Chad Perrin: SOB

3 January 2008

Closed-source glitz — missing the point.

Filed under: Cognition,Geek,Metalog — apotheon @ 04:56

Robin Harris says 2008 is Linux’s year on the desktop. I don’t know that I buy that tight a timetable for open source operating systems to rule the world — and I think he’s a little myopic in thinking Linux will be the one and only major alternative to MS Windows to make such strides. I think when it happens Linux may well be the vanguard, but Macs, Solaris, and BSD Unix systems will be riding in its wake. Realistically, BSD Unix OSes will probably take over the place Linux once had on the desktop (but with more poise and less of the feather boas and dancing routines), while Linux takes a slightly larger role, and OpenSolaris will share a fair bit of server space with BSD Unix. That is, I believe that’s what’ll happen barring unforeseen eventualities that could throw everything out of whack.

. . . but let’s just go with Robin’s “optimistic” (depends on your priorities, I guess) view of things. For the record, I think it would be a step away from the absolute wrong direction, but not strictly the right direction, as such. Anyway, we’re assuming for the moment that Robin is right about how things will go if Microsoft and Apple sit still and let it. Now, it’s time to examine what they might do.

We’ll take the predictions of Adrian Kingsley-Hughes as your cue for this one, in his response to Robin’s take on the future. He tells us How Microsoft (and Apple) will respond to very-low-cost Linux systems, at least in his estimation.

Adrian says:

That’s easier than you think you know. You see, take a look at either Vista or Leopard (it doesn’t matter which) and what do you see? That’s right, a rich, media intensive platform that’s stuffed full of eye candy. It’s no accident that both Microsoft and Apple are embracing media and eye-candy with enthusiasm – this is a very deliberate business choice that both companies have made.

He’s not a man of deep examinations of his own predictions, apparently, because that was almost a third of the entire weblog post. I’ll try to summarize the rest. Italicized text is a paraphrase:

MS and Apple are using their ability to crank out high-end graphical goodies and handle multimedia spiffiness to put those Linux cavemen to shame. Linux can’t handle multimedia content and 3D graphical GUI environments, you see, since they all use text-based interfaces circa 1969. All those cheap, low-end systems being sold with Linux on them will fall by the wayside as people see the Light, and run back to the arms of King and Queen Bloat. Plus, y’know, there isn’t even a price difference, ’cause Linux is too late to the party.

I stripped away a bit of Adrian’s veneer of reasonability there, but it was pretty thin to begin with, so I’m sure nobody will notice. Anyway, now that we’re done with the rosy-lensed Linux-cheerleader prediction, here’s what I think of the Evil Empire cheering section’s predictions:

  1. Microsoft and Apple will of course keep making high-glitz systems.

  2. Microsoft and Apple will also try to get into the low-cost market. Microsoft will fail horribly (look at the example of Vista Home Basic) to make a positive impression. Apple will make surprising headway with things like the iPhone and the Mac Mini serving as early steps in that direction, but all the “real” Mac users will still want the high-end systems Apple offers — after all, that’s what Apple is good at: high-end hardware with a glitzy straightjacket OS. Microsoft’s almost as good at the straightjacket, but its glitz always ends up looking like fifty-seven pounds of pink taffeta.

  3. They will indeed hope their high-glitz systems will hold a niche Linux can’t break into. Of course, unlike Adrian Kingsley-Hughes, they know what’s actually going on out there — they won’t believe that as long as people want glitz they’ve got an unassailable foothold against the encroachment of open source systems on their market. They’ll have to hope that nobody catches on to the fact that open source systems are currently better at glitz and glamour than Microsoft and Apple systems. The order of stylish, shiny eye candy from best to worst goes something like: Compiz Fusion (window manager for the X Window System); Aqua (the MacOS X GUI); Aero Glass (the MS Windows Vista GUI); A bunch of other X Window System window managers like Enlightenment; Whatever Microsoft calls the “no Aero Glass for Home Basic” GUI interface. The order of bloated, RAM-sucking, processor-burning resource consumption from greatest (worst) to least (best) goes something like this: Aero Glass; Aqua; MS Windows Vista non-Aero interface; Compiz Fusion; KDE and GNOME; Enlightenment; All the rest of the X Window System window managers. Note that I’m not entirely sure that, for resource waste, Aqua is fatter than Vista Basic, or that Compiz Fusion is worse than KDE and GNOME.

  4. So . . . open source OSes like Linux and BSD Unix systems (and even MacOS X) can do better in the glitz arena with less resource-suckage than MS Windows, and MS Windows is the only OS family that does worse than MacOS X on either score, overall. Far from being primitive caveman stuff that can’t stand up to the superior technical acumen of those paragons of programming innovation in Redmond, the open source development community seems to have mastered the art of producing prettier, fancier software with less egregious resource suckage — and somehow, Microsoft Windows still can’t provide multiple workspaces worth a damn, despite the fact it’s about fifteen year old technology at this point. Should I go on? How about proper application pagers, journaled filesystems, filesystems that don’t need defragging, true privilege separation, and the ability to extend one’s filesystem by mounting other filesystems at arbitrary points with whatever permissions you want? Hey — at least MS Windows comes with a kitchen sink. Wait — I forgot a really important one: good, comprehensive, stable, secure software management. Microsoft still has no clue how that works, apparently.

  5. Barring dirty tricks (which are sure to arise, with MS Windows and Apple MacOS X in the works), trying to hold the “high ground” through maintaining consumer ignorance while giving up some of the low ground is a losing proposition. After all, as more and more people who want cheap but capable systems are introduced to open source OSes, familiarity with the scary world of “not what came from Best Buy” will grow in the general populace, who will suddenly realize just how easy this stuff really is. It’ll start growing. It’ll spread more quickly, once it passes a particular tipping point, because it’s free — not only free of charge, but free for copying, redistributing, selling, modifying, playing frisbee with the CDs, whatever. Eventually, high-end systems will start cropping up with Compiz Fusion and World of Warcraft running on Free/Libre/Open Source Unix-like OSes. Of course, it won’t get that far easily. As I said, they’ll fight dirty.

So, now that we’ve settled that . . . barring legislation and litigation tailor-made to destroy open source software development, Microsoft and Apple will have to come up with a lot more than just holding the line on glitz-heavy OS interfaces to stem the rising tide of open source software adoption. In fact, the way things have been going for Microsoft lately, it’s going to have to undo some of the damage it has done to itself by driving away many of its customers. Oh, sure, not a majority, or even a significant minority percentage (yet) — but certainly enough to sit up and take notice, and to wonder how things will go this year.

This year. Man. Is it 1998 already?

(edit: I had accidentally said iPod at one point when I meant iPhone. That error has been fixed.)


  1. I think that the Achilles heel of Windows and OS X is that they are utter resource hogs. I cannot speak directly to OS X (odd, I can troubleshoot a Mac over the phone, but I haven’t used one personally since 2000), but Windows has always managed to be a pig. Indeed, every new version of Windows seems to be taxing on even the beefiest system when it is released, and does not run nicely on mid-range hardware until SP1, if not SP2. Look at Windows XP. It was released when the Pentium 4 had just been introduced. Athlon had just switched from advertising real clock speed (I had the Athlon 1.4 CPU, the last of that line) to advertising rough Intel equivalents. And XP ran rather… slowish… on that hardware. I saw a ton of people running XP on Duron or Celeron CPUs at like 700 mHz.

    The point I am making here…

    I think that if desktop anything-but-Windows had an opportunity, it was right around the beginning of the XP era. Let’s look at the conditions then, that lead me to that conclusion:

    • On the resource issue… even the lowest-end PC now has more that enough gumption to run Windows Vista fine, or OS X if that was an option. OK, they might not get the Aero interface. But Vista is the first Microsoft OS since DOS (without Windows on top of it) that ran fine on the run-of-the-mill hardware available when it was released. Slapping Linux on a dirt cheap PC and selling it for peanuts doesn’t make sense for all but the lowest end PCs, where the $50 or whatever Microsoft charges an OEM for Vista Home actually will make a big impact on the sticker price.

    • The home computer market was not what it is today, the boom was just starting, thanks to the sudden availability of broadband being priced the same (or less) as a second phone line + ISP. If the ABW movement was ready to strike at that moment, there is no good reason why all of those folks needed to be hooking up Windows PCs to their cable modems. Those could have been Macs or BSD or Linux installs.

    • XP was the time when home users (and the business users who stayed away from NT 4 and Windows 2000) had no choice but to get off of the DOS/Win9X architecture, and onto the NT codebase, with all of the attendent headaches of that transition. I would expect Microsoft to not do something like that (nor split the codebase so dramatically like Win 9X vs. NT) ever again. That was a window of opportunity that will not only never be open again, it’s been bricked over.

    • I think that the PC as a computing class is getting very, VERY close to past its heyday for non-business purposes. You’ll see me write a lot more about this in my space over the course of this year. Email is nearly useless at this point. Clock speeds and storage capacities are to the point where a PC is not needed for the multimedia. If the cell phone telcos get their acts together in terms of billing data in a way that doesn’t scare users from using it more, all it will take is for someone to get a YouTube client with a “TXT this video to your friends” feature, and you can kiss the PC goodbye in a lot of homes. TXT has already replaced email for a huge number of people, except for when they want to send attachments. And attachments can be solved by 3 items: 1) Storage space 2) Bandwidth 3) Viewers Once you see that storage and bandwidth keep going up, and that 90% of attachments are one of a few MIME tytpes (JPEG, GIF, MP3, AVI, MPEG), and the phone can already follow hyperlinks (even my dinky phone has a browser)… it is really hard to see why a lot of people would want, let alone need, an entire computer. A cell phone, not even a “smartphone” fits the needs of many people now!

    • The sudden proliferation of multimedia devices that started booming around the XP era, helped to drive PC adoption in the home. After all, those digital cameras and iPods needed something to connect to, to get the bits in and out. Those devices are increasingly getting network connections of their own, or are built into devices with connections. My cell phone takes pictures and plays MP3. MP3 players now take pictures and have WiFi. Etc. The PC no longer needs to be part of the equation. I know too many people who, 2, 3 years ago were chained to a laptop who now do everything through a BlackBerry. Pay attention to that trend, it will increase, not decrease.

    In a nutshell, not only do I beleive that the chance to dethrone Windows on the desktop market is past, I think that the desktop market itself is about to start to expire, if it has not done so already. Will be be a fast thing? No, more like a decade or two. But it will happen.

    For me, 2008 is all about setting the claymore mines and determining firing lanes to act as the rear guard in an organized retreat. I think that the change that was poised to happen a few years ago, the transition from data processing to computing, missed its chance. It was hijacked by the entertainment stuff, which can now be done perfectly fine (or good enough) on a “smartphone” today and a cell phone in a year or three. I have little desire to return to the data processing stuff that I’ve done ages past. And I don’t need to, either.

    As a result, I firmly beleive that I will essentially become a technological recluse. The programming paradigms that interest(ed) me are no longer relevant. The death of computer science as an academic subject was the canary in the mineshaft on that one.

    What is really, truly ironic is that the “Joe Six Pack” programmers are making the shift without even realizing it, because they don’t care enough about the work they are doing to even think about data processing vs. computing. And they don’t care, and they are happy. Meanwhile, I find myself preparing to spend the next year (or less) documenting the abortion of the movement that I dreamnt of being part of since my youth, as I slowly depart.

    Wow, that sounds gloomy. :)


    Comment by Justin James — 4 January 2008 @ 12:02

  2. A couple of thoughts came to mind as I read your post and Justin’s response.

    -what do you make of the fact that OS X runs on the iPhone? How does that fit with bloat?

    -do you or do you not think that very low cost machines will a) sell in large quantities and b) run Linux?

    Thanks for the thoughtful response. And yes, I am an optimist.


    Comment by Robin Harris — 4 January 2008 @ 09:06

  3. Justin:

    I don’t think desktop computing (as embodied not only in desktop computers, but also in laptop computers) in the generic is on its way out so quickly. When it does go out, I don’t think it’ll just die and leave us with cellphones that can take pictures and send text messages, either. I think it will just evolve into something different, and at some point we’ll look back and say “Huh. It’s not really ‘desktop computing’ any longer. I wonder when that happened.”

    What it will evolve into — well, we’ll just have to wait and see. I’m pretty sure it’ll involve getting ever-more mobile, though, on one end. On the other end, it’ll get less mobile, but more remotely accessible. In other words, just as with the business world before it, personal computing will diverge from single-system all-in-wonder do-everything machines into a client/server model, where you have your data store and so on at home, and a mobile device with you that will provide all the power of a good laptop (even when it no longer has a recognizable keyboard or mouse) that connects with the server. The only sketchy part of this is whether the server will be at home.

    Keep in mind that predictions like this depend on some assumptions about other things in life not changing drastically. After all, if US politics keeps going the way it has been, the US could potentially fall apart in civil war by the time these predictions might otherwise come to pass.

    Obviously, I don’t think the desktop computer market is going to die in two years based on changes like those you (Justin) posit any more than I think Linux will dominate the market in two years. In fact, I think that’s less likely.


    First of all, welcome to my personal weblog. I’m glad you noticed this and came to contribute to the discussion. I occasionally talk about Linux at the TR IT Security weblog, too, if you’re interested — though I don’t use that as a platform for picking apart others’ weblog posts except as pertains directly to matters of security.

    I think that MacOS X hasn’t gotten outside the realm of reasonability with the heaviness of its standard desktop OS, as long as what you want is a system that maximizes for glitz rather than for efficiency. This is especially true since, unlike MS Windows Vista, the vendor has substantial control over what hardware will be used to run it, thus ensuring that it works reasonably well everywhere.

    Of course, the iPhone’s version of MacOS X is a different animal altogether. It may technically be “MacOS X” that runs on it, but it’s not the same MacOS X you run on your desktop systems. Apple’s able to do that because it’s using a lean FreeBSD-like OS with a Mach-inspired kernel under all the major framework and GUI stuff used to turn it into desktop MacOS X — so if you completely change the lineup of software above the level of all that FreeBSD stuff, you can cut it down to embeddable size.

    Take note as well of the fact that the iPhone has some kickass hardware in it for a cellphone.

    As for Linux selling like gangbusters — I think it will do relatively well on the Eee PC, and I don’t see why it wouldn’t do very well. I’m just not sold on the idea that this particular battle will be “won” by the end of 2009. It may take longer — and companies like MS Windows and Apple, in combination with the stubborn, willful ignorance of millions of consumers, may yet come up with a way to yank that train off its tracks if it takes more than a couple years to reach the next station. I hope that’s not the case, but it very well could be.

    Never underestimate the power of willful ignorance.

    I’m a cynic. As I’ve defined it from time to time, a cynic is just an idealist that has learned from life experience.

    Comment by apotheon — 4 January 2008 @ 11:59

  4. To Chad: I think that MS, with it’s attempts at competing with Google on the Internet (and apparently losing), is over-extending itself. If it loses on the desktop in the next few years, it will be only because it is trying to give great focus to too many areas of computing. Microsoft, in order to survive, needs to re-invent itself and not in the buzzword sense, but in the real sense of determining what is truly important. Either it needs to focus on its core business of desktop/server computing and get mostly out of Internet services (such as their AOL-clone ISP offering) or dump their core services to the way side and focus on Internet services such as Google is doing (although I here talk of a 1st quarter Google Phone). It would help, too, if they paid more attention to the code-base as that would in turn actually make their customer-base happier.

    To Justin: Not sure that you’re right or wrong, and I really can’t say why. Just a gut feeling that your prediction(s) aren’t completely right or wrong.

    To Robin: You’ll have to ask Dell about the second point, as they already sell low cost machines with RedHat on them.

    Also, I think the techno-political climate is heading for 1984 instead of 1998, 2008 or even 2984, and if that doesn’t stop none of what’s said here will really matter.

    Comment by Joseph A Nagy Jr — 4 January 2008 @ 12:02

  5. Well . . . Sun Tzu did say it’s a bad idea to fight a war on two fronts. I don’t think that’s the major reason Microsoft will experience significant, growing trouble in the future — but it certainly doesn’t help matters for the corporate behemoth.

    As for the bit about 1998 — that wasn’t meant to suggest where the world is going, but to express my startlement at how fast the time has gone by. I still feel like I haven’t finished with the ’90s. It feels like the world has skipped a groove, and jumped ahead a decade or so.

    Anyway, barring the dirty tricks winning the day and maintaining the software industry status quo with brand spanking new, shiny shackles on the whole industry, I think the only way for Microsoft and Apple to survive in the long run will be to get increasingly open source oriented — and a lot faster than they have been already (though Apple’s much closer to being on-schedule with Darwin underlying MacOS X than Microsoft with F# and IronPython getting fused to .NET).

    Comment by apotheon — 4 January 2008 @ 12:17

  6. My point (not so well worded at that hour of the night) was that for people who are not mandated by an employer to use a desktop PC (which includes laptops), they won’t. 5 years ago, sure they would. The problem is that everything that the PC does on a day-to-day basis, unless you have a special purpose app like SAP or something else your business uses, is now available in other, more convenient form factors. Even more important is that so many of the things we have been using PCs for have been supplanted by other protocols. IM is being replaced by TXT. Email is being replaced by TXT. Email itself is so spammed out, it has lost all value to far too many users. Much of what people use Web sites for can be (and is, in a number of cases) replaced by RSS (for example, I do not go to your site except to respond, since the RSS feed has the full text). And so on.

    I just see a diminishing future for general purpose, desktop/laptop PCs, except in situations (like certain business apps) where they are needed. And those apps are movong to the Web, so really, all one needs is a docking station that provides a keyboard/mouse, monitor, power for their phone or smartphone to access them. It would either drive them all through the phone, or use the phone for storage and network access. Either way, it gets folks on the Web with local, transportable storage/settings, and immediately solves the disconnect/portability issues as well.

    The idea of a future like this really bothers me. But I think it is sadly where things have to head, and will head.

    As a result, no year will ever be the year of Linux on the desktop.


    Comment by Justin James — 4 January 2008 @ 11:23

  7. I’m sure MS and Apple will keep up with the pretty sparkly stuff but I don’t think Linux and the other OSes need to go overboard with all of the glitz to keep up. They just need to graduate out of the 20 year old GUI look that many of the programs have. A lot of the programs and utilites just look really old and it affects how people perceive their usefulness. People tend to think that if it looks like it was created in 1992 then it can’t possibily be as good as Windows.

    Contrary to what Justin believes, I think Linux does have a chance at getting a larger share of the desktop market. I don’t think the desktop is going away as fast as he predicts. The iPhone and other devices are still in the “entertainment” phase and I think many people still perceive them as toys that teenagers use.

    The popularity of the Linux machine sold by Wal-Mart shows that it can compete and be a viable option to more than just computer geeks. With computers already installed with Linux the users do not have to worry about getting hardware and various devices to work. They just turn it on and it is ready to go.

    Now that the age of installing software via a disk is slowly coming to a close, I think Linux has a chance to shine. One of the major problems in the past was that you couldn’t install software purchased at the store very easily on a Linux machine. Now many software programs are pre-installed or web-based. The software repositories in many Linux distributions are huge and as long as the OS makes it easy for people to just click a few buttons to install software they need, I think many people will find that much easier than dealing with the way you install new software on Windows.

    I honestly think that as long as the OS isn’t completely different than what people are used to, they may not even realize they are using something that isn’t Windows. Heck, I don’t think a lot of people even realize that the OS is something separate from the computer as a whole. If it does what they want then they are happy. If the same machine costs $100-$200 less and they don’t lose any features that makes them even happier.

    Comment by medullaoblongata — 5 January 2008 @ 12:35

  8. […] On Thursday, I posted an entry here at SOB about closed source OS vendors and their strategies for future battles to maintain market dominance against the encroaching threat of open source OSes. I titled it Closed-source glitz — missing the point. […]

    Pingback by Chad Perrin: SOB — 5 January 2008 @ 12:40

  9. Chad: Doh.

    Justin: I see where you’re going now and am more fully prepared to say that you are completely wrong.

    While cell/smart phones are definitely making the need for having a laptop/desktop (in that order) less apparent, the need is still there. As anyone who has used a standard cell phone for txting knows, it’s hell on the fingers to type out messages of any length, and with very little format control in text boxes (such as the one used in blogs such as this for typing responses), it becomes even more obvious why such devices will remain, for the most part, used for entertainment purposes. Sure there are the BlackBerry’s and the PDAs of the world that overcome these obstacles, but they are still prohibitively expensive for most users. Even the iPhone is ridiculously expensive ($400 for the 8GB model, slightly less for the 4GB model). As much as I’d love one, I cannot justify the expense right now (especially since I’m currently unemployed). Even for those who can afford it, unless there is great need for one of the many features, I don’t see adoption being as widespread as we see for something less capable such as the Motorola RAZR.

    Anyways, I need to get back to resume writing.

    Comment by Joseph A Nagy Jr — 5 January 2008 @ 12:59

  10. Justin

    As a result, no year will ever be the year of Linux on the desktop.

    Put exactly like that, I might agree. Somewhere in 2008-2009, though, we might see the beginnings of the year of Linux on the laptop. Since you previously specified that when you refer to desktop use you also meant laptops, however, I’m afraid I have to respectfully disagree.

    I believe that mobility is becoming the new necessity of computing, but not as drastically as you’d suggest. Something that came up in conversation with medullaoblongata is the cyclical nature of emerging consumer technologies, at least in the market for personal computing electronics. The observation she made is that there’s sort of a sine wave cycle (biorhythms?) where things oscillate between pseudo-extremes, and we may be seeing a growth spurt for ultra-mobile platforms like the iPhone, but it’s going to swing back when the hype dies down and that new car smell (aka, the novelty) has worn off for the early adopters, then we’ll be back to the standard desktop PC model and business as usual.

    My take was basically the same, with the proviso that the median state of things tends toward the advancement end of the spectrum very gradually. In other words, while the trend is oscillation between big general-purpose home computers and handheld rich computing appliances at the moment, over the long haul the median between these extremes will trend downward in size and complexity a bit.

    See, we went from minicomputers to microcomputers, then up to bigger and more beefy tower systems and “workstations”, then down to laptops as small as we could make them at the time, then shot back up to a market preference for beefy game-capable midtower systems, then to clunky handheld devices like the original Palms and their ilk. More recently, we’ve gone back up to desktop systems with some trending toward compact systems (think mini towers, iMacs, eMachines, Mac Minis, and micro ATX systems, pretty much in that order). From that, we’ve dropped into the iPhone range. The cycles are coming faster, so that the desktop-replacement huge widescreen laptops (from Apple and Dell, most notably) were pretty much concurrent with the iPhone in their growing popularity, and that’s already waning as people are still ga-ga over the iPhone — but the iPhone’s popularity is waning as well, having spiked quickly and now dropping off faster than gigantor laptops. Now, we’re beginning to see the rise of the ultra-portable laptops, most notably the Eee PC from ASUS.

    Ultimately, I think the median (which, despite the hype, is where most of the users are hanging out for most of their computing needs regardless of cyclical extremes) will only leave the full-featured portable general-purpose computer with built-in keyboard behind when a portable (essentially wearable) computer is developed that you can use with the same facility as a laptop, but even while doing dishes. That, I believe, is the future of general-purpose computing, rather than the apocalyptic replacement by smartphones you envision.

    Of course, part of what will fuel this trend toward smaller, simpler computers in the long haul (where “long” is increasingly an obviously subjective term, as the cycle accelerates over time) will be the increasing ability to modularize one’s computing environment, such that less of it needs to be dragged around with you. By the time we have wearable computers that you can use effectively while doing dishes, they will surely be somewhat like thin clients for dedicated server bricks — relatively portable, but still ultimately meant to be a stationary personal server device. The world is getting more networked, but that doesn’t mean we don’t need local rich computing experiences. It just means that what appears to be happening right in front of you will substantially be, under the hood as it were, happening somewhere else on your server(s), but transparently available via adaptive VPN capabilities.

    And those apps are movong to the Web, so really, all one needs is a docking station that provides a keyboard/mouse, monitor, power for their phone or smartphone to access them.

    1. A lot of stuff is moving to the Web, but a lot of it really can’t — at least, not if anyone is going to maintain any control over his or her own data. Can you imagine losing access to everything just because a backhoe took out the fiber cable running to your house, or because of sunspots? By the way, consider that (as medullaoblongata recently pointed out to me) the sun seems to be entering a long-term period of high levels of sunspot activity in the near future, cellphones (augmented as smartphones) as your connection to all things computer-oriented via the Internet may well become frustratingly unreliable.

    2. Docking stations will not “solve” the need for something with a complete, efficient physical interface. For one thing, the biggest problem with the iPhone’s interface is that it is essentially impossible to use at all without looking at the input interface — a huge detrimental characteristic compared with more traditional cellphones and physical keyboards. For another, limiting your ability to make full, efficient use of the system to places where you have a keyboard and monitor set up for it is extremely inconvenient. A laptop is basically the same thing, but with more computing power, and all integrated into a single unit so that you don’t have to leave your keyboard and monitor behind when you travel. Of course, you could take your keyboard and monitor with you, along with the smartphone, but that’s just ludicrous — it would be less portable than the laptop.

    As a result, no year will ever be the year of Linux on the desktop.

    Just for emphasis, one more time:

    We may not see the Year of Linux on the Desktop, but we will surely see the Year of Open Source OSes on the Laptop. (I still hold out hope that Linux will not mercilessly dominate that trend.)


    They just need to graduate out of the 20 year old GUI look that many of the programs have.

    A lot of that is dependent upon the GUI environment you’re using. When you turn on all the bells and whistles with a pretty theme in something like KDE or GNOME, much of that goes away. I guess a 20 year old GUI look is a hazard of using low to middling resource consumption and fast reaction time window managers like WindowMaker, IceWM, or Sawfish (though Sawfish is almost as slow, and almost as heavy in resource usage, as XFCE, which is solidly into middling at least). I, meanwhile, am using window managers that make IceWM, and even Fluxbox, look positively corpulent — smaller binaries than my terminal emulator (which is smaller than Xterm), with very minimal window decoration so that it doesn’t look old so much as nearly invisible.

    In fact, the most common response I heard from Linux users to the initial beta releases of Vista was how remarkably similar it was to KDE 3.5, which had already been out for almost a year. The sorts of applications that end up looking 20 years old no matter what GUI environment you use, meanwhile, aren’t generally used by the glitz-and-glamour crowd, because they pretty much stick to KDE and GNOME applications.

    I think the real problem with Linux and other open source Unix-like OSes is simply one of image, at this point. People still think of nothing but command line interfaces and crusty old Motif GUI toolkit apps in CDE when they hear words like “Unix” and “Linux”. They simply aren’t aware of the high-octane GUI glitz and comprehensive GUI coverage of common user and admin tasks that can be had on Unix and Linux systems these days. The Eee PC and its successors may well be what it takes to teach such people the errors of their ways — particularly when someone starts selling systems in such a high-profile manner with something like Compiz Fusion already enabled.

    Now many software programs are pre-installed or web-based. The software repositories in many Linux distributions are huge and as long as the OS makes it easy for people to just click a few buttons to install software they need, I think many people will find that much easier than dealing with the way you install new software on Windows.

    This will be a key factor in times to come. If Microsoft and Apple don’t get their heads out of their fourth points of contact with regards to convenient software installation and management (though Apple’s a little closer to the light, through little or no intent of Apple itself), they’re going to be in for a pretty tough row to hoe once the general public gets wind of things like APT.


    Even the iPhone is ridiculously expensive ($400 for the 8GB model, slightly less for the 4GB model).

    . . . especially considering you can get an ASUS Eee PC for under $300 with a powerful and secure (relative to MS Windows, anyway) OS and a full-strength office suite. Hell, it’s only about four times the size of an iPhone — and doesn’t suffer from the impressive flaws of the iPhone (see an upcoming article of mine about blunders in TR’s IT Security weblog).

    In General

    I tend to agree with the positions of medullaoblongata and Joseph, and even to some extent that of Robin Harris, over that of Justin — as should be fairly obvious by now. It’s possible my opinion might just slightly less strenuously disagree with Justin’s than either Joseph’s or medullaoblongata’s do, but I definitely agree more substantively with them than with Justin. Err, if that makes any sense.

    Comment by apotheon — 6 January 2008 @ 01:54

  11. Don’t forget to mention portage, which I still tout as superior over apt, especially in the area of dependency resolution (I have almost NO problems at all with portage in that area).

    Comment by Joseph A Nagy Jr — 7 January 2008 @ 10:15

  12. As I recall, the problem you had with APT was that you tried to install a newer version of KDE from outside the APT system — which mean that the cause of your problem wasn’t APT itself.

    Anyway, portage is just a cheap imitation of the FreeBSD ports system.

    Comment by apotheon — 7 January 2008 @ 10:28

  13. No, I was trying to use an apt source, it was just completely borked, and I had a great deal of dependency problems in stable while trying to use apt. That was my greatest Debian frustration. Even before I tried using something outside of the “official” apt sources list (still set up for apt, just not “official”), I was getting a lot of dependency problems. Anyway, that’s neither here nor there. Apt is a good tool, I just feel it lacks the refinements of portage, even if portage is just a rip of the FreeBSD ports system (the sincerest form of flattery is imitation, after all).

    Comment by Joseph A Nagy Jr — 8 January 2008 @ 01:46

  14. In my (admittedly limited) experience of portage, things broke rather more often than they did in APT. I guess we’ve just had differing experiences.

    Comment by apotheon — 8 January 2008 @ 10:10

  15. Not surprising there. (:

    Comment by Joseph A Nagy Jr — 9 January 2008 @ 12:42

  16. […] the IT Security weblog. In my SOB entry titled Closed-source glitz — missing the point, I posted a comment in the ensuing discussion in which I […]

    Pingback by Chad Perrin: SOB » In case you were waiting for the security blunders article: — 9 January 2008 @ 01:39

  17. As for Linux selling like gangbusters — I think it will do relatively well on the Eee PC, and I don’t see why it wouldn’t do very well. I’m just not sold on the idea that this particular battle will be “won” by the end of 2009. It may take longer — and companies like MS Windows and Apple, in combination with the stubborn, willful ignorance of millions of consumers, may yet come up with a way to yank that train off its tracks if it takes more than a couple years to reach the next station. I hope that’s not the case, but it very well could be. I agree 100%

    Comment by Jason — 4 February 2008 @ 07:13

  18. […] This isn’t the first time I’ve written about AKH here. I just thought you might like to […]

    Pingback by Chad Perrin: SOB » Sometimes Adrian Kingsley-Hughes pisses me off — 2 July 2008 @ 01:53

  19. […] If your primary criteria are particular to the gleam of fancy, flashy tricks and bells and whistles, OSes that use the X Window System are more than desktop ready, if MS Windows is considered desktop ready. The glitz and glamour of Compiz puts Aero Glass to shame. […]

    Pingback by Chad Perrin: SOB » How do you really know it's desktop ready? — 19 July 2009 @ 11:56

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

All original content Copyright Chad Perrin: Distributed under the terms of the Open Works License