I don't believe that any particular desktop will be the 'One OS to Rule Them All.' The market seems to be moving in a direction where the operative terms are 'heterogeneous' and 'browser-centric.' The desktop environment, more and more, is simply a means to get to a standards-based browser, so you can do those productivity tasks you wish to do.

Linux has a lot to offer (stability, ease-of-use, security come to mind) if it could only be allowed to compete on a level playing field. Get the Top-tier vendors and retailers to start delivering linux pre-loaded on systems, and let the market decide. Most folks don't load OS's anyways; they're most likely to use whatever the computer came with.

If linux continues to make incremental progress in desktop market share, it will compel proprietary software makers to really compete: on price, on features, on ease-of-use, on security and actually take into account user requirements. Just look at feature improvements in Internet Explorer that weren't there before Mozilla-Firefox started to gobble up market share. Let's see more, fair competition. It can only benefit us as end-users.

The list was meant as a thought piece to drive discussion. I specifically said that the items were only possibilities and that the audience (those at LinuxCon) could and would, of course, add their own. This was one portion of a much larger talk and was not meant to be comprehensive.

Bob Sutor

"And just like Google and Amazon and other huge consumers of Linux, they'll have a built-in GPL dodge and share only whatever code they feel like sharing"

I have no shares in Google, but Chromium the OS (including the publicity and credibility) and SoC are big givebacks aren't they?

What about that filesystem they are developing?

Do we actually know that they run some souped up kernel doing weird and wonderful things?

Just asking

The place to start here is to try to get a balanced view of what Dr. Sutor actual said _and_meant_ during his presentation. I have no idea if the person using the name "Bob Sutor" in the blog comments section is acutally Dr. Sutor of IBM. But a balanced report of what he said is given by Brian Proffitt on the www.linux.com website (http://www.linux.com/news/enterprise/biz-enterprise/73050-linuxcon-desktop-should-be-better-than-windows).
.
That being said, I don't believe that IBM or Red Hat (or some of the other large, corporations) are ever going to be the ones to push Desktop Linux forward. Mostly, these are companies who's mandate is to make money for shareholders. For these companies, this is the most direct route to $$$: Linux servers/services -> Profit! Sorry, sudo-religious (couldn't resist the pun, for you Debian fans) arguments or beliefs of what Desktop Linux is or should be are not important, here--simply put it in a speadsheet and see if the shareholders are making their quarterly gains. But to suggest that the money they have put into Linux (albeit for their own, self-centered reasons) hasn't benefited Desktop Linux is hard to justify. Just supporting Linus et al for kernel development, to name one thing, is extremely helpful to everyone using Linux.
.
So, that leaves Desktop Linux with benefactors--folks that are willing to spend sweat-equity or their own money into pushing things forward. Various community distros, Mepis and Ubuntu fall into this category. For example, Warren and his faithful community of "elves" do an excellent job with Mepis, while Shuttleworth continues to throw money at Ubuntu (although he states that even he has his limits for no profits). Individual Desktop Linux advocates also fit into this category--who can find a more dedicated and effective Desktop Linux believer than helios? Ken is out there, making a difference every day.
.
This is the landscape I see, and my prediction is that this is how it will remain for the foreseeable future. I'm not forecasting some big, watershed moment that all those Windows users will suddenly see the light. Instead, it is going to be an incremental, a-user-at-a-time type of an approach. Thus, I don't see a "Year of the Linux Desktop" ever happening. But, maybe that isn't so important. ARM-based netbooks are the next possibility for a big jump forward in Desktop Linux usage, but even that area will not be a "gimme" for the penguin.
.
-dB

quote: Bob Sutor said:

The list was meant as a thought piece to drive discussion. I specifically said that the items were only possibilities and that the audience (those at LinuxCon) could and would, of course, add their own. This was one portion of a much larger talk and was not meant to be comprehensive.

Yeah right Bob, your (IBM's) past record of support for Linux on computers speaks for itself.

Your (IBM and the rest of the "Community" at Linux Foundation) speaks volumes too, a bunch of Corporate suites who's latest passion is Cloud computing, where YOU get to control my data. Oh sure once that's in place you may find Linux relevent, so long as it's tethered to your Cloud.

Yay Carla! this is the first article on the subject I have read that did not read as all claws and teeth :), I like the part about not looking for a saviour.

Stability? In my experience the Linux Desktop stability is worse than that of e.g. windows XP.

It may be that a failing Linux desktop will not bring down the entire box but that helps me little when I'm writing a document in OpenOffice and X hangs. My guess the problem often have to do with bad graphics drivers and bad support from graphics hardware venders, but again it doesn't matter why something fails if you have work to do.

On the server side, on the other hand, Linux just rocks

@anonymous...

You will hold Linux accountable for that while not holding the others that way? Nice.

If that's not what you meant to say, then you should have qualified your remarks...

I don't have a good feel that you even KNOW what might be causing your claimed problem. More often than not, it's not dodgy drivers or a dodgy OS, but rather bad parts- even a power supply not being quite right can cause issues like the one you've described. I've seen that and it's not Linux, but the parts that were the issue. And I'd have had even stranger issues with the Windows side of things with the hardware glitches I've seen in the past.

Even if it was a bad driver, here's a hint: Would you put up with a "bad driver" under Windows? No? Why, you'd work towards getting different hardware once you fingered the driver in most cases, right? WHY would you do any different in Linux? If the vendor's the provider (i.e. AMD or NVidia...) then you need to drub them instead of blaming Linux. If you can't replace the GPU, then you might want to reconsider IGP based machines as being unacceptable for future configurations as you can't rely on the vendor's support being robust.

I'm not against proprietary companies making money with Linux & FOSS, but I am against them trying to LOCK-IN Linux or FOSS into their revenue model the way Microsoft is trying to use MONO to capture the Linux desktop API.

The "Year of the Linux Desktop" for me was 1998. Linux has been gaining on the market share leader every since, and in the last two years Linux has captured more than 10% of the desktop market share away from Microsoft, according to Mr. Ballmer. In some countries that market share is over 15% or more. Everyone that I've asked, who is running XP or VISTA, said they had NO plans to upgrade to Win7. And why should they pay SEVERAL HUNDRED dollars more to get what XP gives them already?

As far as the "Cloud" is concerned, anyone who owns a home PC isn't going to start shelling out money to run the same apps "remotely", and expose their personal information and data even more that Windows already exposes it. Security and affordability is driving MILLIONS to the Linux desktop. They are not going to throw that away by moving to the Cloud.

It's all about the money.
If you're a multinational or government with 50 000 desktops or so then IBM or Red Hat or whoever will sign a contract for desktop support but for you and me or the corner store or the local doctor or lawyer etc; no way.
Desktop support is a done locally by little shops and they do MS stuff because they get a regular income from a small enough area.
Linux is too reliable for them to get a good enough income from a managable area; ergo they aren't going to change to Linux unless the customer base forces them to; and if it does a goodly number of them are going to go out of business; not enough work because not enough security holes to patch up or fix up after the event.

Chris B

@Chris B

Not necessarily, issues can do occur even if its not the OS's fault after all. There will those who still are just users and wouldn't know what to do if they accidentally mucked something up on their Linux desktop. Most likely people who go to shops to fix their PC's will probably still do so with Linux. In fact I would suspect they might get more visits if Linux was more wide spread. Linux is great but its not perfect. There will be issues (bulk of which might be user-created granted but still, never underestimated mankind's ability to be stupid.)

GreyGeek: you keep claiming that Linux has more than 10% share according to Ballmer. THAT IS FALSE. Please stop repeating it.

Apparently you got that from that PowerPoint slide (I can't retrieve the URL right now, but I'm sure you know which) showing Linux with a larger slice of the pie than Apple. Add that to other sources claiming that Apple has 10% share and you get your "more than 10%".

But that's wrong because in Ballmer's slide, Apple DOES NOT HAVE a 10% slice! If they were 10% and more than 10%, together they'd make for almost a quarter of the pie (25%). They're clearly much smaller than that. My guesstimate is that Ballmer set Apple at 4% and Linux at 5% (if you want to check that, print the slide, measure the angles with a goniometer and divide by 3.6 to get percentages).

Actually, I've read claims that 4% is a more realistic estimate of Apple's share worldwide, while the 10% figure might be correct in the US only.

@ Anonymous

I agree with Frank that you're holding Linux and Free Software to a standard that you're not holding proprietary software to. You may be forgiven slightly as one of the selling points of Free Software in general and Linux in particular has been its stability. However, methinks you need to do some troubleshooting.

I have a box sitting here that shows an uptime of 220 days, 8+ hours and all that time I have been logged into KDE 3.5.10. Of the open apps Iceweasel has to be restarted on occasion as does Fldigi if my radio control goes wonky but that is rare. So far I have found KDE 4.3.1 to be stable and usable on my laptop but I don't have the uptime figures to tout as I jump from Debian Sid to Kubuntu Karmic to see what is new.

So I would suggest you have a bit of homework to do to figure out the cause(s) of your crashes as they're unusual to say the least. OTOH, you could be a troll, but I feel like giving you the benefit of the doubt this morning.

Chris B, why do you assume that the main way to make money on Linux is by putting out fires?

@Bob Sutor: Did I make any factual errors? I'll be glad to correct any. I think my analysis is right on, and that IBM is one of the last companies in the world I'll listen to about the the Linux desktop. At the risk of being tediously repetitious:

1. Bigtime commercial vendors, even those who profess great love for the penguin, are not interested in selling a Linux SOHO desktop, or even offering a minimal level of support for their existing Linux customers. None of them are interested in selling a nice polished Linux LAN server either.

2. There is a marked trend away from customer premise servers and desktop, and towards hosted services, which takes control away from the customer.

3. All those nice Linux fans who have been wanting corporate involvement all these years are finally seeing their wishes come true. And that is fine, just don't expect those titans of industry to have your interests at heart. You still have to do your own heavy lifting.

When it comes to OEM Linux, stick with the independent Linux shops like ZaReason, System76, Penguin Computing, and so on. They're clueful and responsive to Linux users, and don't have to do any weird keep-Redmond-happy dances.

I've been using GNU/Linux exclusively since 2001, and curiously enough, during the last few years, I decided to settle for a distribution that's almost explicitely not meant for the desktop: CentOS, a clone of Red Hat Enterprise Linux. Red Hat has repeatedly formulated its lack of interest in desktop products.

And yet, to get an idea what you can do after a little tinkering and sanding down the edges:

http://www.microlinux.fr/captures.html

Sometimes I can't help feeling that Red Hat behaves a bit like an elderly soft-spoken homosexual claiming overtly his lack of attraction for women, yet proves to be a much more satisfying partner in daily life than all those self-proclaimed latin lovers who feel like they're God's own gift to humanity in general and women in particular.

Stability? In my experience the Linux Desktop stability is worse than that of e.g. windows XP.



Please, I have been using GNU/Linux on the desktop and notebook since January of 1999. Yes, in the workplace some employers require me to use Microsoft Windows usually installed on their computers. In the rare times I need to run Microsoft Windows (XP) on my own computers it runs within VirtualBox. The difference in stability of both operating systems is sufficiently close to zero but GNU/Linux comes out ahead by a thread's thickness. At least with MSWinXP the Redmond Giant finally got it right.



It may be that a failing Linux desktop will not bring down the entire box but that helps me little when I'm writing a document in OpenOffice and X hangs. My guess the problem often have to do with bad graphics drivers and bad support from graphics hardware venders, but again it doesn't matter why something fails if you have work to do.



Yes, GNU/Linux occasionally hangs itself due to some driver issue or a memory release problem with an application. Mozilla Firefox is a chief culprit here. However, I can generally kill the offending application from the command-line or in SUSE Linux a couple clicks of the application 'close' button tends to bring up the Terminate Application dialogue.


Bad drivers are usually the fault of the vendor because they either release buggy code or refuse to release the specifications for the device so the community can build a better device drive.


Is GNU/Linux perfect? No. Is Microsoft Windows XP much worse than GNU/Linux? No. is GNU/Linux ready for the desktop for most people? Yes.

"The Linux desktop has been outperforming Windows for a long time, which is a low standard to beat. Apple has the reputation for prettiness, but Linux has long beat the pants off both of them in flexibility, features, customizibility, power, and stability. It is doing quite well for prettiness too."

The author and the Linux crowd are so out of touch with the real world, there's no wonder they don't understand why the vendors do what they do. In what universe is Linux superior to Windows and Mac? People continue to have problems with simple things like sound and video resolution to this day in Linux. If they do manage to get them working, an upgrade will quickly change that. And there's no dearth of upgrades. There's one every six months. Look at the Ubuntu forums to truly understand how ridiculous Linux is.

IBM has actually tried to convert to Linux desktop in its own company, and failed. Perhaps the IBM folks know a little more about the Linux desktop than the bums cheering for Linux from the sidelines. If there's money to be made in the Linux desktop, trust me, somebody is going to be making that money. There should be no need for any kind of 'promotion' or charity.

The author says applications are all that Linux needs. Even if that is somehow true, how come we have so few commercial apps for Linux after all these years? Linux is in development for 18 years. We have been hearing for at least the last ten years about how great the Linux desktop is. But where are the apps? There are any number of reasons for this, but the Linux crowd is utterly incapable of figuring them out. They are too deluded to wonder why they should expect the situation to change now if it didn't change in lo these many years.