14 January, 2008

Sweet Follows Sour

I think it's really necessary to respond to some criticism seen on the reactions to the latest OSnews article.

I won't go into the article itself, imho it's rather negative, but hey. From an user's perspective, it makes sense to only review 3 or 4 parts of KDE 4 and complain about them, and ignore all the other brilliant pieces of work in there, right?

On to the responses, I found this reaction by dagw to be the most typical.

Let me quote the most important part:

The KDE team obviously shot themselves in the foot with calling it 4.0. I'm sure they had a reason for not calling it Beta or Developer release, but whatever the reason it was a bad one. Especially since every complaint is met with a response of "well what did you expect, it's a Beta software". No matter which way I look at it, the KDE team screwed up this release, and it would probably be in their best interest to admit it and just flat out say, we jumped the gun.


Well. That's painful. So, is he right? Did we make the wrong decision? Let's look at it from a broader perspective for a while. Let's see it in the Grand Scheme of Things to Come.

The big question that should come up is: couldn't we have released what will now be KDE 4.1 as KDE 4.0?

No. Seriously, no. If you think that, I see why you would agree with what dagw said. But it's wrong, for many reasons.

One of those reasons can be summarized as 'community dynamics'. You need to get people into release mode, and we wouldn't have been at KDE 4.0.0 stage right now if we wouldn't have committed to releasing it. Many users will start using KDE 4.0.0 and start reporting bugs, so many corner issues the developers themselves would've NEVER found will be fixed in 4.1 - those would have been there if 4.1 would be our first release. Sure, the current 4.0 won't be picked up by as many ppl as the '4.1-4.0' would have been - but by more than if we would have released another alpha or beta.

A second issue is packaging. KDE 4.0 is relatively hard to package, not due to it being that difficult - packaging it is far easier and faster than KDE 3.x. But it is new, and new always needs adjusting to. CMake, SVN, many new dependencies, many new architectural pieces, changes in the internal structure of the major KDE packages like KDElibs and KDEbase. It'll take a while to get used to those. We probably can't expect distro's to put out KDE 3.5.x quality packages for at least a few months. By the time 4.1 is released, though, they will have some experience, and get it done rather quickly. (if you don't believe me - just check out a few different KDE 4.0 distributions... They differ wildly in terms of stability, features, everything...)

Third, we didn't want to hurt KDE-edu, KDE-graphics, KDE-games and the other parts of KDE that were ready for a release up to a year ago - for an explanation, read my previous blog - Why KDE 4.0 now.

Fourth - underlying issues. Many of the problems in KDE 4.0 can and will be fixed by the KDE hackers (many of them hopefully in KDE 4.0.1 already). But many can't. By pushing the boundaries of technology, you'll be pushed back. We've exposed issues in drivers, architectural issues in X, windowmanagement, Qt, all over the place (if you want to read up about it, aaron seigo has some excellent blogs about it). These simply would've appeared in '4.1-4.0', and would've bit users just as hard as they're biting now.



What I'm trying to say is of course the typical stuff: it is easy to say a decision is wrong if you're standing on the sideline. But the issue is often much more complicated than you think - and indeed, it is. Please, take that into account when you criticize the decision we (as in the KDE community) made. My bold statement is: No good would have come of delaying the release any longer. We would just have delayed progression. Would you want that?

(have a nice day and see you on the other side of the ocean)

30 comments:

  1. You couldn't beat sense into some people.

    ReplyDelete
  2. I don't feel there's any reason to respond to it. If you didn't what would happen? Nothing. Some people don't understand open source and .0 releases and they haven't been around long enough to remember other .0 releases from major community projects. You basically feeding the trolls. Just drop it and move on this will all be forgotten until KDE 5.0 then everyone will forget about the 4.0 releases and talk about how terrible the 5.0 release is.

    ReplyDelete
  3. don't bother about thom and his osnews morons.
    just ignore the small minds.

    they're using gnome and the deserve it

    ReplyDelete
  4. I think it's good that KDE released 4.0. I just think they should've added a couple words to the "4.0". KDE 4.0 could have done what FreeBSD did with 5.0 and released it with "Technology Preview" or whatever they used. It's probably not too late to append something as simple as those two words to references to the release on KDE web sites.

    ReplyDelete
  5. > We probably can't expect distro's to put out KDE 3.5.x quality packages for at least a few months.

    Typo? KDE 4.0.x?

    ReplyDelete
  6. >> We probably can't expect distro's to put out KDE 3.5.x quality packages for at least a few months.

    >Typo? KDE 4.0.x?

    I read it as "packages with the quality of KDE 3.5.x packages".

    ReplyDelete
  7. You are presenting a false dichotomy. There are more options than release what you have as 4.0 and put off the next release until you have 4.1.

    What I believe the author was getting at, and what I believe would have been a much better idea, would have been to release KDE 3.9. It makes it clear it's not yet finished, and still allows all the benefits you claim are only possible with the unfinished 4.0 release.

    I honestly think that this fiasco has severely harmed the KDE project as a whole, and I think the attitude of some developers to blame end-users for expecting 4.0 to be complete is disgusting. I'm no newbie, I've been using KDE since the 90s, and it's seriously made me consider switching to GNOME.

    ReplyDelete
  8. You guys need to stay out of the feedback sections on these sites.
    The KDE and Gnome groupies always troll ( even if this was kde 3.6 and super stable ).
    I do find it amusing that they compare it to commercial software and there releases, Vista sure had issues, so did the latest OS X, they were more stable than 4.0.0 probably but had enough issues to prove that this whole naming issue is BS, you have to release stuff to get eyes to report bugs etc.

    ReplyDelete
  9. The KDE team correctly identifies different groups of involved parties:

    1. KDE core / libraries developers
    2. application developers, consume (1)
    3. distributions & users, consume (2)

    Since (1) is considered to be stable, KDE 4.0.0 was released with the mention that (2) was not top notch yet. IMHO, there's nothing wrong with that at all.

    What is wrong, is that Thom learned the classic trolling trick of stating the exact opposite of what he knows to be the truth, mixed in with half truths and obvious things. That way he can generate a lot of traffic to OSNews.com to get more ad revenues. The next article can then again go about the same thing and the subject can be milked to oblivion.

    Don't feed the trolls, stay away there.

    Matt

    ReplyDelete
  10. No good is really coming out of release it early too. I was really looking forward to a KDE 4.0, but instead I will just wait for KDE 4.1. Distro's will not pickup KDE 4 for their default desktops, people will not install it. You are now in the same boat as you were with a developer preview.

    Playing the revision game is just going to hurt the KDE project and reputation it has built for a stable and complete DE and I am just disapointed as I thing others are as well.

    Another thing is that you must take reviews for what they are worth. Yes you got a bad review from OSNews. The points were 100% valid. This is what happens when you release something as incomplete as KDE4. Live with it.

    ReplyDelete
  11. Whatever the reasons to release I am quite happy to test and use what is essentially a developer prerelease I like helping out, I have used KDE 4 as my default desktop for a number of weeks from SVN. I think there must be others who share this sentiment too.

    I love kde4, be sure to keep up the good work, don't worry about detractors or people who are ignorant just keep doing what you do best.

    ReplyDelete
  12. I think that the PR people have done a good job at explaining that 4.0 is not a release for the average user.
    What I see missing is : "now, the next step".
    What is going to happen till/for 4.1 ?

    As long as the kde project does not inform clearly what are the next steps, milestones, feature plan for 4.1, the average user will probably only see the regressions.
    Jos, is there any plan to provide this type of information soon ?

    ReplyDelete
  13. Good point, Alvise. I'll try to discuss that at the release event.

    ReplyDelete
  14. I'm not sure what is so negative about this article? Sure there is critism, some of it debatable (e.g. his rants about bold fonts), but the issues are there, this is a fact! Blindly ignoring that doesn't help anyone, it just causes frustration. And he also states things he likes and what to expect for the future "the potential is just oozing out of every pixel"!!! This doesn't sound negative at all to me! I'm a true KDE user and fan since the good old days of Beta 3 and I'm more than exited by what you guys are creating for KDE 4, but I fear a have to agree with him. I tried several of the Live CDs and it was usually, well ... buggy. Of course it doens't help that a central piece as the panel is hardly usable... This is not bad, it's the way software development works. And the progress could be seen with every release! But calling it 4.0 was perhaps not the wisest decision. This is not about corner issues, it's about obvious problems and bugs. I *really* don't want to accuse anyone for that, but it's the way it is. As for perfection never being reached: nobody seriously expects a feature completness like KDE 3.5.8!!! But, also given the high expectations, IMHO a bit more time for 4.0.0 to mature would have been advisable, in order to get the basics right and stable. IMHO 4.0.0 is more what at most RC1 should have been... I guess this sounds negative to you as well... Personally I don't care about version numbers, I use it when it's usable. I'm just concerned about the public reaction... Nevertheless I will upgrade my distribution for being able to install and use KDE 4.0!

    As for the stupid comments, ignore them!

    ReplyDelete
  15. What do I consider negative about the article on OSnews: the fact that it ignores so much good stuff in KDE 4.0. I don't deny most criticism in there is quite valid, but hey, you can find bad stuff anywhere. Trying to balance the bad with a bit more good stuff would be great. Of course, that's the authors decision - just as it is mine to call it negative ;-)

    ReplyDelete
  16. As long as you don't charge for KDE why should people care? I go to OSNews for the news, not the e-peen in the articles and I go into their articles, like any blog, as the potential to have an informed opinion.

    I trust KDE has a better finger on the pulse of their users than OSNews's comment section.

    In short, I can't wait until 4.1 but I won't cry over 4.0.

    ReplyDelete
  17. Take it on the chin and move on, you'll only feed the trolls.

    I tried a Kubuntu live CD release on VirtualBox and saw much which impressed me and much which didn't.

    As a XFCE user, I like to keep on top of Desktop Linux as I use both KDE and Gnome applications.

    My decision was a simple one:-

    "It's not for me yet, lets see what happens with the next release"

    It was a bold move to ship 4.0, whether it was the right one is a moot point, as the baby is now out in the wild.

    Beware Trolls!

    ReplyDelete
  18. Although I knew about the .0 issue, I was a little disappointed when I first tried kde 4.0 a few days ago. However, this has changed. A lot of the issues which bothered me were due to packaging problems and they got fixed quite quickly by the kubuntu crew.

    Although there are still some minor issues (some shortcuts don't always work, assigning a shortcut to the windows-key + somekey doesn't work, rendering issues etc.), I really like 4.0 know. I think I'll even switch to it for my daily usage. Of course I still need to use a lot of KDE 3 programs as there are no equivalent KDE 4 counterparts but that's just a matter of time. So I'll stay with KDE 4 and hope that 4.0.1 gets out soon with a lot of bugfixes and maybe even some little improvements (Please let us change the taskbar size again).

    ReplyDelete
  19. OSNews has been very quick to point out problems with Linux. They seem to prefer Windows a lot more. This is quite obvious from their selection of articles and the summaries. Vista was a major disappointment and leaves Windows far behind OS X. Windows supporters realize that KDE4 is the next major competition and more likely to win over the vast majority once it gets going. Thus the various attempts to take the wind out of KDE4's sails by raising non-existent issues and claiming a major problem where none exists.

    ReplyDelete
  20. I don't think that denying facts does any good to KDE. This was a blunder, maybe not in a technical way, but still. You could have called this release KDE 4.0 - Platform, or whatever, and told everyone that it will be followed by a KDE 4.0 - Applications soon. That is much easier to understand for eveyone than accepting that "4.0" is really 4.0alpha, and the "real" 4.0 is 4.1... Because this is not what people expect from a .0 release!

    And no stupid "they like Windows!" or "they are using Gnome!" bullshit (because it's not even true) makes this go away. And similarly, yes, he stressed the bad points, but that's what happens if someone is disappointed; because he expected more.

    And let's face it, things like plasmoids cannot be placed on the panel are posing the question if even the core is ready. But that's just a technical detail.

    So for the time I'll stick to 3.5.8, because I just love what you did to KDE 3, and it just blows everything else out of the water. And this is why I am also disappointed in 4.0.

    ReplyDelete
  21. Well, I like the new KDE. I don't remember such a big innovation being done for the past 5 years. This is a big step ahead, and surely, there are some issues, but that's why a new releases are forthcoming :)
    I can't think of any other flexible technology available on the world, and you cannot have all in one. Some things you might consider as bugs are actually features. So don't troll, see if you can help with relevant reports or digging for critical bugs.

    ReplyDelete
  22. Jos, good comment. I like KDE 4, and I think it's great. However, allow me to nitpick:

    "A second issue is packaging. KDE 4.0 is relatively hard to package, not due to it being that difficult - packaging it is far easier and faster than KDE 3.x."

    Packaging is made unnecessarily harder because of the kdesupport dependencies. These are very hard to package. There are packages such as qimageblitz where the only way of getting them in source form is through a SVN checkout.

    It would have been great if KDE released a "kdesupport" source tarball with source packages of that software with versions guaranteed to compile with KDE 4. For now, one must dig in the bowels of the KDE SVN tree...

    ReplyDelete
  23. I've been tying to get a username all morning at OS News.. maybe when they get their website fixed they can complain about a desktop environment's panel not being adjustable in size eh?

    Frankly, KDE 4.0.0 is already my main desktop. I'm still pretty new to Linux and Open Source. I don't get what the fuss is about. I've PURCHASED software that was MUCH less functional, and not configurable at all...

    It's not perfect... big deal? I really don't get what all the fuss is about. The KDE team never said it would be. In fact if anything their comments led me to have much lower expectations than I should have had. KDE4 impresses me already.

    Menu.. drag the traditional one from the menu to the panel - fixed.

    As for the panel not being configurable.. oops.. that's not a shiny spot for the release. But guess what, my daily computing has been working just fine even though I'm not able to drag the panel up and down my screen.

    So I can't visit kde-look.org and go config wild... to be honest this is the first version of KDE I've seen that there's not really that much that needs to be changed. It's really pretty. Even the big panel is growing on me. (menu -- got to go!)

    The thing is stable and the technology works, and the whole is usable by a normal user that means .0 release to me.

    I think some people have been spending too much time watching those clever I'm A Mac commercials on TV and forget how Software works in the real world.

    Kev

    ReplyDelete
  24. Forget trolls, and anti-trolls, and people that see OSNews as the enemy. As someone else pointed out, the issues are there. I do not care and neither should anyone with a sane mind, about whether this should be 4.0.0, 4.1, 3.9 or Delta Prerelease or whatever the name. It's supposed to be the next KDE. The better KDE. It doesn't matter whether it crashes or whether there is tearing or if compositing sucks, that's bugs to be fixed and those are understandable. What is not understandable though is the design of things - why replace kcontrol with it's many and useful options (I've seen users complain about few hacking options for window positions and scrollbar opaqueness ) - why have a panel that cant be done anything with ? no right click , no widget docking nothing. These are not bugs. These are design flaws. Not implemented yet you say. Okay, are you waiting for user feedback to pick up the idea that people want to use something the same way they have been using it for over a decade ? I dont think so. Yes, you started from scratch, yes implemented tons of new libs and APIs. Ok that's fine. Why rush then ? Do not hide behind your fingers, because kde4 had been largely behind schedule, and release dates being pushed back and back. When you decided to release it, you only did that because you felt worried people would criticize the OSS development model, KDE, linux, and maybe even get the designation of vaporware, and then what ? So, off you went, and botched together whatever was ready at the time, made sure it compiles and doesnt catastrophically die (as in segfault) - and call it a release. It is half empty, not half full.

    ReplyDelete
  25. When Microsoft releases a product before it's ready, they're criticized for not properly testing the product.

    When OSS releases a product before it's ready, it's fucking progress.

    Get your head out of your asses.

    ReplyDelete
  26. Hi KDE-Developers,

    don't be too touched by the OSNews-Article; Thom Holwerda is far from being authoritative in any way - sometimes he tends to be a bit narrow and opinionated. What the KDE-team did with the 4.0 Release was completely correct, just go right on.

    ... Old Arab Proverb: "The dogs are barking, but the caravan continues on its path."

    ReplyDelete
  27. Hi all,
    I'm just a pure user, not new to the floss world, or KDE, by any means, but just a user --I sometimes give a little money here or there... my entire contribution.

    I'm only commenting because I want you to know that I downloaded the kubuntu-kde4.0-i386 'live' CD and tried it.

    I'd like to put the next part in all caps --I won't yell --but I'd like to:

    It is a fantastic release!

    And I couldn't even figure out how to get it to deal with my statically assigned router settings... which means zero about much, except that I'll eventually figure it out, or someone will fix it, or whatever... I'm just trying to say, 'less then perfect' well yes... And so what?

    I always thought that any X.0 release just means something like; 'Here we start a new way'.

    And so far it looks to be a really grand 'new way', minor problems not withstanding.

    As a plain 100% user I can clearly see where this will become very special.

    So there you have it from what is probably the only view that really matters about usability --someone that simply uses KDE, day in and day out, and has for years.

    Yes, it's not yet got all the polish of the 3.5.x releases, but it none-the-less, shines much brighter.

    Good show, and thanks, to each and every one of you.

    ReplyDelete
  28. Release naming: spiral of evil

    I think that KDE 4 is a great project and I really appreciate the technology progress under the hood (I'm not a KDE developer, but it seems to be promising).

    But the approach "if we do not release 4.0.0 nobody would test it" seems to me like a spiral of evil.

    Why do you thing people are not testing betas by using it? Probably they are not ready for daily use. Probably packagers from users' favorite distributions came to conclusion that it is not usable yet and did not prepare (not even) testing packages.

    What is better? Release clearly marked as testing (Beta) or .0 release surrounded by rumors that it's not yet ready? A couple of my friends said: "I've already installed KDE 4.0.0, but I do not use it. It's not for daily use." This is the reason to release .0? To encourage people to install it and then left it unused.

    I think people will start to use it when it is ready. No matter what is it called . Surely, there's a momentum. When people are used to the fact that betas are not usable they won't use it until some courages geeks say "Hey, I've tested the latest beta and it's pretty stable! Do not afraid to install it."

    And if people find out, that .0.1 release is usually not OK, they will wait until .0.3 or .1.1 ... this is the spiral.

    This can be clearly observed when you look at what versions are coming to stable branches of distributions. There are .0.2 releases, .1.6 releases, but there are also some RCs and betas, 0.1 releases, and so on. Users and their Linux (or other OS) distributions will decide, what is usable and what is stable. Software developers can just make it easier by marking their releases clearly.

    Yes, "release early and release often", but mark your release clearly.

    ReplyDelete
  29. I think that all explanations, justifications, rationales, ... don't bear one simple thing in mind: Many users had high expectations for KDE 4.0 and are now disappointed.

    Even if you're right with everything you say: It's from a psychological point of view not a good idea to explain someone who is disappointed that it's his own fault. Most of the users know and understand the reasons, but that doesn't change the fact that they are still disappointed.

    Understand that they're sad with the status of 4.0 and give them hope that it'll get soon better. And admit that this is also not the KDE 4.0 that developers hoped to see, as some of the explanations for the release of 4.0 sound in the meanwhile.

    ReplyDelete
  30. Is it so hard to admit you were wrong ?

    ReplyDelete

Say something smart and be polite please!