i get asked fairly regularly how one can starting using kde4 right now. well, you certainly can grab what's in svn trunk/ and have at it, but you may be disappointed by the constant compiling (the libs are still in massive flux, and there's more where that came from) and how the big visual changes are not there yet.
so what's a kde fan to use? kde 3.5, of course. it needs love too, after all, and it is continuing to be improved if at a more calm pace. today i committed a change that removes one more annoying bevel from konqueror (the one around the view but inside the tabwidget; such small changes can make such a big difference IMHO) and i'm hoping to get the thumbs up from the usability pros to change the default for lines between toolbars in plastik to off. new translations, documentation and lots of bug fixes continue to make their way into 3.5. and of course the 3.5 apps continue to roll out of the various projects.
but someone on a mailing list pointed out in passing that many of those in the "bleeding edge" crew of people will likely move on to other options during the kde4 devel cycle just because it's going to take a while. we're not talking about an e17 or duke nukem forever type schedule, but it will be longer than our usual "what, it's 9 months already? new release!" standard operating procedure.
i wish this weren't so, but in my gut i can't help but think, "yep, people will move on in search of the latest and greatest something." this has a really negative effect on open source projects, as this fickle attention span can make it rather difficult for us to do longer development cycles (and therefore larger changes). why?
well, user base is everything for large projects. it's what keeps the q/a going, where we get new contributors from, where user support comes from, the pool for regional support at things like tech shows, what packagers use to gauge what packages to give more love to and more ...
in the proprietary world they just lock their users into their platform with file formats, hardware platforms and other nasties so they can take their time if they need to: their users ain't goin' anywhere. we don't do that (because we respect Freedom), and so our users are free to roam.
but when they do roam in larger numbers, that can impact the project. when the project does release that spiffy new version that took a year and a half to complete, it often has to start building the user base back up to where it once was. and no, most projects don't usually have the resource to simultaneously develop two trees indefinitely.
the impact of migration-due-to-boredom can be so bad that some projects actually consider not doing anything that might take too long, such as making larger architectural changes that would bring the project a quantum leap forward. so the fickle nature of our user base can, and sometimes does, have the unhappy effect of retarding important developmental jumps. some things just take longer that 6 months to do.
ok, i'm sure you get the point by now. =) so if you're a kde user today that really enjoys the platform, stick with us during the upcoming longer-than-usual release cycle. sure, check out what everyone else is doing too (don't do that already anyways? ;), but ditching kde 3.5 just because there hasn't been a major release for 9 months (as will happen sometime in the second half of this year) will only make kde4 a more difficult release for the project.
there will still be lots of exciting app development for kde3 apps this year to track, and 3.5 itself is still improving (such as the aforementioned visual improvements). by keeping with us during this slightly-longer-than-usual cycle you'll help support us on the way to kde4.
of course, you could also get your hands dirty and start contributing as well. i find that helps pass the time quite nicely ;)
Subscribe to:
Post Comments (Atom)
31 comments:
I think you are speaking a wrong language here....
Here is how I would say what you just said:
Users! Especially corporate users! Happy news!
This is a unique opportunity to have a platform stable for a long time. KDE 3.5 line is SO GOOD that all we can do to it is just add some cosmetic changes.
You can expect 3.5 line to remain "current" code for at least a year and a half! This is how we, the KDE community, can recognize the needs of ISV's and accomodate them. THIS IS THE PLATFORM TO BE ON AND RELY UPON FOR (up to 2) YEARS!
As the pending release of KOffice 1.5 will show, we mean productivity and user-empoverment without the shocks of escalating upgrade cycle.
We welcome you all!
You really should wash your mouth with soap, Aaron.
yes, the future of 3.5 is a god send for those who need stability.
it's the bleeding edgers and community base that are the concern here =)
Me as a bleeding edger,
I can't WAIT for KDE4,
but that's just curiousity.
Because I will keep running stable KDE 3.x on my normal desktop until KDE 4.x is proven to be quite stable.
(I will of course test KDE 4.x out the minute it's in alpha/beta/we on my test pc) :D
But I'm in great need of some Graphical changes and screenshots/videos of KDE 4.x,
but I know they aren't working on gui just yet?
Another thing though, is that it's better to wait long enough with the final release,
and make KDE 4.x include everything planned, and maybe even more :)
I looked at a Gnome 2.14 preview yesterday, and the gui is really cool now ;) not so "busy" as KDE 3.x.. So I do hope that KDE 4.x can keep up :) !!
"today i committed a change that removes one more annoying bevel from konqueror"
And this will be in 3.5.2? I love you man :). I know that I can be a pain in the ass in the mailinglists (FYI: I'm the main offender in the huge discussion in KDE-quality....), but I really do appreciate your (and others) work on KDE.
we, the home/end users of kde, are not single minded people, we love kde 3.5.1 for its stability and love to hear about all the bells and whistles from kde4.
I agree with you Aaron, the KDE marketing people or whatever the name you give them, should emphasis the importance of debugging the current desktop for a faster dev of kde4. (here better coding of kde4 = faster working kde4 !)
I'm so happy with 3.5 if kde4 was released today I wont upgrade ;)
i think the decision to allow a bit more than just bugfixes in the KDE 3.5 tree was a very wise one... keeps the 'i-want-new-stuff' guys on board, while it won't scare away the corporate users (if you do it slowly).
profox: beware the "previews" ;) but yes, we're trying to make kde4 kick as much ass as possible without taking overly long. when we do start releasing alphas/betas, i look forward to seeing your bug reports ;)
janne: yes, it'll be in 3.5.2
"today i committed a change that removes one more annoying bevel from konqueror"
Hooray! That extra beveled frame has annoyed me greatly since at least KDE 3.3, so I'm thrilled that it will be going away soon.
Could you by any chance do the same thing for Konsole? The bevel in Konsole annoys me even more than the bevel in Konqueror.
First: There is a poll on kde-apps which is related to this topic: here.
Besides that:
When I read your post I got the feeling that it was written for people like me: SIne I know how to make a svn checkout I compiled KDE 3.5 alpha-something from source to get an idea of it how it is and posted screenshots.
And since KDE 3.5 is stable, I check out "KDE 4" regularly and try to compile, start, play around - and waiting for the first really interesting new thing to post some nice screenshots.
But I also see what's going on at other places, and I already got scared if KDE is failing to keep up with gnome.
THe problem is that I see real screenshots of cool gnome stuff everywhere at the moment - but nothing than mockups and promises from KDE somewhere. :-/
I already know that I will have much more looks at gnome because of AIGLX, XGL, but also other stuff like networkmanager and LUKS-Hal integration (which is very, very important for me), and it is sad, because I would prefer KDE much, much more (face it: nothing can beat KIO-slaves in konqueror :-D).
To be honest: I have no perfect solution to get out of this dilemma. One idea would be to show the KDE bleeding edge users that there is already real code which is running on some machines - but even if you show only a short video it would lead to other trouble and would raise the question why this code is developed "hidden" - don't forget, you were one of the first people who complained about hidden development!
My preferred way would be: set yourself a date from which on you release the first, blowing away release (like a quite stable alpha or something). Until then you just havce to hope.
Even if some people are using more gnome for a few weeks, that does not mean that they switch their heart also ;-)
And, hey, I'm learning a little bit more c++ every day at work, and I really hope that I can use it at KDE development in some distant future to give a little bit back :-)
Indeed, but kmail is acting strange as reported in 119510. It's functionality that always worked, and because somebody not in the real (user) world removed a hack to keep it working.
Organizer has problems saving files on remote locations. At least it reports this constantly and due to this it is not possible to use kontact inside korganizer.
Often kmail marks a folder ias readonly or something like that, while the session was closed down
nicely. It keeps om retrieving large imap folders constantly over a slow link while the folders are already retrieved n dimap mode.
Yeezzzz, it is really hard to keep on using kde (ahum kmail), with such bugs. The 1st and the 2nd I reported already. But the 1st one is just neglected. Frustrating for the user....
Aaron: I haven't seen REAL official previews of KDE4 yet ?
I have checked out all available mockups though..
I have great hopes for KDE4, and I do keep an eye open for KDE4's development.
It scared me a little though... It seems like SO much is planned..! There has to be alot of work going on if you want to do everything in a relatively short timespan!
Thumbs up if KDE4 will have the GUI and all the cool new technology i'm hoping for, while still having as low as possible System Requirements / CPU usage / RAM usage, AND if it's stable enough for desktop usage..
Somewhere I really doubt it if this will be possible, but it's open source, so nothing is impossible :) . . .
"But I also see what's going on at other places, and I already got scared if KDE is failing to keep up with gnome."
I sometimes get that feeling. And yes, GNOME IS gaining momentum, and usually at KDE's expense. Fedora is GNOME-centric, Ubuntu is GNOME-centric (and as good as Kubuntu is, it still plays second fiddle to Ubuntu), Novell is GNOME-centric. There are still few KDE-bastions: SUSE and Mandrake come to mind. But Mandrake has been losing mindshare, and SUSE has been beefing up their GNOME, so it will soon be a GNOME/KDE-distro, instead of KDE-distro.
Yes, I know that Novell supports KDE. Just look at KDE's technical working group, and the number of Novell-folks there. But I think that everyone agrees that Novell is more focused on GNOME than on KDE. Just look at their new desktop-technologies. They were demonstrated on GNOME, KDE was not mentioned at all. All their new technologies are for GNOME. Where does KDE fit in here?
That said, I still see KDE improving at a trendemous pace. And the abovementioned perception might be just that, a perception. GNOME-guys often make big headlines about some cool new stuff, while KDE-guys quietly create something similar, without making a fuzz about it.
But still... I'm just not seeing that much support towards KDE from Novell, espesially when compared to their GNOME-efforts. Mark Shuttleworth is prodding up Kubuntu, but Ubuntu is still their primary distro. They have several people working on GNOME, how many people are there working on Kubuntu? Fedora's efforts are all GNOME-centric, while SUSE is paying more and more attention to GNOME. While there are quite a number of distros that ship with KDE, even using it as their default distro. But the number of distros that do active desktop-developement (as opposed of merely shipping some desktop, maye with their branding) is quite small. And most of those seem to be focused on GNOME.
What the hell is going on here? Maybe Trolltechs support compensates party for the difference, but I would still love to see more distros doing active KDE-developement, instead of merely shipping KDE.
Is my perception completely off-base, or is there a kernel of truth here? If I'm at least partly correct, it begs the question: what can we do about it?
> or is there a kernel of truth
> here? If I'm at least partly
> correct, it begs the question:
> what can we do about it?
what's true is that Novell purchased both Ximian and SUSE, and that the Ximian people continue to be rabidly pro-GNOME and that bleeds through in a lot of Novell stuff.
but i really wonder at the amount of time people spend obsessing over distros. they come, they go, the rise in imortance, they diminish, they change focus, etc, etc... they are part of the scene, but not the irreplacable core of it.
what *really* matters, and what people are often really looking to the distros for, is:
that a project has a healthy community of developers
that a project is continuing to meet real world needs
that a project's software is being deployed widely
that a project's efforts are supported by third party ISVs/IHVs as appropriate.
kde is doing well all on all those fronts, and historically we've done that w/out the dominant support of distros.
"but i really wonder at the amount of time people spend obsessing over distros."
Well, I'm not, I'm obsessing over KDE :). I don't really care if some distro disappeared, and some other distro took its place. What I do care is what those distros do to KDE (or GNOME).
It would be very nice if we had distros throwing their weights behind KDE, like Ubuntu, Fedora and Novell do with GNOME. I see SUSE doing some of that stuff, as well as Mandrake. But, I already mentioned the problems with SUSE and Mandrake.
Yes, KDE does have wide support with the distributions. But most of them seem to merely ship KDE with their branding. Compare that to Fedora, Novell or Ubuntu. They not only ship GNOME, they have hackers working to improve GNOME. All that whiz-bang 3D-technology Fedora and Novell have demonstrated? All GNOME-centric. We do not have any distros saying "look at this cool piece of technology we have developed! We will be using this in our next version, and that version will be running KDE".
This might be a timing/PR-problem as well. KDE4 will propably have comparable technology and eye-candy, but as things are today, GNOME gets the press.
I'm not saying that KDE is doomed if it doesn't get any major support from the distros (as opposed to merely shipping KDE). But I think that everyone agrees that things would be better, if we had some major organisations throwing their weight behind KDE. GNOME has Fedora/RH, Ubuntu, Novell all actively pushing and developing GNOME. KDE has SUSE (diminishing), Mandrake (losing mindshare), Linspire (do they do major KDE-coding? I remember some of their more interesting stuff is closed). Plenty of organisations and distros are actively using KDE, but they are not actively developing KDE.
The reason that Sun, RedHat, and now Novell are primarily Gnome shops is already known, but it's one of those things that is forbidden to talk about in the KDE community.
> Ubuntu
well, ubuntu *is* putting weight behind kde by investing in kubuntu. the reason it's not more obvious is because the public faces for ubuntu aside from mark are pretty rabidly into gnome (similar to how i'm pretty rabidly into kde =)
> They were demonstrated on GNOME,
> KDE was not mentioned at all.
you can thank nat friedman and the rest of the ximian boys for the current state of public affairs at novell.
but this does not reflect all internal software development there.
> All that whiz-bang 3D-technology
> Fedora and Novell have
> demonstrated? All GNOME-centric.
not at all. compiz is designed from the ground up to be desktop agnostic, for starters. and XGL is by its very nature non-desktop specific.
the only desktop specific stuff is that novell did some early integration with gtk+ and red hat is working on metacity. that latter is absolutely no surprise since that's what they've been doing all along.
and do you remember who shipped a composite enabled window manager first that could do transparent windows? yep, kde.
i think you're getting a bit lost in the current marketing bluster and putting more emphasis on the importance of distributions in general.
> But I think that everyone agrees
> that things would be better, if
> we had some major organisations
> throwing their weight behind KDE
of course; but distributions aren't the only organizations out there. and having distributions behind them hasn't exactly done GNOME the greatest set of wonders in the world.
the feature lag and lower % of the open source market speak volumes in my mind.
> Linspire (do they do major KDE-coding?
they are a smaller shop, but yes they do contribute such as paying matt rogers to work on kopete.
> Plenty of organisations and
> distros are actively using KDE,
> but they are not actively
> developing KDE.
this is, thankfully, patently untrue. i know of many companies that develop kde apps and technologies, several of whom do it as their primary business.
"well, ubuntu *is* putting weight behind kde by investing in kubuntu."
True, but Ubuntu is still their "primary" distro (or so it seems at least). Kubuntu seems to have only a fraction of the resources that Ubuntu has. Hell, their (Kubuntu included) release-schedules are tied to GNOME!
"and do you remember who shipped a composite enabled window manager first that could do transparent windows? yep, kde."
Yes, I do remember that :). But it was created by KDE-developers. It seems that GNOME enjoys lots of attention from distros and others, whereas most work on KDE is done by KDE itself.
"i think you're getting a bit lost in the current marketing bluster and putting more emphasis on the importance of distributions in general."
Like I said, I don't think that KDE will be doomed if it doesn't get any support from distros. What I AM saying is that for some reason GNOME seems to have lot of developement-support from distros, whereas KDE has less so. Of course we can downplay the importance of the distros, but in the end, things would be better if distros supported KDE more actively than they do today. GNOME receives lots of such support, whereas KDE seems to receive less. of course, I'm not a developer, so I might not see the "big picture" here.
And marketing does matter.
"this is, thankfully, patently untrue. i know of many companies that develop kde apps and technologies, several of whom do it as their primary business."
Well, no. I didn't say that none of the distros/organisations that use KDE, develop and improve it. What I meant is that GNOME seems to have an advantage in that their users (distros etc.) are often developers as well. Fedora, Novell and Ubuntu do major work on GNOME. And while there are many KDE-using distros out there, they don't generally speaking work on KDE as suchb (with few forementioned exceptions). Mandrake and SUSE do, as does Linspire, but that's about it. Most others merely use and ship KDE.
In short: Ubuntu ships GNOME, and they also spend lot of resources to improve GNOME. Same thing with Fedora and others. And that seems to be rarer with KDE, with SUSE, Mandrake and Linspire being the major distros actively developing KDE, as opposed to merely shipping it. Kubuntu might fit in there as well, but it's still in the shadow of Ubuntu. And looking at the KDE-camp of distros, I do get a bit worried. SUSE (THE KDE distro) is beefing up their GNOME these days, Mandrake has lost mindshare (to Ubuntu primarily I think), and Linspire isn't that widely used. There is Kubuntu, but it lacks the support Ubuntu seems to have. In the GNOME-camps things are a bit different. Ubuntu is the current sweetheart of distros, with Fedora being widely used as well. Novell is a big business with lots of resources.
How could we turn "distros that ship KDE" in to "distros that ship and improve KDE"? KDE would propably survive without active support from distros, but I bet that those additional resources wouldn't harm KDE, quite the contrary ;).
What I AM saying is that for some reason GNOME seems to have lot of developement-support from distros, whereas KDE has less so
I guess I'll have to spell it out for you Janne. Trolltech owning Qt and charging for non-GPL compatible software is the reason why corporate interest is in Gnome.
KDE made their bed a long time ago and now they have to lay in it.
I suspect aaron will go into full-fledged denial, but all the developers know it's true - even if they don't want to publically admit it.
> Kubuntu seems to have only a
> fraction of the resources that
> Ubuntu has.
there are a couple of people who package gnome for ubuntu (i think it works out to something like 1.5 people), and 1 guy who packages kde for ubuntu. there's one guy who works on adept as well.
others on the payroll in ubuntu work on OS level issues that both desktops use.
what does make it seem otherwise is that jeff waugh is the ubuntu face to the world and their ubuntu weekly newsletter is gnome centric.
> Hell, their (Kubuntu
> included) release-schedules are
> tied to GNOME!
kubuntu's releases are in step with ubuntu's releases, which are every 6 months. that was decided a good idea by the same people who decided a 6 month release cycle is a good idea in gnome. there's nothing more to it than that.
> Ubuntu ships GNOME, and they
> also spend lot of resources to
> improve GNOME
no, actually, they don't.
> And that seems to be rarer with
> KDE
so ... novel, red hat and ubuntu line up behind gnome. ubuntu offers meager resources to the mix; red hat isn't doing a lot of gnome development these days; ditto for sun which used to be in that line up every time someome started on this old saw but since they've gone into the background marketing and development wise everyone just ignores them.
on kde's side we have suse/novell, mandrake, linspire, kubuntu, xandros, red flag ... of these suse, mandrake, linspire and kubuntu all put resources in.
but you know what's more important to me? the fact that the european space agency puts money into kde devel (via the kst project); that sourceXtreme puts resources into development; that KDAB, Creadativ and other European tech companies do the same; that Trolltech puts serious money on the line; and on and on...
those resource are far more interesting and important to me than the flagging interest of distros who are try to sell desktop linux into the enterprise where it isn't really being adopted very strongly.
> but I bet that those additional
> resources wouldn't harm KDE,
depends. i haven't seen the distro involvement exactly do wonders for GNOME's community developers. i haven't seen it lead to a project with a coherent, forward-thinking core. they've done a lot of good, but it's also come at a non-negligable price.
there are distros that i think do very well, but some distros just want way too much control and to drive projects towards their business goals.
it's not the cut and dry "obvious good" that so many people seem to think it is =/
or do you think it's accidental that with our more community and non-distro corporate backing that kde tends to move faster, have a more coherent project and a larger user base?
> Trolltech owning Qt and charging
> for non-GPL compatible software
> is the reason why corporate
> interest is in Gnome.
yes, some companies don't like this model. thank goodness there are other options for them in the open source world then!
most of these people are simply concerned about paying any money whatsoever in any circumstance (gotta love those who won't give back =), because as you probably know Qt is GPL'd (aka "Free") and we have a contractual relationship with Trolltech (which we recently updated and renewed) that ensures that Qt is never made less Free.
BUT ... this is not the overarching reason for "corporate interest in GNOME". in fact, much of the "corporate interest in GNOME" is marketing-driven myth. for every one company developing with GNOME tools i can show you several developing with KDE tools. for every GNOME deployment i run across (and they are growing in number! hoorah for the open source desktop!) i run into several KDE deployments.
Red Hat went with a gnome investment strategy because at the *time*, Qt was not under an OSI approved license and they strongly believe in Free Software. Qt is fully Free on all platforms now, but Red Hat's investment is non-negligable and so not something that will move quickly if ever. and that's ok. they also support KDE these days by shipping it and providing paid support for it (that's why they ship it ;)
Novell's movements have everything to do with their Ximian purchase.
Sun and HP have both touched and been burnt by the flame of the Linux desktop, so their choices are moot and, i think, interesting given their success level.
how is that for spelling it out?
Sorry Aaron, but most people realize a simple fact of lowering the barriers of entry into a market already dominated by Microsoft. The LSB desktop working group realized this (at least at one time) and had a policy of "no strings attached".
Now you can give all sorts of alternative reasons why RedHat and Novell went with Gnome, but the simple fact is that they realize that if the Linux desktop ever takes off (looks more and more unlikely now), these companies can't go to their customers and say "By the way, there's this little Norweigian company that owns all the copyright to the underlying toolkit that powers our desktop, and if you want to write something not GPL-compatible you'll have to buy developer licenses from then. And yeah, we know that every other framework and toolkit is free to do with you want, but we chose KDE because it's a technically superior platform".
Now I expected your denial Aaron. I don't think you could stomach the political fallout from actually admitting what everybody already knows. And at this point, for people like you, there's probably no benefit from admitting it since you can't do anything about it. But most of us that have been around Linux for the past decade or so, but don't get religious about open source or Linux or any software for that matter can look objectively at why desktop linux has failed to be anything, but yet another niche market.
If Qt had just been a community effort instead of driven by corporate interests and given a proper library license such as LGPL or GPL with exceptions then Gnome probably wouldn't have been started and KDE would be "the" desktop for Linux, with ankle-biters such as E16-17 and others in the background.
P.S.
most of these people are simply concerned about paying any money whatsoever in any circumstance (gotta love those who won't give back =),
That's just a retarded statment since you have to "give back" to projects like gtk+ or any other GPL project if you distribute modifications. What you mean is that if you're not going to let your own independent software be affected by the viral Qt license then you'll have to pony up to the Trolltech corporation.
A couple more points...
See, what people need to understand (Aaron works for Trolltech, so his opion is pretty much irrelevant on this), is that Trolltech and KDE have competing interests that are not necesarily compatible.
What could have happened, and what would have much better for KDE, is for Trolltech to have liberalized the Qt license and wrote apps on top of it for sale - like everybody else does. That way other people wouldn't have been beholden to whatever pricing scheme Trolltech devised at one time or another. The other scenario would have been for someone to buy out Trolltech, liberalize the license, and write proprietary apps on top of it.
Of course you would then have the GPL/Open Source jihadists screaming foul because of "evil" proprietary apps, but if you listen to them then you're doomed anyway.
A lot of people in KDEland have been talking about marketing and promoting lately. The problem that desktop Linux has always had and most likely always will have the problem that there is no "the desktop linux". It's hard for ISVs to target because there are no standards - even though some people talk about it, but nothing ever seems to come of it.
So to reiterate, if Qt would have had a proper license, Gnome would not have been started, KDE would be "the desktop" for linux, and there would be something for ISVs to target and they could freely develop for KDE without any strings attached like all the other operating systems. Linux would probably have at least 10% (if not more) of the desktop market by now.
But instead now you have all these "distros" with their own little private repository universes, and two desktop environments, and a major deployment headache for everybody that isn't in some repository because of dependency hell.
At this point in time, the only hope is for someone like Google (if they actually cared, which they probably don't), is to take the kernel, Xorg, some GNU tools and write a real desktop operating system from the ground up.
KDE is a great desktop and it rocks architecturally, but it's always going to be hamped by the Qt license. Aaron, I'm not going to try and convince you, since your already tied into Trolltech and KDE, but for people like janne that don't understand the situation.
> Aaron works for Trolltech, so
> his opion is pretty much
> irrelevant on this
go look at my stand on this before last summer when Trolltech started sponsoring me. you'll see it was no different then.
> If Qt had just been a community
> effort instead of driven by
> corporate interests and given a
> proper library license such as
> LGPL or GPL with exceptions
first off, Qt isn't the only toolkit with "corporate interests"; if you look at Gtk+ it's very much driven by corporate interests. and that's not a bad thing.
as for the licensing, yes, it is unfortunate it wasn't GPL'd from day 1. thankfully today it is, and it has been for many years now.
> can look objectively at why
> desktop linux has failed to be
> anything
sometime if we get the chance we can grab a beer and discuss why we today have somewhere between 2-4% market share. but it isn't Qt's licensing. i wish it were, because then it'd be an easy win.
no, there are many other issues and of far greater impact such as IHV support, easily purchasing systems with Linux on them, working in a monopoly situation, having started only 9 years ago (GNU started 24 years ago), etc, etc, etc...
good news is that new deployments of open source desktops are happening more and more. it's speeding up, not slowing down and not stopping.
you may feel it's a failure already, i see it growing. as do most others who are actually involved and in the know.
> the viral Qt license
you mean the GPL.
> the only hope is for someone
> like Google
who uses KDE as the default desktop on their internal distro, btw.
> Aaron, I'm not going to try and
> convince you, since your already
> tied into Trolltech and KDE, but
> for people like janne that don't
> understand the situation.
i hope you'll forgive me if i don't just stand back while you share your mistaken viewopints on this. i don't expect to convince you of anything either; you've already made up your mind and that's fine. everyone's allowed to be wrong ;)
where you an i differ is on this: it isn't licensing issues (one way or the other) that's helping or hurting the open source desktop; that the open source desktop is growing just as the open source operating system did in the server space in the 90s; i don't feel it's a lost cause, and i do see the connection between the desktop succeeding and the progress on the server side.
Aaron
I'll put aside the Qt license issue aside for the moment, except to respond to your gtk+ corporate interests remark by saying that the difference is that RedHat no Novell nor anybody else owns all the copyright on it.
But putting that aside, one of the biggest problems I see for desktop linux adoption is the distro repositories and dependancy issues.
It's so limiting to be an ISV (or just random guy that wants to distribute a binary of his program), and not being able to deal with dependencies problem in a sane way.
What I see as needed, is a distro neutral package system (maybe autopackage) that can tie into the native distro package management system, but can also deal with dependencies in a distributed fashion with fine-grained versioning. Have you checked out conary packaging that is used by the Foresight distro?
There does seem to be more cooperation these days between freedesktop, KDE, Gnome, etc..but I think there needs to be much more. Hopefully Dbus and some type of cross desktop component embedding will be a common denominator in the future.
But I think both of these issues just illustrate the fact that there isn't enough cooperation among distros, desktop developers, kernel people, other interests in competing against the likes of Microsoft and Apple. That's just one of the pitfalls of open source that is almost antithetical to presenting a unified desktop system.
Let's face it..having two dominant desktops instead of one isn't good for desktop linux market share.
Vista coming out this year and Apple going with Intel doesn't bode well for desktop linux either.
> that the difference is that
> RedHat no Novell nor anybody
> else owns all the copyright on
> it.
and that would be problematic for Qt if it weren't for the FreeQt Foundation and the contract we have with Trolltech that prevents them from removing Qt from the GPL (the penalty: we get to BSD the whole thing)
> Have you checked out conary
> packaging that is used by the
> Foresight distro?
yes. it's quite interesting. unfortunately packaging is mostly out of scope for my areas of responsibilty in the community, but i agree with you that it's a problem right now.
mostly the binary compat and API reliability issues. these are things the LSB is working on, that the Portland Project (from a desktop perspective) are working on and that FreeDesktop.org is working on.
we do have a couple decades of crap behind us to clean up and it's taking a few years to do it.
> Hopefully Dbus and some type of
> cross desktop component
> denominator in the future
DBus is already fulfilling the IPC part of that equation; common component models are a ways off, but it's not entirely clear they are absolutely necessary. have a small number of interoperable ones may turn out to work just as well. we'll see =)
> having two dominant desktops
> instead of one isn't good for
> desktop linux market share.
that's a tough one to call. it's certainly caused an amount of uncertainty, and it's certainly created division lines amongst distros.
on the other hand it's given us the ability to tap larger audiences of developers; ensured that the mistakes of one small group of people don't endanger the whole thing (risk spreading); allowed us to experiment more and gave us a team to compete with.
hard to say which has been bigger: the down side or the up side.
from the deployments perspective, we continue to pick up steam, and that's all that's important in the here and now.
(i'm really not happy with how involved distros are in technology direction these days, as their business priorities and the needs of open source often lie along different lines. they can be made to harmonize nicely, but they aren't the same.)
> Vista coming out this year and
> Apple going with Intel doesn't
> bode well for desktop linux
> either.
it's just more of the same from them. no different than their last big releases (save for perhaps the DRM issue). they still don't offer the future data guarantees, vendor freedom or flexibility we do.
all the while we're closing the gap on hardware and integration issues.
> Vista coming out this year and
> Apple going with Intel doesn't
> bode well for desktop linux
> either.
it's just more of the same from them. no different than their last big releases (save for perhaps the DRM issue). they still don't offer the future data guarantees, vendor freedom or flexibility we do.
all the while we're closing the gap on hardware and integration issues.
See, there's your problem Aaron. You just dismiss Vista and the Macs with "more of the same" when that's obviously not true.
Vista will have a whole new presentation layer (all the eyecandy), a managed programming model, and win32 subsystem for backward compatibility.
Apple's move to intel should offer better price, performance, and choice. And I'm sure you've already seen some of the switching that has gone on with open source Linux hackers to the Mac.
So you just dismiss these two platforms and in 2016 desktop linux is still hobbling along as a hobbyist desktop platform, and you'll wonder what happened. Actually, it's 2006 and you should be wondering what happened. Year 2000 - it's year of the linux desktop. 2006 comes along. What happened?
We know what happened. Factionalization, horrible deployment problems, no unified desktop to market, bad licensing decisions, getting religious about source code, etc...
I know you have to be optimistic for obvious reasons, but by just offhandedly dismissing the competition you're just denying the reality of the situation.
> Vista will have a whole new
> presentation layer (all the
> eyecandy), a managed programming
> model, and win32 subsystem for
> backward compatibility.
let's get back together in 3 years time and see where all that took them, shall we? =) microsoft's technologies are always compelling until they release them. they are also the incumbent monopoly and have nowhere to go but down.
> Apple's move to intel should
> offer better price, performance,
> and choice.
i'll believe that when i see it. well, i believe the performance bit but the choice and price? heh. yeah. keep wishing. Apple is a company and they have realities to meet.
> And I'm sure you've
> already seen some of the
> switching that has gone on with
> open source Linux hackers to the
> Mac.
yep. there's been some the other way, but there have been more move to MacOS X. some will stay, but i'm confident we'll get many of them back. in the next year we're delivering most of the reasons people switched, and Apple still doesn't have the benefits we started with.
> Year 2000 - it's year of the
> linux desktop. 2006 comes along.
the people who proclaimed 2000 as the year of the desktop, or any year since, were quite frankly stupidly wrong.
there never will be a "year of the linux desktop" since it's a multi-year journey of progress year after year. there never was a year of the server, either. just years that more people became aware of it. but it was a similar long (though exciting) haul.
"I guess I'll have to spell it out for you Janne. Trolltech owning Qt and charging for non-GPL compatible software is the reason why corporate interest is in Gnome."
I'm well aware of that "issue". But I don't care about it one bit. Seriously, since when did this free-software thingy turn from creating kick-ass software that is free (in speech and in beer), in to satisfying the whims of peddlers of proprietary software? Why do I keep on hearing people whine "Qt is great! if only it allowed others to create proprietary software for free". I don't say this often but.... If you are so gung-ho about proprietary software, maybe you should just use Windows? Seriously? It has all the proprietary software you could ever want!
I care about free software. I don't give a flying f*ck about proprietary software (Yes, I do have to use it sometimes). And I really fail to see that why should we (as a free-software community) suddenly start to care about the well-being of creators of proprietary software. I use Linux to get away from them. Why should we feel sorry for them? Why should we try to make their lives easier? They are not interested in helping us out, yet we should bend over backwards in order to satisfy them and their whims?
If what you say is true (and I doubt it), then I want nothing to do with it. This is about free software, if we start to jump to the tunes of proprietary software, we are selling our souls, as far as software and it's freedom is concerned.
Why do I doubt what you say? Well, look at Fedora/Red Hat. They support GNOME, but they do not create any proprietary software. Same thing with Ubuntu. I haven't seen anything proprietary GNOME-stuff come from Novell either. In short: I think you are full of it.
Hi Aaron. I was just trying out Akregator for the first time, and lo & behold, a couple of your blog entries were available. Because I remember the favour you did for me in 2001 [you let me come over to your house, even though I was a stranger, and get a copy of glib or glibc for my RH 6.* box], I thought that I'd check it out.
Regarding this entry, I think that it's sad that KDE 4 is a ways off, but I'm happy for the stability KDE 3. I'm using an old Pentium Classic, and running the latest stable Gentoo KDE. It's slow, but I'm very happy with it. At this point, I'm very, very pleased with features and performance. Even the crashes and occasional freezes aren't a problem. Speaking of features, I just did a spellcheck for this comment. I'm just so impressed. Opera is really the only non-KDE app that I enjoy, and I enjoy it pretty much only because of the speed.
Thanks for your work.
--
Sincerely, and with thanks,
Eugene T.S. Wong
After reading all the comments here, I feel that a big virtual hug and thank you are in order. Whether or not KDE is successful in the future of the free desktop is irrelivant(<-yes I know I spelled that wrong, thanks to Konqueror!) when compared to the amount of generous work Aaron and other KDE hackers do for us. The technology they create will be used for many years to come. If not in KDE, in something else. That's how open source works, and why it's what I believe in. So thank you Aaron and all the other KDE devs out there.
Post a Comment