1. Glorious design

2. Shitty first pass, released anyhow and made a standard

3. Laboriously fixing it over the period of years until it converges on "Slightly-but-only-slightly less sucky than what it replaced"

4. Goto 1

has become the standard operating procedure of the Linux desktop world.

I am not convinced that the Gnome and KDE projects are good organizational strategies. Replacing a lumbering behemoth accountable to its many customers with lumbering behemoths accountable to essentially nobody is not an improvement, no matter how "open" the latter may be.

(Oh, and I was happy only because I'm a professional programmer and I still prefer the command line over any "semantic desktop". I freely acknowledge this is not the common case.)

[1]: http://www.joelonsoftware.com/articles/fog0000000018.html

My favourite anecdote about linux's degradation is this: Earlier, applications like nautilus , cdburner etc would accurately predict and show the amount of time for operations like copy to complete. Moreover, copying stuff via the GUI was not significantly slower than command line.

With the latest versions of KDE (and even gnome), we've gone back to windows98 levels of inconsistencies (seeing the kde time-estimate bar during a file copy is just painful).

If you want a distro which doesnt take 1G of ram to run a filemanager, the latest one you can try is Ubuntu 7.04 - runs much faster on my 6 year old laptop (with 200MB ram) than my modern desktop (with 1G ram).

Patrick's emphasis on pragmatism & performance means that he simply wouldn't have incorporated stuff that didn't work if he could avoid it.

While the issues mentioned in the article are valid, it's still so much easier for me to maintain all family (and several friends) pc/laptops around, after I migrated them from Windows to Kubuntu/Xubuntu. The desktops just work and don't break, which is not something you can say about the Windows tamagotchi thing - you need to care about this one, or keep it restricted.

Also, 2-4 cores are standard these days. 8 - 24 cores would be high end.

Having an opinion about the miserable state of the Linux desktop has nothing to do with open-source advocacy. The success of many open-source projects including Linux itself (on everything but the desktop) is proof that the model works, and can produce quality software, often better than the proprietary alternatives.

I too am an open-source advocate, and I've met insult and disdain when I rant about the desktop "holy cow". There's just something there that makes people blind to the issues, and it's the same thing that makes the issues pile up and not get fixed. It just doesn't happen very often now because I seldom give enough of a crap about this long-lost battle to even enter the discussion.

And a big part of the difference is the environment:

On the server you have a relatively homogeneous hardware platform, many deeply technical users who can and will fix things that are broken, and good connectivity of related environments so that useful innovations spread quickly and the people who made those innovations possible benefit from their spread.

On the desktop you have a much less uniform hardware environment, relatively sparse numbers of technically able users, multiple obstacles to connectivity, and almost no individual recognition for people who make important improvements.

And the problems that are being solved are different, desktop software usually has to put up with a lot of bureaucracy and rules; writing desktop applications requires that developers be aware of the conventions of the system they are operating in and play well with a host of other software for something as simple as drawing a dialog box for the user and capturing input from it. On the server it's relatively easy to partition processes from each other so that they don't interfere with each other; whereas on the desktop you have to support every conceivable action that someone could possibly take.

That said, Linux "winning the desktop" isn't necessarily a worthwhile goal. Don't get me wrong; having Linux desktops available is a good thing all around, it keeps Apple and Microsoft honest. And it's unfortunate that intellectual property concerns have hobbled the development of multimedia in open source to the extent that they have.

Do we really need more WIMP style desktop operating systems that are attempting to be familiar to both Mac and Windows users?

Shouldn't Open Source Desktops be a hotbed of innovation? A carnival of new ideas, metaphors, experimental interaction designs?

But if you look historically at most major open source projects, they aren't innovative. IMHO, most open source projects seem to stem from a desire to have a free (in both senses of the word) version of xyz proprietary/closed source software.

So we end up with lots and lots and lots of copies/equivalents to closed source software, only with a few esoteric additions for whatever itch the author needed to scratch and half the feature set gone.

Sometimes, the OSS version gets better over time, and even supplants the closed source version. But just as often, once the author scratched that itch, development stalls and very little innovation happens. Or alternately, we'll see dozens of entirely different reinventions of the wheel, where somebody didn't like the particular way the OSS version worked, and then just rewrote the whole thing again. I mean really, how many half baked personal finance software tools do we need?

In all of this, what doesn't tend to happen is novel innovation in desktop software. Open source games are rehashes of closed source favorites (I mean really, can anybody point me to a completely novel OSS game)? Desktop software tends to be free (as in beer) copies of expensive proprietary software (GIS tools, image editors, 3d modelers, laboratory tools) or non-user friendly academic research projects released into the wild. Taking gaming as an example, almost all of the innovation is happening in the closed source, indie gaming scene.

Basically, you can kit yourself out with less polished version of hundreds of thousands of dollars of proprietary software for free (that you can play with the code a bit), but that's about it.

OSS is sadly trapped in a cycle of non-innovation and has been for years.

iTunes and WMP are bloated crap that force you to sit through DVD warnings. WMP has gotten a lot more stable since Windows 7, but that's actually a case where it has taken Microsoft years to come up with anything approaching the usability of VLC. (And their fealty to the MPAA prevents them from taking the final user-friendly step and allowing us to watch what we want to watch on the DVDs we've purchased.)

Chrome is widely regarded as the front-runner among browsers. And to be clear, Google didn't make Chrome because they wanted a free (in both senses of the word) browser. They built it because they wanted a browser that doesn't suck. (Though I guess you could make the argument that KHTML was just a me-too product.)

I don't think Inkscape has any competitors, open or closed. It's an intuitive, full-featured vector graphics editor. Adobe Illustrator by contrast is certainly superior, but it has a learning curve that makes it unusable for people who aren't graphics designers. Inkscape is extremely innovative.

Other open source products I can think of with fantastic UX are Transmission (the BitTorrent client) and Colloquy.

I'm not sure that is a victory of OSS as much as it is that they have more freedom to play fast and loose with patents than a commercial/closed source developer would have if directly selling their software (particularly in the US market).

Most software isn't innovative, and open source samples from the same intellectual gene pool.

EDIT: This is more like what I used to look for back in the day: http://www.osnews.com/topics/1 There's a list of available topics at http://www.osnews.com/topics/

Only if you don't want them to be used. Which is totally okay, but it'll keep most people (myself included) far, far away.

Last time the Linux desktop felt innovative was the release of enlightenment 0.16. I had 9x virtual desktops and a completely skinnable UI with a reasonably robust scripting language running on some POS Dell laptop, and it was speedy.

You are right, and my comment wasn't very clear on this part.

It was just to avoid the comments saying that I am an open-source hater.

The true beauty of Linux is the way that the OS improves over time. Problems are ironed out, and legacy hardware support is often ongoing. I think this is unique to Linux.

Decisions about where to go next re. Desktop paradigms are tricky ... but I think the main problem is lack of strong and decisive leadership; largely due to the problems inherent to an open development hierarchy. A paradigm shift often requires a leap of faith. Without that, any changes becomes a patchwork or amalgamation of multiple people's ideas - which in my mind isn't optimal.

I reckon Mark Shuttleworth is trying to create a paradigm shift with Ubuntu .. and I applaud him. There's so much life left in Linux as desktop platform. To state otherwise is shortsighted, and ultimately damaging to an incredibly vibrant and worthy platform.

I'm sorry, but the statement that linux improves over time, and thay's unique to linux has really no basis in reality. Both Windows and MacOS do improve over time, we could argue which of those three does the best job, but I don't really think linux would be the winner in that comparison either.

I don't have any experience with other OSes than those 3 mentioned, so somebody else can chip in.

I think an open source model encourages iterative improvement which isn't tied to commercial product releases.

This was the basis for my comment.

Your original comment was a little bit ambiguous: you were primarily thinking of legacy hardware support as in "legacy hardware support is often ongoing". I read the first part of the sentence "Problems are ironed out" as if that's only valid for linux, which I don't think it is.

Now try running Linux 3.0rc on, let's say, an aging SPARC 5 workstation. Just Works.

But hey, I use linux for the last two years for all my desktop computing (and more for servers). I love it and find it superior to windows. Except my bluetooth doesn't work: and it did with windows.

C'est la vie, I guess...

Believe me, they are not. They might be history for you, but not for everyone.

Many non-standard setups (using Jack, using pass-through, using bluetooth headsets) are still broken. Applications using Phonon have very different behaviors with or without PA. KMix and pavucontrol still behave very differently from each other...

There are also the hurdles to jump through of signing up for a bug tracker account, searching for existing similar bugs, watching my report get ignored because it's not an "average user" use case, etc. For someone like me who's experienced with ALSA, written a sound card driver, and had things "just work" for a very long time, there's greater incentive for me to just rip out PulseAudio and get back to work.

Gnome3 and Unity, I leave them the benefit of the doubt, because they are new, but GNOME people are usually the ones who are pushing those technologies.

And I tried out 'airfoil' the other day (I think that was the name) that let me push out audio from different apps on my mac to 'foil speakers'? - basically other software running to capture and playback that audio. I can essentially run and manage just the audio portion of itunes on mac 1 from software on mac 2. And it 'just worked' (to reclaim the ubuntu mantra). Ahh - here: http://www.rogueamoeba.com/airfoil/mac/

I hate how it keeps touching the volume level for the system/hardware.

Power management with linux on laptops has been going downhill since sometime and today I get atleast 30% less battery life as compared to Win7 (maybe Phoronix is right with its power regression hullabaloo)

Suspend and resume still sucks (https://bugs.freedesktop.org//show_bug.cgi?id=28739 presumably fixed in 2.6.38 but still gives me problems frequently). Displayport/HDMI out on my laptop does not send out audio and until I used a custom PPA kernel, multi-monitor out on my laptop used to lockup frequently.

All of these things work seamlessly on Windows 7.

I am willing to pay good money for a seamless Linux desktop (~100$ for something that works seamlessly and timely updates) - if that's what it takes to get someone to focus on something of value, like TuxOnIce, rather than placement of close buttons on the left vs right.

oh, oh.. jumps around excitedly. I know this one!

  sudo apt-get install powertop
  sudo powertop

  pkill npviewer.bin

P.S. I dont completely trust Phoronix, but I'm beginning to think that the power regression thing is real http://www.phoronix.com/scan.php?page=article&item=linux...

Apple does it and that's why their hardware mostly works. (But to be fair, I've had plenty of issues with Windows and OS X. Software sucks. Hardware sucks. That's why you get a supercomputer with 15 hours battery life for $500, though. If you wanted it to be perfect, it would cost you $100,000.)

I imagine it's cheaper just to support the hardware you are using for other OSs though?

Dell were shipping Linux based laptops and presumably chose this route over the "create hardware specifically for Linux" route. Why did Dell fail in this respect?

Say you only support hardware from the last 5 years (Drop all pre-2001 compatibility.) - 5 major laptop makers,3-4 hardware/motherboard lines. Can a software player not raise 20K$ to buy hardware and build in full out-of-the-box compatibility ?

Diaspora raised 200K for vaporware - can this not be funded ?

FWIW, this is somewhat incorrect.

First of, Konqueror is an application, not a browser engine; you're thinking of KHTML.

Now, it's true that KDE is still developing KHTML and KJS (and it has to, considering KDE commits to ABI/API stability within a major release series and the Qt port of WebKit wasn't close to ready by the time KDE 4.0 was released).

However, Konqueror offers support for using either KHTML/KJS or QtWebkit/JavaScriptCore - technically, both are just KPart components, along with the Okular PDF viewer and various others Konqueror embeds to display documents of various sorts - and some other parts of the "core desktop" use WebKit exclusively (e.g. the Webview and Web Slice widgets).

(And yes, on my video-playing box, I use PulseAudio, and listen to stuff with Bluetooth headphones. Turn them on, audio switches to Bluetooth without the mplayer session even being affected. Turn them off, back to the TV's speakers. It just works. And I set this up by clicking pretty things in a GUI.)

http://www.thechromesource.com/netflix-plug-in-for-chrome-an...

As it is, the Linux desktop is still working towards "yesterday's experience," since we don't have good enough browsers and there are tons and tons of legacy apps. Unfortunately, supporting yesterday's experience is harder for usability since it's conceptually richer, and most of these concepts are dueling at the bits-and-bytes level, where they're hard to reason about and present in digestible ways. The browser can blackbox a lot of stuff behind the standard APIs, something which the OS does not enjoy.

Another problem, just as large, is that there continues to be lots of new hardware, and correspondingly new drivers to support and specs to reverse engineer(since many manufacturers remain unwilling to assist). This is a problem that can only be solved by maintaining a hardware taskforce roughly proportional to the amount of new hardware released, but it's not particularly interesting or sexy, and it's not a core goal of most of the Linux ecosystem to create premium-grade drivers for all hardware.

Now, I'll admit, I'm a minimalist. I use Arch Linux, run no end-user applications except Emacs and Chrome (and I use w3m as the browser often enough), dwm for window manager, and so on. But really it seems that the enormous complexity of these new solutions makes them near-impossible to support. PulseAudio, PolicyKit --- all of these had working systems that were far, far simpler. Maybe instead of PulseAudio, we needed to look at the ALSA code, clean it up, and add the few features people really cared about (per-application volume control) there. Yes, the semantic desktop might be this great pie-in-the-sky result. But really, too complex. Unity, Gnome3? Same thing. We don't need Wayland, we need to simplify and clean up X.Org (something, admittedly, that /has/ been happening. X.Org is the one of those software packages that /does/ improve monotonically with every release). We don't need the total solution to all our problems. We need to simplify and polish what we have now.

And, to be honest, GNOME 2 is still pretty lacking. I know a lot of GNOME team members past and present, and I like and respect most of them for a lot of work put in on a pretty thankless project, but the results aren't there compared to Windows or OS X.