUSPEND enabled
in your running kernel.

Read
http://elliotli.blogspot.com/2009/01/safely-remove-usb-hard-drive-in-linux.html
for an detailed explanation.
EOF
exit 3
fi

[[ $VERBOSE == 1 ]] && echo “Suspending $DEVICE by writing to $POWER_LEVEL_FILE”
echo ‘suspend’ > “$POWER_LEVEL_FILE”
fi

Found this with google search “linux command safely remove usb”

http://forums.debian.net/viewtopic.php?f=5&t=52627

Re: “Safely remove” equivalent in CLI?

Postby bash » 2010-07-26 15:07

I’ve found one way this could be done via echoing “suspend” to power/level of the appropriate device, but this has been removed in 2.6.32 kernels and I’m using 2.6.34 since today.

Hi, I had the same problem and found 2 solutions to it:
1. You should find you device in sysfs and then echo “auto” to power/level. Than you have to unbind you device as described here http://lwn.net/Articles/143397/ For example, I have my external HDD on fifth port of first hub. And than my command looks like this:

Code: Select all
echo “auto” > “/sys/bus/usb/devices/usb1/1-5/power/level”
echo “1-5:1.0” > /sys/bus/usb/devices/1-5\:1.0/driver/unbind

Default timeout is 2 seconds on my machine. To speed things up you may consider using:

Code: Select all
echo “0” > “/sys/bus/usb/devices/usb1/1-5/power/autosuspend”

It will allow device to suspend immediately if there are no pending operations.

2. Simpler way is to use remove attribute of the device:

Code: Select all
echo “1” > “/sys/bus/usb/devices/usb1/1-5/remove”

Both solutions works for me, but slightly different. In first case device can’t be powered down while it is mounted or there is any operations with it. Also, device can be easily bounded back by using bind attribute. In the second case you can run into trouble because it is possible to remove mounted device, WOW! Also, I didn’t find a way to power up this device again without replugging it. Maybe these are not the best solutions but they are the only I found so far.

Re: “Safely remove” equivalent in CLI?

Postby juru_piotr » 2012-02-05 11:31
Sorry, but I have to refresh this old topic – I’ve found more appropriate solution. It should be done by means of “udisks” command 😀

Code: Select all
udisks –unmount /dev/device && udisks –detach /dev/device

Solution found on http://ubuntuforums.org/showthread.php?t=1821645

Advertisements

10 thoughts on “The Most Important Things a Linux Desktop Newbie Needs to Know

  1. I just want to tell you that I’m newbie to weblog and truly loved your blog site. Most likely I’m planning to bookmark your blog . You certainly have excellent articles. Many thanks for sharing with us your web site.

    • Thanks, Martomed; I’m flattered: but you are giving the credit (for webmastering) to the wrong party: It isn’t ME, it’s wordpress’ software–and the fact that my work is TEXT-ORIENTED–very few images or vid/active-content. But I’ll accept the “masterpiece” part of your kind compliment, Thank You. -L.L.

  2. Hey there just wanted to give you a quick heads
    up and let you know a few of the pictures aren’t loading correctly. I’m not sure why but I think its a linking
    issue. I’ve tried it in two different internet browsers and both show the same results.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s