OS != Distro

There is nothing better than seeing people “getting it” as they go along. Mark Shuttleworth of Ubuntu is one of these people. In his latest blog post he discusses that projects must co-operate and interoperate with each other, and one first step to do that is to have common release schedules and bugzillas. He is so right about this. The No1 thing that makes GNU/Linux “clunky” and “difficult” is the fact that each project does its own thing, without architecting their software in a way that plays well with a grander scheme — because there is no grand scheme as a whole. This is the reason why Gnome/KDE feel so “cut off” from the underpinnings of the OS they run on. But Mark gets it, and he gets it because he is in the business of the OS desktop. I am sure that Red Hat and Novell people also got that in the past, but their primary business were always the server market, and therefore never cared to comment or pursue a grand plan of transforming a “distro” (as in collection of random software) to an “operating system” (as in one logical entity that was engineered to interoperate perfectly). However, I am sure that when Mark and others in Ubuntu came together to create Ubuntu Linux they didn’t have such ideas but they realized the needs of a modern desktop engineering as they went along.

Testament to this is my personal experience with Bernd Korz, the CEO of YellowTAB. When Bernd first came over for dinner a few years ago, he had just started in the desktop OS market with his Zeta OS. He was over-optimistic, an OS enthusiast. He did not understand basic things about what needs to be done in order to create a desktop OS that “just works” in order to sell. Fast forward two years later. Bernd comes over for dinner again during one of his US visits. This time, I see a changed man. This time, I see a professional who understands the traps, problems and overall machinations that he needs to overcome in order to get such a product out and succeed.

Also, a few years ago I asked my JBQ (who is an ex-Be,Inc. media/kernel engineer) “what made BeOS so good and consistent? It felt that it was a living, breathing life form rather than an OS“. JBQ replied “it was the fact that when I had a problem with my media code or performance, I could just walk down the hall, find the kernel guys Cyril or Travis and talk the problem out in minutes.“. BeOS 5 back in 2001 was delivering the same smooth experience as Mac OS X offers today (as long as you had a compatible PC that is).

Mark started Ubuntu with somewhat idealistic and naive ideas, but as he went along and the users that were coming from Windows demanded this or that, it becomes clear to him now what needs to be done. And what needs to be done is full interoperation, co-ordination and most importantly, a grand plan of architecting the various libraries, kernel, X11, Gnome and the whole shebang. In other words, a modern OS must be developed in a way that it’s well thought as an overall plan, instead of putting individually-developed libraries together afterwards. That’s what’s mainly creates the underwhelming performance and experience of the desktop Linux, and Mark seems to get it. A semi-fork might be in order too (read the last paragraph here to see what I mean by “fork”).

As a sidenote, I had more than three times in the past the experience Mark describes in his blog. I found a bug, and the bug was shared among 3 projects (the latest bug has to do with libtiff, shared-mime-info and Evince). I had to make 3 bugzilla accounts in 3 different places and the bug is still not completely fixed (only Evince fixed it as of yet).

Post a comment »

Jay wrote on December 28th, 2006 at 1:50 AM PST:

The thing is that it is “cut off” from the underlying operating system for portability. Things like DBUS, HAL, and things like NetworkManager and gnome-power-manager that are built on top of those, seem to be the step in the right direction.


Mark wrote on December 28th, 2006 at 7:38 AM PST:

“I am sure that Red Hat and Novell people also got that in the past, but their primary business were always the server market, and therefore never cared to comment or pursue a grand plan of transforming a “distro” (as in collection of random software) to an “operating system” ”

Which explains why Red Hat is investing much more on the desktop including primary developers of HAL, DBus, Network Manager etc which ties freedesktop.org and major portions of cross desktop functionality together while Mark shuttleworth has a explicit goal of using the desktop to get into the door and eventually turn it profitable from server software.

Sometimes I do wonder if people ever bother to look at reality rather than just someone’s propoganda


This is the admin speaking...
Eugenia wrote on December 28th, 2006 at 8:35 AM PST:

> Which explains why Red Hat is investing much more on the desktop including primary developers of HAL, DBus, Network Manager etc

Red Hat’s policy and resources was shifted AWAY from the desktop 2 years ago. Red Hat only has engineers working on desktop-related things ONLY if these things truly benefit their server solution too.

>Sometimes I do wonder if people ever bother to look at reality rather than just someone’s propoganda

It is you who don’t see the reality. Even Havoc Pennington, the manager of RHAT’s desktop team and the driverforce behind everything desktop on RHAT, does not work FOR the desktop of Linux anymore. He took away 5-6 of the desktop engineers (including the GTK maintainer) in order to create an user application instead of system software. RHAT does not do, not even 30% of what they used to do for Gnome/desktop before 2005.


Mark wrote on December 28th, 2006 at 9:57 AM PST:

“Red Hat’s policy and resources was shifted AWAY from the desktop 2 years ago. Red Hat only has engineers working on desktop-related things ONLY if these things truly benefit their server solution too.”

Explain how HAL or DBus or Network Manager or dozens of desktop components listed in here benefit the server?

“It is you who don’t see the reality. Even Havoc Pennington, the manager of RHAT’s desktop team and the driverforce behind everything desktop on RHAT, does not work FOR the desktop of Linux anymore.”

HP has not working on the desktop for long time before mugshot started. Connecting them both is distortion of reality to fit your agenda.

“He took away 5-6 of the desktop engineers (including the GTK maintainer) in order to create an user application instead of system software. RHAT does not do, not even 30% of what they used to do for Gnome/desktop before 2005.”

GTK has multiple maintainers. I assume you are referring to Owen Taylor. Have you looked at Mathias Clasen? There isnt 5 desktop engineers working on mugshot at all on a dedicated basis. If you want to claim otherwise, can you name those 5 people? Where does the 30% come from? Can you quote sources? Do you know the size of the current Red Hat desktop team or what projects they are working on? Who do you think is developing the above mentioned projects? Red Hat certainly has more people working on desktop than the entire Canonical team.

Mark Shuttleworth very explicitly stated in a interview that he is using the desktop as a gateway for his planned server business and that the desktop makes no revenue in Linux. Where is the Canonical desktop team working on? Have they started a single new desktop project? They have again a stated policy on not working on new projects because it doesnt yield revenue for Canonical.


This is the admin speaking...
Eugenia wrote on December 28th, 2006 at 10:08 AM PST:

>to fit your agenda.

What agenda?

Mark, this is the second time you reply to my blog and you accuse me of things that are not true. You come here with PREJUDICE against me. If you have a problem with me please include a REAL email address so we can settle this directly.

And please use HTML to represent links, as it’s stated above the form.


Martin wrote on December 28th, 2006 at 10:24 AM PST:

“He took away 5-6 of the desktop engineers (including the GTK maintainer) in order to create an user application instead of system software. RHAT does not do, not even 30% of what they used to do for Gnome/desktop before 2005.”

Like Mark asked you before can you give us some proof/source for your claims. I am also very interested in those figures and where they come from.


This is the admin speaking...
Eugenia wrote on December 28th, 2006 at 10:35 AM PST:

I can see how fast/slow Gnome evolves. Red Hat does not actively pursue Gnome development anymore. Most of the RH-based projects are sterile for months now. For example, even NetworkManager is slow to get updates, shared-mime-info is pretty much abandoned, GTK goes a lot slower than it used to be etc etc. DBUS and HAL and Fedora itself continue in a faster pace because thankfully, third party people have taken control of them. But other projects were not as lucky. I don’t need to have an insider’s leaked paper to see that resources were PULLED away from the desktop at Red Hat (and recently at Novell too). I can just see how things are going. Remember, I am also running Gnomefiles.org, it’s my job to see how that “market” goes (and I am not talking about third party apps only here). Also remember, this is my personal blog and I write my opinions the way my experiences with several things have shaped it. I don’t need any “hard proof” before I publish something, because this is not NYTimes or WashingtonPost, not even OSNews. It’s a personal blog and therefore it sports opinions based on observance.

BTW, the actual topic here is cooperation of various projects under the same umbrella, please try to stay on topic.


Luis wrote on December 28th, 2006 at 11:00 AM PST:

Being realistic about the Linux desktop, I think that in it’s current state it could be defined as a “0.8.x” version, i.e., it’s very usable for people who like computers but it still needs some polish for normal users. My guess is that it will enter 0.9 by mid/late 2007 and 1.0 by the end of 2008 or beginning of 2009. Then it will be ready for prime time (yes, I think that 2009 could be the year of desktop Linux at last).

All the hype from 2001-2004 you talk about was just that, hype. It normal, since at that time it was progressing very fast. But that’s just because it evolved from a 0.3 to a 0.7 state.

However, even in 2009, when Desktop Linux is completely ready, it will have to face it’s biggest challenge: Windows’ 96% market share. It is very difficult to compete against an OS that has such market share even if your product is superior and free (gratis), since you not only have to compete with the richest company in the world, but also with all the hardware and software vendors on their side – well, in (some kind of) hardware it’s not that bad thanks to the server market share of Linux.

I agree with you that most of the work required from now to a 1.0 release is about getting all the pieces together, increasing compatibility, cooperation, etc… but I think that things are going in that direction already, so no need for big changes or a semi-fork to achieve it. It’s just a matter of time (2 years from now is my guess).


Martin wrote on December 28th, 2006 at 11:02 AM PST:

Writing your own opinion is one thing but using figures which are “sucked out of your thumb” is another.
Stop using figures which you cannot support.

Next time you will tell us that an increase of 80% productivity into Gnome development has emerged from your article.


This is the admin speaking...
Eugenia wrote on December 28th, 2006 at 11:10 AM PST:

>Stop using figures which you cannot support.

The figures are estimations and there to make a case, not to be substantiated in any way as 100% accurate. Nobody can give accurate numbers about anything like this! My blog –as I have written many times– is a way to write down my thoughts, the way are inside my head and the way I see things around me. You think too much of me as “Eugenia of osnews.com, a tech news site”, rather than “Eugenia, the geek”.


Max wrote on December 28th, 2006 at 11:30 AM PST:

Luis is completely right. FreeBSD nor major Linux distros are not there yet. They all are not ready for desktop

market. That’s it. You can use those systems for non-critical server environments like web-server, clusters, etc.

but there needs more to be done to make them ready for desktop.

Usability, consistents and compatibility … is needed.
Unix systems are known to lack all of the 3 factors badly (except MacOS X of course). Even MacOS X is

inconsistent and not a good point for compatibility, but has a good usability.

There is BeOS and Windows (especially the WinNT platform) which are known for be the best in this categories.

BeOS is obsolet and no one need to shed crocodile tears about a product which has been dead for more than 5

years, beside some rather unsuccessful methodes to reactivate it.

So there is only WinNT series left. It has the advantages of Win16/32(/64) subsystems of earlier Windows versions

and the powerful kernel architecture based on VMS with some ideas from OS/2. WinNT has the best compatibility,

which reach 20+ years back for applications and 13+ years for drivers. Hell, you can execute a Win 1.0 apps in

WinVista without problems and you can even use several WinNT 3.1 driver. The usability and consistents are

similar good as MacOSX and BeOS (as speaking of WinXP).

The major drawback is how Microsoft act and what it’s roadmap looks like. TCPA here, DRM there, limited licenses

overthere, etc.

That’s why I have great hopes for ReactOS. It’s the only operating system which is heavily based on VMS

architecture like WinNT series but also incooperate the former secrete native WinNT API. Plus it provide using

WineLibs and code written from scratch a Win32 subsystem.
This means, ReactOS is growing to an alternative operating system similar to WinNT (XP/Vista) series and is even

highly compatible to it. The biggest advantages are that it is a free software project and even open source.

Means no licenses, no DRM by design inbuilt, compatible to the biggest software base.
ReactOS is a full operating system and there is no need for distros which only lead to more incompatibility, a

famous example for this is our well known overhyped unix clone named after his creator Linus.

Get it?

Why should I bother about the overhyped “linux” kernel or any other unix based software which has a long way to

go and is in the end still incompatible to most software?
I for myself, will stay as most user with the Win32/Win64 software for the next decades. And ReactOS will help me

in 1-2 years to stay independent Microsoft.


Martin wrote on December 28th, 2006 at 11:31 AM PST:

“You think too much of me as “Eugenia of osnews.com, a tech news site”, rather than “Eugenia, the geek”.

Then you must insist to remove the direct link to this story on the OSnews homepage. You write and link articles yourself on Osnews so its not strange people mix-up your tech and fantasy figures.

On Topic:
I think you mix-up two things;

a programmer/team who has to program an application wich has been designed by some think-tank team of a big software house and a or A programmer/team who has lots of fun in programming an application of which he publishes the source on the net.

You try to fit an open-source way to fit in a commercial jacket. Working together does also mean forking, new ideas, new chances. I know form my own experience that lots of new and good ideas do often not make it into the software because of commercial thinking. You need both to get the best. There is no “getting it”


This is the admin speaking...
Eugenia wrote on December 28th, 2006 at 11:40 AM PST:

>Then you must insist to remove the direct link to this story on the OSnews homepage.

I won’t insist anything. This was Thom’s report and he took the decision as an active editor in chief to link to my blog because he judged that it was an on-topic reply to Mark’s blog (I was sleeping when Thom did so, I found out hours later, after I woke up). Besides, this is not the first time we link blogs! We do that ALL the time, on all kinds of blogs! It is up to the reader to figure out that some blogs are more ‘official’ in their content (e.g. Havoc’s or Mark’s) while other are more personal. Mine, is a personal blog. I mean, come on, it’s called “Eugenia’s RANTS and THOUGHTS”. And no, I won’t add a disclaimer that my opinions are not OSNews’, because I am NOT employed by OSNews.


Oliver Herold wrote on December 28th, 2006 at 11:58 AM PST:

Beeing very realistic or just to speak tacheles, there should me more cooperation in open source than just “hey let us do something new” (inventing the wheel over and over again), but the drawback would be a huge loss of variety imo. This “dictatorship” within Redmonds software development is maybe one part of its power, but on the other hand it’s a nemesis too. Because people tend to over-regulation, so in the end you have one nice feature like described above, but you have some nasty “addons” too over the time.


Ludovic Hirlimann wrote on December 28th, 2006 at 12:18 PM PST:

>BeOS 5 back in 2001 was delivering the same smooth experience as Mac OS X offers today (as long as you had a compatible PC that is).

On a lot less powerfull hardware.


CPinto wrote on December 29th, 2006 at 3:50 AM PST:

Ciprian, I sincerly doubt that Mono will save GNOME from anything bad. A programming language, by itself, doesn’t make up a platform nor does it make it easier to create/develop a complex platform. Small disclaimer: I like Mono and choosing between Mono and C I’ll take Mono any day of the week mostly because my GNOME projects are hobbies/proof-of-concept and I want to get them done as quickly as possible so that I can enjoy them as soon as possible.

That said, I think it’s time for hardcore programmers to take the passenger’s seat and let architects take the driver’s seat for a change, without having them go through the nonsense of proving themselves as programmers. In my opinion, much like Eugenia says, there aren’t enough people out there thinking about the bigger picture and it shows.


Mark wrote on December 29th, 2006 at 10:24 AM PST:

“What agenda?

Mark, this is the second time you reply to my blog and you accuse me of things that are not true.”

You keep claiming incorrect things and I am repeatedly correcting you and you drop the conversation after that. I asked several questions which went unanswered yet again.

“You come here with PREJUDICE against me. If you have a problem with me please include a REAL email address so we can settle this directly”

There is no prejudice. Whenever I see incorrect claims made by people like you who are linked from news sites like osnews, I felt the need to correct those claims since people might be misled by this and would assume you have authentic sources for this.

I dont want to get into a personal bickering over email. If you want to refute something I said, lets have a public discussion. Do try to verify your claims next time you make them so that atleast the basic facts are correct.

“The figures are estimations and there to make a case, not to be substantiated in any way as 100% accurate”

How did you make a estimate of 30%? Simply pulling figures out of nowhere is not a good strategy. Stats are meant to be used on a scientific basis.


Mark wrote on December 29th, 2006 at 10:33 AM PST:

“GTK goes a lot slower than it used to be etc etc. DBUS and HAL and Fedora itself continue in a faster pace because thankfully, third party people have taken control of them”

Yet another misleading claim. DBus and HAL are primarily developed and maintained by Red Hat developers even today. J5 and David Zeuthen if you didnt know the name. I dont know where you get the idea that third party people have taken “control” of them.


Ciprian Mustiata wrote on December 29th, 2006 at 11:17 AM PST:

Eugenia: I’m fully agree with GTK+ development, that was slow down, right now only Mono can save GTK as a platform. The future of GNOME is as Miguel de Icaza said: that GNOME 4 should be written mainly in Mono. That will solve problems of performance, because Mono start became better and both in combination with GTK. GNOME needs to be a good platoform for develop, to make easy to maintain applications, to make easy to emerge the “Windows world”. Is hard to a C# developper to go to Linux and see that is no tool. Is hard for a C++ developper to work with GLib + GTK because are bad documented and has bad tools as you notice. And at the end is easier to go from C++ to C# than from C++ to C. I hope both in a cleaner implementation of GTK and a autogenerated documentation from source (like doxygen) at least or to move completly to C#, and that to make easier to maintain all GNOME project.


This is the admin speaking...
Eugenia wrote on December 29th, 2006 at 12:06 PM PST:

At last, a reader who while doesn’t agree with me, he replies his disagreement in a very nice and balanced way rather than taking the easy way out of personally accusing me of agendas, the Gartner-effect on percentages and having eaten all the cookies from the jar (I might have done the latter though).

Oliver, thank you very much for the comment. Noted.


johnSchaffer wrote on December 30th, 2006 at 11:27 AM PST:

GTK has never been good enough to build a platform on top of it, and as a result Gnome is seriously lacking as a development platform. As time progresses, the size of the desktop grows, and what is demanded from the platform is bigger than before. GTK is not up to the task, and this is why Gnome is evolving at a slower pace than before.

Gnome hackers are aware of this, and Mono is only an example. I don’t think Mono is going to save the day though, for both technical and political issues.

KDE on the other hand is a brilliant development platform, and has the technical abilities required to make desktop Linux happen, but until this day has not received a serious push from the rest industry. The perception is that no one wants to rely on Trolltech for their propietary developments. Not sure how much of that is true, though.

Anyway, with Gnome very far technically from being up to the task, and with those doubts hanging over KDE, the future for desktop Linux does not seem very bright.

More on topic… that Linux needs urgently standards and a global vision is the only sane thing to think. Maybe now that a millionaire is singing this same song something will happen and Linux will stop being chaos. Time will tell.


Comments are closed as this blog post is now archived.

Lines, paragraphs break automatically. HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

The URI to TrackBack this blog entry is this. And here is the RSS 2.0 for comments on this post.