MAR
28
2008

Aaron Seigo Talks About Kontact's Bright Future

Sirius' Tom Callway interviews KDE's chief-hugger Aaron Seigo. He talks about KDE 4, communication within the project and what effect the 'new platforms' (Windows and Mac OS) have on KDE development. Seigo also talks about the KDE 4.0 release and how that compares to the 2.0 and 3.0 releases of KDE. About Kontact, he explains, "We are going to see some very interesting developments happening when Kontact is available on all platforms. For instance, finally there will be a groupware solution that looks and behaves exactly the same on all platforms (a support win) that lets you choose your groupware server (a server side win). Kontact represents the client side of the first realistically competitive threat to the Exchange-plus-Outlook hegemony. And that's just one application."

Comments

"Windows offers *keyboard only* driven GUI (you can virtually do anything without using a mouse) - the thing which is still *not* possible in *any* Linux DE."

"What can be done with keyboard on Windows that cannot be under KDE?"

Well, I do not know, but unfortunately there are certain things that cannot be driven by keyboard in KDE.
One issue in question are the vertical "side tabs" in Konqueror. Navigating between the various tabs (root directory, home directory, services etc.) can only be done by mouse (I tried to find a possibility to assign a key combination, but I failed. So if there *is* a way to do it, I would welcome any suggestions.) The same with Kaffeine and Amarok (they also have these vertical "side bars").


By Imruska at Wed, 2008/04/02 - 5:00am

with tab you should be able to give these tabs focus, and then you can open them... Haven't tried it, though.


By jos poortvliet at Wed, 2008/04/02 - 5:00am

I could open sidebars with tab but I couldn't focus in them. Anyway, I don't se any reason to navigate such an interface with keyboard - with keyboard typing in paths with completion is much more convenient. I don't think many users choose Windows because they can navigate such things in Windows.


By Grósz Dániel at Sat, 2008/04/05 - 5:00am

Well, I will only respond to two points here.

First, I've seen (and used) Qt apps run under Windows and they certainly feel native to me.

However, even if they were slightly different in an aspect here and there...well, they're in good company then. Not even Microsoft's own software maintain a consistent look and feel to them. Vista looks and behave in one way. Internet Explorer is slightly different and not consistent with Vista as a whole. Office 2007 deviates even more, and so does Windows Media Player. Windows users don't seem to mind, so if Kontact (for example) doesn't blend in completely is unlikely to be a major showstopper. It may or may not be suitable for other reasons but it's un-nativeness is essentially a non-issue.

And as far as your beloved KDE3 apps go...what's stopping you from using them in KDE 4? Those I've tried run just fine in KDE 4 with no recompiling necessary. True, they don't take advantage of the new features but a Windows app made for XP won't take advantage of Vista's new features either (but there's no guarantee that they will run. API breakage is not exactly unheard of in Windows or MacOS X either).


By Jonas at Sat, 2008/03/29 - 5:00am

>> Do you know why Windows is still tenfold more popular amongst developers than Linux and even Java?

Because most of the best applications and the most useful applications for users out there are written for Windows. And the users are not programmers, they will not care if about binary compatibility ;).

>> My second point is that ... we already have Java. We have .Net managed code

Use java to build a ultra-complex Graphical User Interface to achieve cross platform and avoid recompilation? Thanks, but no.

I have already had enough experience with it. Java is good for server-side applications and client side applications that don't need too much ultra-fancy painting capability. Java VM + bytecode interpretation will make your program slow as turtle (I don't care if people say Java is not slow anymore, my measurement still says Java *is* really slow)

>> Because Windows offer extremely *polished* desktop experience
Maybe true


By fred at Sat, 2008/03/29 - 5:00am

> Windows offers *keyboard only* driven GUI (you can virtually do anything without using a mouse) - the thing which is still *not* possible in *any* Linux DE.

A friend of mine has a notebook with a broken touchpad. He's using his KDE 3 just with the keyboard, and that does not slow his workflow down.


By Stefan Majewsky at Sun, 2008/03/30 - 5:00am

Do you know why people still prefer to *pay* for buggy, virus prone, registry breakage prone Windows? Because Windows offer extremely *polished* desktop experience. Once set up properly Windows doesn't tend to crash here and here. Windows offers *keyboard only* driven GUI (you can virtually do anything without using a mouse) - the thing which is still *not* possible in *any* Linux DE.

Polished? Windows is anything but polished... Start ms word doc, minimize it, open a second word doc, the other one is maximized - totally crazy. Or try alt-tab - most apps open a 'hidden' extra window, making alt-tab incredibly annoying. Need more examples? Open a bunch of webpages so the taskbar starts to add all windows into 1 button. Now enlarge the tab bar - no matter how much room there is, it won't put em back individually. And have you ever heard of focus stealing prevention? Redmond clearly hasn't (Yeah, it's Lubos, our KWin maintainer who invented it some years ago). And I can go on and on... Linux does ALL of these things perfectly, has done so for years.

And you complain about keyboard driven gui - most Windows apps can't change shortcuts, entirely unlike KDE - where you can always change all shortcuts. We're lightyears ahead in that area...

If there is ONE desktop which is extremely unpolished, it's MS Windows. By far.


By jos poortvliet at Wed, 2008/04/02 - 5:00am

> My second point is that ... we already have Java.

How much market share has Java on the Desktop? Right.

> We have .Net managed code. We already have XML driven user interface. While > KDE4 is a fantastic DE, the world has already moved further to a point where
> recompilation is not necessary at all.

KDE also has "XML-driven User interface" (KXmlGUI). Of course that doesn't cover all your application, as the actual GUI is always only a tiny part of the app. And in Java, the vendor has to maintain binary compatibility as well.
E.g. you can't add a method to a public interface without breaking BC.


By Frank at Sat, 2008/04/05 - 5:00am

I think Aaron takes the right lesson. For the platform it would have been better to make a gradual transition. Eg. to first port KDE 3.5 to QT4 as a KDE 3.6 and then work on the next big things. Also the Suse kicker is available for a long time. You need a platform with users, then the rest will just emerge.


By andy at Sat, 2008/03/29 - 5:00am

Porting unmaintained KDE 3 code to Qt 4 that is to be dropped and replaced anyway would have been waste of time.

"the Suse kicker is available for a long time"

Do you mean Kickoff?


By Grósz Dániel at Sat, 2008/03/29 - 5:00am

Hello,

and here I wonder why we keep getting told that KDE 3.5 _is_ still maintained for years? That doesn't seem to fit into your statement.

And Plasma team is investing a lot of time into explaining why Plasma can't do anything at all yet.

To me, it's clear. Kicker and kdesktop are applications. As these they were _not_ within the scope of KDE 4.0.0, which was to provide the libraries for the applications.

All in all, KDE 4.0.0 was taken hostage by dropping kicker and kdesktop, to the point, that KDE 4.0.0 had to be delays by months and _then_ still was not feature complete in _only_ that, for even the most simple and core desktop tasks.

Saying that investing a few weeks into something would have been a waste of time, means to somehow not recognize the damage that has been done to the project. I can very well imagine it wouldn't have been fun. I can very well understand that nobody did it. But then, did anybody see a chance to do it?

There were people highly enthusiastical about KDE 4.0.0, like that guy who wanted to do all the linguistic tricks. But now I guess, KMail 4.0 won't be able to spell check different languages per paragraph. Happy that we got some cool gizmos instead? Me not.

Probably I making up options, probably that guy would have droped out anyway, but as somebody who actively tried to run SVN, I can tell you, that it turned me away that there simply was no desktop shell for many months.

I strongly advise KDE to never ever let anybody put anything on a criticial path before he or she has shown the code. Were Plasma really such a great thing to must have it in 4.0.0, we certainly should have had the code first, not?

But coming back to sanity, hey, how was every decision to be made to be 100% right, or else we get mad at each other? A lot wise things were decided for KDE4.

Besides, not porting the workplace shell is not among them, but then again, the one who now understood how great an idea it is to port to MacOS and Windows, probably will understand as well, why Plasma should be ported.

Well, once people will be able to stop laughing about Plasma as a suggest replacement for anything now functional.

Yours,
Kay


By Debian User at Sat, 2008/03/29 - 5:00am

So all the Plasma developers are all magical super-good PIM programmers? Not all programmers are proficient or even enjoy to code in all areas of programming. Plasma is not the cause of KDEPIM being delayed to KDE 4.1.


By Jonathan Thomas at Sun, 2008/03/30 - 5:00am

No, Plasma is not a cause for delay of 4.1 which was supposed to be the release for applications to catch up. Plasma is the cause for delaying 4.0 and should itself always have been part of 4.1 only.

The delay is work for KOffice besides, I think they will miss the 4.1 time frame as well or drop almost everything application-wise from their release. A pity.

The thing that Plasma did, was making the SVN version unusable for a long time, which definitely has turned away people from trying to be a part of it.

Let me put it that way: All the arguments why 4.0.0 needed to be released with Plasma in its nascent form are actually arguments why 4.0.0 needed to be released without Plasma.

Yours,
Kay


By Debian User at Sun, 2008/03/30 - 5:00am

As much as I respect the work going into Plasma, I'm afraid I must say I'm inclined to agree. Do you want to know how Kompare was ported to KDE 4? Using a kdelibs4 package on KDE 3.5! To a point where I proofread my changes in... Kompare from kdesdk 3.5 (because that's what was integrated in my desktop)!

That said, Plasma _is_ improving quickly, 4.0.3 is essentially usable (thanks also to the feature backports; I must say I really don't understand the people complaining about those, they were very much needed!) and I'm no longer worried about shipping that in Fedora 9. (Shipping with 4.0.0 or daring to ship one of those so-called "release candidates", however, would have been a disaster.) There _are_ some remaining annoyances though, e.g.:
https://bugzilla.redhat.com/show_bug.cgi?id=439636
http://bugs.kde.org/show_bug.cgi?id=160073
(Yes, that's the same annoying bug, we asked the reporter to report it upstream too.)


By Kevin Kofler at Sun, 2008/03/30 - 5:00am

I was talking about the wait for 4.0.0, and how KDEPIM was pushed back to 4.1.


By Jonathan Thomas at Sun, 2008/03/30 - 5:00am

Hello,

except for base, games and edu, no applications were supposed to be ported for the release of 4.0.0 if I am not all wrong.

That release was to get kdelibs in a perfect state and to integrate more techs in the framework that apps would then start using.

I don't recall KDEPIM being delayed. Actually KDEPIM more or less has had an independent release schedule and was focused on stable releases with incremental feature additions, wasn't it?

I don't mind anything being pushed to later. I do mind that Plasma was not pushed back and as a consequence was pushing everything else back.

Yours,
Kay


By Debian User at Sun, 2008/03/30 - 5:00am

So you say KDE 4.0 should have been released without a desktop?
(to clarify, without plasma = without desktop as porting kicker & kdesktop would realistically not have been possible - just read and think about it)


By jos poortvliet at Wed, 2008/04/02 - 5:00am

Not a bit. The problem then would be that we'd have a "stable" KDE that would be continually breaking its API every few months. You think that's better? It's certainly not a recipe for quality & stability.


By T at Sat, 2008/03/29 - 5:00am

sometimes i wish people would learn to lurk. just hang out and read, and then read some more, then read a bit more.. before they start posting their 'opinions'. that or i begin to wish sites would stop providing the 'user feedback' option. given, user feedback is a driving force that makes opensource software meet users needs better in ways 'corporate' goals don't always address. too many only understand how to complain without understanding the importance of constructive criticism. sometimes it seems people think developers of free software should knock on their door, ask how they can make FREE software that works for them, and ignore how others may want to use the same software.

in aaron's keynote speech (at google) he did an excellent job outlining the benifits and potential of the various kde4 frameworks. there is a huge amount to be excited about. but people are not patient. and software development requires patience. many of the replies to this post say people either failed to read, listen, or understand what kde4 is about. and the potential it holds. if a particular freely provided software doesn't make you happy... just go somewhere else.. unless you can add something useful. but it seems interpretation of what something useful actually is fails to gain understanding.

transition periods are difficult and i can only hope aaron and others can ignore comments such as those here during this transition period. from what i've understood of kde4 and its future, the kde developers are trying their best to provide the best solutions they can. i think they are doing a wonderful job, and i applaud them.


By tooth at Sat, 2008/03/29 - 5:00am

To those who complain:

What has happened so far is:
1. Make huge under-the-hood changes
2. Release a version to put out your API and get testing by early adopters
3. (in progress) Fix bugs, add missing features

So step 3 is going to be continuing throughout the rest of the year. Why complain about it? Who is forcing you to use it? You already have a great, free, product to use, why complain about another product that is also given freely and will someday be truly great? And while the developers get their work done, it's not like they're blocking your driveway, or spewing toxic fumes into your living room. In fact, their work should not be impacting you in any negative way whatsoever, since you're free to use whatever software works well for you.

Also, why complain about the enthusiasm & evangelism the developers show for their work? I think it's wonderful. Free software progresses by attracting developers who care for a product, and without promoting the advantages of the framework that has been developed, it is hard to imagine how others will learn enough about it to become intrigued and join the project.

Bottom line: if you don't like it, just wait until it is better. Nothing more is being required of you, and you'll save a lot of saliva by not foaming at the mouth all the time. If you are a patient and constructive person, then you can contribute by providing constructive feedback that acknowledges the value of what is being so freely given.


By T at Sat, 2008/03/29 - 5:00am

I cannot disagree with you and I'm really sorry for my complains.

However I've always wanted Linux to become a real alternative to Windows and it just cannot happen if we don't have a set of stable APIs.

ISV just cannot spend their time and resources compiling their applications even for five most popular Linux distros. They will not spend their resources because Linux distros and API's are maturing at such a speed, so that your *binary* application is not guaranteed to run even in a one year term.

And closed source software is here to stay for at least the next 25 years.


By Artem S. Tashkinov at Sat, 2008/03/29 - 5:00am

For a couple of years now there is the LSB and basically all distributions do have support for it or are even certified.

The LSB is also moving at a slower pace and additionally has a long deprecation period.

However you don't see ISVs actually using it.


By Kevin Krammer at Sat, 2008/03/29 - 5:00am

LSB was nice ... in theory.

Alas, in reality LSB is more of a moving target. The target which is moving at a very fast speed. New LSB releases appear almost every year - that being said we are again at the point of unstable APIs and developers who shun writing their applications for Linux.


By Artem S. Tashkinov at Sun, 2008/03/30 - 5:00am

It doesn't matter if there are new releases of the LSB, because they add things and at most deprecate older ones, thus staying backwards compatible.

Deploying on more than one platform is alway more work than deploying on just one, but developing for that additional platform, or as you put it "...writing their applications for Linux..." is not.

Have been doing this for years and I really can't hear the whining anymore, usually from people who made an unfortunate choice regarding and are now locked into a single platform or would have to resort to "porting".


By Kevin Krammer at Sun, 2008/03/30 - 5:00am

Even though theoretically most newer Linux distros comply with the latest revision of LSB, the fact is that Linux distros have:

1) Different versions of libraries (and applications are not guaranteed to work flawlessly with different versions of libraries)
2) Different sets of file and directory structures (I mean menu files hierarchy, system settings, etc.)
3) Different packaging systems (for now we have at least four: RPM, DEB, TGZ/SlackWare, Sources/emerge/Gentoo). Even RPM implementation greatly differs amongst RedHat, Mandriva and Suse - e.g. the same libraries have different names.

Take a note that a very rich VMWare Co. spends a considerable amount of time just testing how their products work with different versions of Linux distros. Most shareware writers/programmers cannot afford such expenses ever. Thus most shareware and freeware utilities (some of them are very useful and necessary) will never reach the world of Linux/OpenSource.

And we (open source users) are thrown ourselves at mercy of those who have a spare time and desire to write something (which may not even work as expected) ... While Windows users enjoy thousands of high quality flourishing software releases because its writes get *paid* and they will not break their heads trying to find out why their piece of code doesn't work in distro X.

Without stable API Linux will be at the bottom of Desktop for a very very long time.

Have you ever considered why Wine has become so popular and attracts so much attention recently? Because Wine has a power to preserve mostly stable Windows API even in the light of lacklustre Vista reception and possible Microsoft demise.


By Artem S. Tashkinov at Sun, 2008/03/30 - 5:00am

> Different versions of libraries (and applications are not guaranteed to work flawlessly with different versions of libraries)

If this is about libraries not covered by the LSB then it is not likely that a third party vendor will be using the installed one but rather bundle a specific one with their product, just like they do on other platforms.

If this is about libraries covered by the LSB, can you given an example? Libraries in LSB are specified at symbol level and thus even newer versions should at least provide everything an application expects them to provide.

> Different sets of file and directory structures (I mean menu files hierarchy, system settings, etc.)

I don't know of any desktop environment not using the freedesktop.org menu specification. As far as system settings are concerned, most applications do not anything to do with them anyway. But I am open for a list of applications that do.

> Different packaging systems (for now we have at least four...

Not from the point of view of an LSB application and traditional software vendors are more likely to use an installer anyway and there are plenty of those, proprietary and free ones.

> Take a note that a very rich VMWare Co. spends a considerable amount of time just testing how their products work with different versions of Linux distros.

Yes and they are one of very few applications which needs to integrate at kernel level. Most desktop applications don't.

> Without stable API Linux will be at the bottom of Desktop for a very very long time.

Ah, nice try. Both GNOME and KDE APIs as well as any API they are building on have been stable for years.

Besides, ISVs do follow changes on other platforms as well, moving to the lastest and greatest of the platform APIs, but somehow do this without complaining.


By Kevin Krammer at Sun, 2008/03/30 - 5:00am

"For a couple of years now there is the LSB and basically all distributions do have support for it or are even certified."

The LSB, and projects like Portland, are failures. Nobody is using it.


By segedunum at Sun, 2008/03/30 - 5:00am

I wouldn't say the Portland project is a failure. They have produced xdg-utils, which is really useful, and many packages in Fedora which aren't part of a specific desktop have been configured or patched to use xdg-open to open their files. That way, they will follow the settings of the desktop of your choice. Given that the previous preference was gnome-open, this is a particular improvement for us KDE users.


By Kevin Kofler at Sun, 2008/03/30 - 5:00am

Alas, Portland project is irrelevant to this topic - though I highly appreciate people's attempts for unifying. I'd even say that such projects can lead some day to a bright future where we have slightly different Linux distros with very similar cores (libraries, configuration, etc.) where they only differ in their artwork, installers and booting process.

I was talking about 100% binary compatibility between distros and it's nowhere to be seen.


By Artem S. Tashkinov at Sun, 2008/03/30 - 5:00am

"I was talking about 100% binary compatibility between distros and it's nowhere to be seen."

I'm afraid you've been proved wrong there. Conclusively. Installing between distros is a problem. Binary compatibility? Nope.


By segedunum at Sun, 2008/03/30 - 5:00am

> Installing between distros is a problem.

It's not. You can rpm2cpio rpm package, untar deb package, `tar -zxf TGZ package` and simply copy it into your file system directly.

> Binary compatibility? Nope.

Really, nope. Applications compiled for Qt 4.3.x may or may not run with Qt 4.4.x, the same goes for GTK :-)

I've seen that, I'm not imagining.


By Artem S. Tashkinov at Sun, 2008/03/30 - 5:00am

As far as I can download a Mysql rpm for RedHat, use alien to convert it to deb and install it without a flaw, LSB works


By Vide at Mon, 2008/03/31 - 5:00am

> And closed source software is here to stay for at least the next 25 years.

This is nonsense. Have you tried to run 16-bit-Win-Apps on Vista? Have you tried to use your MS-Office-97-Macros in Office-2007? (Have you ever tried to use it on an "i18n'ed" Office?

Have much fun...


By Andreas at Sat, 2008/03/29 - 5:00am

> This is nonsense. Have you tried to run 16-bit-Win-Apps on Vista? Have you tried to use your MS-Office-97-Macros in Office-2007?

Office 97 runs just fine in Windows XP SP3 right now. The former was released in 1996, the latter will be released soon.

So, as you can see, you can run 12 years old software just fine. *Most* 12 years old open source software will not even *compile* under newer distros.

Have fun either.


By Artem S. Tashkinov at Sun, 2008/03/30 - 5:00am

"Office 97 runs just fine in Windows XP SP3 right now. The former was released in 1996, the latter will be released soon."

That's not what you were asked. XP was released in 2001 (mentioning SP3 is totally irrelevant - I'm sure you didn't do that on purpose;-)), and Office 97 four or five years earlier. Of course it will run. Getting Office 97 to work on Vista is slightly more problematic:

http://www.google.co.uk/search?hl=en&q=%22office+97%22+%22vista%22&btnG=...

Have you tried running 16-bit Windows applications in Vista? DOSBox provides better support for DOS apps than Windows, and at the rate things are going WINE will provide better backwards support for Windows apps than future versions of Windows. Microsoft have lost control of their own APIs.

"*Most* 12 years old open source software will not even *compile* under newer distros."

Twelve year old open source software doesn't need to compile under newer distros.................because it's open source. There is a far newer version of a piece of open source software available now with more features than what was available twelve years ago. Why would I want to do that?

As for twelve year old binary software, Motif applications still run, I have 8+ year old games that still run on any Linux distro and provided you have KDE 3 libs, KDE 3 apps will run, as will KDE 2 apps if you have KDE 2 libs installed. As demand increases those libs will continue to be shipped, and see better integration with older applications inheriting new features. On top of that, the ABI stability of our favourite compiler has got infinitely better over the years.

You painted right over the real problem with Linux desktops and distros today, and that is general software installation, but as for backwards compatibility, things look pretty good indeed.


By segedunum at Sun, 2008/03/30 - 5:00am

> Have you tried running 16-bit Windows applications in Vista?

Why are you so possessed with running 16 bit applications under Vista? After all there's a free DosBox which perfectly runs even under 64bit Vista.

> Twelve year old open source software doesn't need to compile under newer distros

Why are you so certain about that?! There are plenty useful old Linux applications which do not run in newer Linux distros. And they don't compile either. And you have decided *for me* that I don't need them.

What I say is that in Windows (even in Vista) you ***can*** run 12 years 32bit old useful applications and utilities (e.g. screensavers). And programmers who wrote them have changed their profession a long time ago, thus leaving you with *no support* but with *working* applications.

And Wine's 16 bit windows API is *very* incomplete - you can believe me.

And still Open Source applications work perfectly in Windows.


By Artem S. Tashkinov at Sun, 2008/03/30 - 5:00am

"Why are you so possessed with running 16 bit applications under Vista? After all there's a free DosBox which perfectly runs even under 64bit Vista."

Because that's what you were asked. Will Vista run these older Windows and DOS applications out of the box without having to resort to projects like DOSBox and WINE? No. You're spinning I'm afraid.

"Why are you so certain about that?! There are plenty useful old Linux applications which do not run in newer Linux distros. And they don't compile either. And you have decided *for me* that I don't need them."

You're talking about twelve year old *open source software*. Read that again. Versions move on and latest versions are shipped with distros. I'd be pretty surprised if you could get a lot of Win95 code to compile and run perfectly under XP and Vista (much has changed that requires umpteen compatibility settings). On a Linux system, all you need are the relevant libraries and headers in most cases, and as I'd pointed out, Motif applications will still run as will 8+ year old games - on any distro. You don't have a point here.

"What I say is that in Windows (even in Vista) you ***can*** run 12 years 32bit old useful applications and utilities (e.g. screensavers).........thus leaving you with *no support* but with *working* applications."

The people trying to get Office 97 to run properly wouldn't agree with you, and the standard response is that it isn't supported now. WINE provides better support for Office 97 than Vista does.

"And Wine's 16 bit windows API is *very* incomplete - you can believe me."

Vista and XP's support is almost non-existant, and you'll be very lucky to get a lot of apps working. WINE's support for such apps, and improvements to make them work, is in better shape and is still ongoing.

"And still Open Source applications work perfectly in Windows."

Errrrrrrr, yer, because the source code is there. Even then, there are still a lot of unsupported features and software when moving from Win9x -> NT for a great deal of open source projects. A lot of binaries you might have had running on Win9x won't run on NT, and support is generally dropped.


By segedunum at Sun, 2008/03/30 - 5:00am

Ever heard of static linking? That's how most closed source software work, even on windows.


By patcito at Sun, 2008/03/30 - 5:00am

I have direct experience of moving a bunch of 2001-era programs running on Windows 98 SE to Vista.

Outlook 2000 *does not work* on Vista. Even with address book hacks (Microsoft changed the system address book completely) it crashes. Fireworks 4 and Dreamweaver 4 do not work properly on Vista. All three of these programs crash upon exit. I won't even mention the unsupported hardware and broken drivers.

ISVs will not spend time and resources making their old application versions work on new Windows. They may be willing to sell you a newer version. Gee, thanks.

I hate closed-source software and am doing my best not to buy any more. (Though when getting it for free... it's harder to say no.) And I really regret getting new hardware preloaded with Vista; XP would have worked much better and a Linux distro + WINE would probably have about the same level of incompatibility with 7-year old software that Vista has.

I realize this doesn't contradict your precise points, but Windows stability is less rosy than you portray. After two or three releases, you're compelled to buy new software, often from new vendors.


By skierpage at Mon, 2008/03/31 - 5:00am

While i agree on toning down discussion i cannot understand why you developers take your products being criticized and your views thwarted.
Of course you are free to go your way, whatever people say, but that is not the open source community many people have in mind.
Open communities should be able to welcome professionals from different fields and simple users, making every effort to understand their needs and complains, so as to be able to exploit such wide diversity of competences. You appear to go the opposite direction.
Ok, such efforts should be be made by both parts, but i have noticed several times on these pages, you seem to close the door to any confrontations.

--
Pol


By Pol at Sat, 2008/03/29 - 5:00am

> Open communities should be able to welcome professionals from different fields and simple users, making every effort to understand their needs and complains, so as to be able to exploit such wide diversity of competences.

When I work for a client, I listen to his needs and complains, when I work on free software I do it for me only, if you happen to like it, good for you, but don't expect me to listen to your needs.
That's free work I'm doing for fun or for my own needs, "free work", get it? When was the last time you did something for free? When was the last time you did something for free for others? Hacking can be pretty boring when you're not hacking on things that are of interest to you, that is why free software developers only works on things that answers their own needs, otherwise it gets boring. I hope users could understand that one day. Buy or download yourself a C++ book, start learning and programing, you'll soon understand :)

Cheers


By patcito at Sun, 2008/03/30 - 5:00am

While I agree that you're basically right with this statement, you nevertheless should be aware that this attitude will get you absolutely nowhere as soon as you try to adress more than the open source geeks (making up probably about 0,000001% of all computer users).

KDE cannot aim to be a desktop for the masses if, as soon as there is criticism, key players retreat to the "leave me alone or code it yourself" mantra.


By anon at Sun, 2008/03/30 - 5:00am

Patcito, despite being right, is not answering anyone. Of course he can code whatever he wants to the way he wants too. So what?

KDE is much more than a collection software made, for free, by guys who are coding for themselves and don't really care about potential users, and that might just happen to suit other people's needs.


By ad at Sun, 2008/03/30 - 5:00am

> KDE is much more than a collection software made, for free, by guys who are coding for themselves and don't really care about potential users, and that might just happen to suit other people's needs.

Very few people are being paid to work on KDE, amarok, kate, digikam, konversation to name a few, most plasma developpers etc...
So yes, most people working on KDE do it for free and for their own needs, it just happens that KDE attract a lot of people and so many needs and wish gets answers and solutions. Sometime though, no KDE devs have interest/fun/needs in developping something so it just doesn't get done. The good thing though is that as this is Free Software so you can always hack it yourself OR pay someone to hack it for you OR you can use something else. I couldn't care less what you use, use whatever you want and if you don't like it, fix it, pay somebody to fix it or switch to something else, as simple as that.


By patcito at Sun, 2008/03/30 - 5:00am

"So yes, most people working on KDE do it for free"

Sure.

"and for their own needs"

Bullshit.

"I couldn't care less what you use, use whatever you want and if you don't like it, fix it, pay somebody to fix it or switch to something else, as simple as that."

That's fine. Just not the attitude of most kde devs though.
And it's actually disrespectful towards many of them to say it is.


By ad at Sun, 2008/03/30 - 5:00am

Certainly they work on things that they find interesting but they also answer feature requests from users. If they only did it for themselves they wouldn't bother would they. I imagine the kde devs get a lot of satisfaction from people using something they have created just like someone else.

Therefore having an open attitude helps everyone. Telling someone to go away, I don't want to do it or generally being curt helps nobody. There just wouldn't be a community, nor would this site be here if KDE devs didnt care about users


By Fool at Mon, 2008/03/31 - 5:00am

I still maintain that if a feature present no fun/interest/need for the developer, it won't be implemented unless he's paid for it. It's not about telling the user to go away, just "sorry but I have no time/interest adding your feature, maybe some day, somebody will do it". Sorry if you believe that not wanting to implement your pet feature is "being curt", that's not very accurate and fair.


By sloppy bobby at Mon, 2008/03/31 - 5:00am

I don't imagine their are many shills posting to dot.kde.org so the negative response by some here is surprising.

If you don't like KDE the code is there for you to change it.


By Tom Callway at Sun, 2008/03/30 - 5:00am

Slightly off topic: Can't wait for Aaron Seigos KDE 4.1 keynote.

I think we can make KDE 4.1 seem even cooler than Steve Jobs' Reality distortion field. At least ours is real. :)

Please let us know if you need any donations for video equipment or a conference hall to shoot a KDE 4.1 keynote.
Your last keynote is why KDE 4 has been in the minds of so many college students and professors.
Please also make it cool like the last keynote.

Don't forget to include coverage of AMAROK 2.0. It will draw people to KDE, the same way iTunes and the iPod drew people to Mac OS-X. :)


By Max at Mon, 2008/03/31 - 5:00am

Until some key long standing issues in KMail are addressed ( some of which are 4+ years old in bugzilla ), Kontact is not going to be any kind of competitive threat to Outlook.

I am talking about bugs like 86463, which is 4 years old now and has over 700 votes - I know developers want to work on what is fun, and I understand that, but things like preserving HTML formatting are so absolutely critical to any modern organization I don't see how people expect KMail to be taken seriously.

I mean, I work in a hybrid office environment. 1/2 the developers use Linux, 1/2 use Windows - so we are definitely not a windows-centric organization, even though we run Exchange. And yet, I don't know of a single person who uses KMail, because it is simply not up to the task. Even I am forced to run Thunderbird in my KDE desktop.


By Jason Keirstead at Tue, 2008/04/01 - 5:00am

Pages