Ars reviews KDE 4.0.0: “KDE 4.0 was officially released last week after extensive development. The long-awaited 4.0 release ushers in a new era for the popular open-source desktop environment and adds many intriguing new features and technologies. Unfortunately, the release comes with almost as many new bugs as it does features, and there is much work to be done before it sparkles like the 3.5.x series.” They were also at the KDE 4.0 release event.
Good article. Points out again that there is a crucial difference between “KDE 4” and KDE 4.0.0, and that what we have so far is an unfinished development release to stimulate interest in the whole KDE 4 cycle. I really hope this puts to rest the naysaying and wilful misunderstandings. Lack of documentation for third-party developers sounds a bit of a worry but no doubt that will improve. And, yes, there are lots of bugs, but then we all knew that, surely.
Just my 2 cents as a user, but what we have here looks as if it will go on to make a superb platform. It’s badly needed, too. Nothing is static in this game, and if FOSS doesn’t refresh itself from time to time the competition will do it for them and we’ll all be the poorer (literally, too).
reading the Ars comments, there still seems to be plenty of willful misunderstanding going on.
Meh, in the end some people may be motivated to do something useful and the rest will have moved onto to the next topic de jure to whinge about, usually at a volume reversely proportional to any contributions they make.
The controversy over the maturity of KDE 4.0.0 challenges the dynamics of the free software community. It’s an expectations gap caused by a cultural shift within the community that deserves some consideration.
The “enthusiast” class has become far less tolerant of immature code than it has been in the past. The same type of user that used to be willing to jump through some hoops and shake out some bugs in order to try out the latest and greatest now expects nothing less than a finished product.
That’s why every thread on this issue devolves into a discussion of numbering and naming conventions. Enthusiasts no longer have any patience for rough edges, they don’t want to touch it until it’s “done”, and they don’t want to hear any excuses from the developers.
I worry that our community of peers is decaying into a mixture of producers and consumers. Enthusiasts don’t want to see themselves as partners with the developers, and the developers don’t want to see themselves as service providers to the enthusiasts.
All software, both free and proprietary, goes through a continuous maturation process. The free software ecosystem was built on the assumption that, from the perspective of any given project, the community is layered like an onion. The core developers are in the middle, followed by project contributors, distributors, third-party developers, enthusiasts, and end-users.
The theory is that, as the software matures, it can be exposed to more layers of the onion. But in the case of KDE 4.0.0, one or more of the follow happened: 1) it got exposed to enthusiasts before it was ready, 2) enthusiasts had unreasonable expectations, or 3) the project failed to make the distinction between enthusiasts and end-users.
What is in a name? Could the expectations gap have been addressed simply by selecting more conservative names for KDE 4.0.0 and the releases that preceded it, or is the problem deeper than that?
Is it still possible to release software with rough edges and expect a productive response from the enthusiast community? Do enthusiasts understand and accept their role within the community? Do enthusiasts and end-users self-identify as such, and are they comfortable with those identities?
Just like an economy collapses for lack of a strong middle class, the free software community will collapse for lack of a vibrant enthusiast class with enough skills and patience to help steward fresh code to maturity. If developers begin to doubt that there is a middle class in the free software community willing to contribute some effort toward a better future, then they will be reluctant to ever challenge “the big friggin’ wall”.
For those of you who don’t follow American politics, the major narrative right now is about how “change” happens. Some think it comes from shrewd work behind the scenes, some from a spirited fight against the opposition, and others from rallying people around a compelling vision.
In my view, the engine behind the growth of free software has been the latter, an idealistic and optimistic belief that if we work together in a transparent and inclusive community, then we can turn the conventional wisdom of the software industry on its head and challenge the entrenched interest in the status quo.
The strength of our community and its ability to produce fundamental change (like KDE4) lies in its ability to engage and empower the user in the development and maturation process.
If we reject this role, if we don’t care to see how the sausage is made, if we don’t want to have anything to do with immature software, if we choose to become passive consumers of finished products, then the status quo will prevail, and our hopes and dreams will be subordinated to the conventional wisdom.
The KDE project has articulated a compelling vision for the future of the free software desktop, and they’ve delivered a strong foundation for realizing this potential. It’s impossible to please everybody with a finished product let alone a work in progress. All I ask of anybody who cares about free software is to either get on board or get out of the way.
It doesn’t do any of us any good to criticize a project for reaching for new heights, for daring to set out in a bold new direction. What bothers me the most about KDE4 is that other projects are perceiving this as a referendum on ambitious development. If we reject the premise of substantive change, then we deserve the stagnation that might follow.
The problem isn’t that enthusiasts don’t want to test. I test every time I have a test machine available (I currently do not because I currently only have one production computer and no spare ones available for testing).
What many of us have been arguing here is that KDE 4.0 is improperly tagged and is actually only at the Beta 1 stage of development with regard to Plasma and KWin. I believe that most of us understand that most of the underlying foundations are feature complete and have stable API’s and ABI’s in most cases (some frameworks and many applications will not be released until KDE 4.1). However KDE 4.0 is not just a set of frameworks for a desktop environment but also the GUI (in this case Plasma and KWin) for those applications developed by that framework and other frameworks. It is the instability of KWin and Plasma that is the problem not the whole of KDE 4.0. I’ve only played with KDE 4.0 on a live CD but it proved to be unstable for me, without the effects turned on. Plasma and KWin are not release ready yet they released Beta software with a Production-Ready tag and now the KDE release team is getting a lot of heat for that choice and rightfully so in many people’s opinions.
I have nothing but respect for the KDE development team but I feel that they made a mistake in tagging the KDE4 series with a production ready tag at this time. This release has more than a few rough edges. It is barely usable as a beta much less as a production system. What KDE released was a Developer Preview as far as I’m concerned and it should have been tagged as such. This has nothing to do with the fact that it is KDE that did this because I would be just as critical of GNOME, Apple, or Microsoft. I feel like Plasma and Kwin just needed another 2-3 months worth of work and then they could have been appropriately labeled KDE 4.0. I believe that KDE 4.0 will become a usable desktop before KDE 4.1 but they have tarnished their reputation for high quality releases with this release. I think a lot of the problem is that an organization that normally would not do this has and it worries people that they will now be releasing shoddier software in the KDE4 series than we have seen from KDE in the past.
I also don’t think that they would have been in danger of entering an E17 style of constant beta development by tagging this by what should be its proper name in an ideal world and that is KDE 4.0 DR1 (Developer Preview 1). Most of us are complaining about the fact that we think those numbers actually do and should mean something. We understand the argument that the KDE release team is making and we think it is a bad argument regardless of who it would have come from. We are sharing our ideals with others who don’t agree and that is fine but we are not going to be silenced just because it is inconvenient for certain people. A lot of people feel like they have committed a form of fraudulent marketing by tagging KDE 4 as they have done at this time and I feel that the x.0.0 means certain things and I further feel that KDE 4.0.0 does not meet this standard that I (and apparently many others) use for all software releases. I feel that this fraudulent marketing will be ultimately detrimental to the KDE team’s efforts. I prefer KDE over GNOME and Windows and maybe it’s just others and my fears that KDE is lowering their standards. I feel like KDE has always had higher standards than other projects, wrongly or rightly, and myself and others fear that those standards are now being lowered.
I will not stop using KDE software because of this but I will not be using KDE4 again until KDE 4.1 is out. I hope that the fallout from this does not hurt KDE 4 in the long run but I’m not going to be silent about what I feel is in effect fraudulent marketing whether the software is FOSS or proprietary, although I guess I feel that FOSS should have higher standards than proprietary software does as this kind of thing is something I would expect from corporations and not free software, IMO.
KDE4 is probably the biggest rewrite in the history of the free software movement. It’s completely unprecedented, and as an enthusiast, you have to adjust your expectations accordingly. KDE isn’t lowing their standards or releasing shoddy software. They’re navigating an extremely difficult transition, and many enthusiasts have not been very understanding.
I think it’s fair to say that KDE made a mistake by tagging 4.0.0. But I take a long view on these things, and I’m deeply troubled by the message that the community is sending to other project leaders. The community is saying we will hold major rewrites to a high standard, and we will rip into projects over any tactical errors that may tarnish the initial release.
You’re not wrong, but your attitude isn’t helping. The conventional wisdom says that rewriting software isn’t worth the trouble. You’re giving more ammunition to the cynics who tell us it can’t be done. KDE took a big risk, and they’re taking a lot of flak over a triviality.
It’s times like this that make it hard to justify working on free software. I don’t make a habit of accusing users of being ungrateful, but I really think that this is a example of how no good deed goes unpunished. This wasn’t easy. People have poured their hearts and minds into this monumental undertaking. Maybe outsiders don’t understand, but I expected better from free software enthusiasts.
I expected our community to accept KDE 4.0.0 for what it is. It’s not mature code, and it wasn’t a flawless release, but what KDE has done should be celebrated and rewarded. We want projects to take big risks. We want projects to rewrite their code when it gets long in the tooth. But the community response has been awfully discouraging, and that’s a real shame.
Well, uhm, yeah. I will treat a free software project in exactly the same way as a proprietary project. They need to deliver quality product, just like their proprietary counterparts, and if they don’t, I will call them on it.
Being a free software project must be quite handy these days. If you do it right, you prove the superiority of the open source model. If you do a major fcuk up, hey, look, don’t hold us responsible, we’re just open source and we do it for F/free. In psychology, the attribution theory has some interesting things to say about this one.
Being open source does not absolve you from responsibility. Especially not an important and big project like KDE or GNOME.
It’s absolutely astonishing how many people just don’t get open source software development, and the way it has always worked and the only way it can work. Let’s put this another way, because you only seem to be able to understand software development in the context of a Vista or OS X release (long paragraph coming up):
If Microsoft or Apple decided to have a much more inclusive development model that included enthusiasts willing to be on the bleeding edge (and they basically admitted that .0 releases are problematic), and they basically created a branch of Vista and OS X where they had fully working, unrestricted and installable .0, odd numbered releases like the Linux kernel or something like that, with explicit warnings that there are no guarantees as to what it does or provides, and where they told OEMs not to install it (just as distributors won’t make KDE 4.0 the default install) – you honestly believe that these people have just cause to criticise a lack of features, criticise bugs and run around like the prophet of doom as to how inadequate it is?! In this phase, if you aren’t going to be a part of the solution then sod off.
You mention the open source model there. What process do you think is going on now with the release of 4.0? Developers, testers and bug hunters are users in the development of open source projects. How do you expect this process to happen? Proprietary developers have already proved the worth of the open source model in how inadequate their .0 releases are, and in the number of patches and service packs available. Those only came about because people tried it and reported back.
That’s the way it has always been. You don’t pay license fees of any description, so you have no comeback as you do with other software (if you have any comeback at all). People started using open source software for a great many purposes exactly because people used it themselves, or people reviewed it, and those people said “Hey, it’s good enough.” For many people it’s only good enough when it appears in the package repository of their distribution, or when it comes pre-installed.
No it doesn’t, and it doesn’t justify silly decisions made within a project just because a developer can, but nobody is saying that is the case here.
You’re tying yourself in knots here, because this is a paradox that will not square. If a project is upfront about developing completely in the open, how do you expect a quality release, that everyone starts using because it’s good enough, to ever come out?
Segedunum, please, for the love of god, stop trying to pull the discussion into a different direction. I’ll put it in caps, maybe that will get it through to you: I AM NOT ARGUING THE USEFULNESS OF THE RELEASE EARLY, RELEASE OFTEN DEVELOPMENT MODEL. Got it? Stop trying to make it seem as if I am!
I fully support this development model (check my articles on Windows 7, for instance), and heck, I enjoy it too. I love how long before it’s even ready for alpha, I’ve been able to play with (and devote time to) the Haiku project, or how I’ve been able to play with E17 for years now, long before it goes gold.
I am not arguing the release early, release often software development model – I am arguing the KDE team’s implementation of this model. The release early, release often model does not mean: release a gold release as soon as possible – no, it means: release early in the development process. There is nothing in the RE/RO model that mandates that you should get a .0.0 release out the door as “early” as possible – it just mandates that the development process should be marked by a steady stream of snapshots, alphas, betas, and release candidates during the development process leading up to a final release (the ‘often’), a stream that should start right at the beginning of the development process (the ‘early’).
Heck, the RE/RO model is not even unique to the open-source world – something that might surprise some here. There is absolutely no reason why proprietary software vendors cannot use this concept either. Again, see BeOS for a software project that also employed the RE/RO model – despite being as proprietary as one can get.
Haiku and E17 seem to be doing just fine. E17 has been in a very stable and mature state for years now – exactly because they properly employed the RE/RO model. They have reached a completely stable and mature state without ever being tempted by a .0 release before it’s done – and it’s a project much smaller in size (on the level of coders, users, and lines of code) and scope than KDE 4. If KDE 4 needs to employ trickery to gain enough development steam, and E17 does not – what does that say about KDE 4?
Edited 2008-01-22 13:18 UTC
I’m not. You just cannot get this conceptually, unfortunately.
OK, I’ll put this in caps so you understand it:
YOU DO NOT GET TO DECIDE WHAT GOES INTO ANY GIVEN RELEASE OF A PIECE OF SOFTWARE UNLESS YOU HAD A HAND IN IT, YOU DO NOT GET TO PLACE YOUR OWN REQUIREMENTS ON THAT PIECE OF SOFTWARE AND DEVELOPERS CAN AND WILL HAVE DIFFERENT REQUIREMENTS FROM YOURS.
The fact that you talk about a ‘gold release’ shows how you are conceptually unable to wrap your head around this. There is no such thing as a gold release, as you would call it. A release is only gold when enough people use and review it and say to everyone else “This is good enough.” This is the way many people have started using open source software for all sorts of purposes – not that something has reached .0. It goes against anybody’s experience of using any given piece of open source software, or indeed, any software.
Where is the development process in an open source project, and when does it occur? Do you see where I’m going with this………….?
Who determines whether this is too early? Not you. The .0 release is decided when they feel they have achieved their aims, goals and requirements at a given checkpoint.
What development process, and where do you think this occurs?
Proprietary vendors can never fully utilise release early and release often, because by definition, they have to get to an acceptable .0 release for their users who are paying license fees.
Remember that the reason why you use RE/RO, and how successful it is, is entirely determined by how much feedback you actually get and how much of a known starting point you can actually work from for each release.
Oh dear. I have no idea why E17 and Haiku haven’t released .0 releases (particularly E17), but have you ever thought that they haven’t yet reached the goals, aims and requirements that they wanted in a .0 release? I mean, E17 by your definition has been stable for years and should have released a .0 release some time ago. You should be criticising them as well.
I will bet you any money in the world that when they do release a point release, an absolute ton of bugs, difficult corner case bugs, bugs that only happen in certain circumstances, feature requests and stuff none of the developers have ever thought of will come flooding in like a tsunami.
You can’t claim that something constantly developed behind closed doors, so-to-speak, without ever having reached a major release, is successful, stable and feature complete. Many people would have thought the CTPs and RCs of Vista were successful, but once you get it out there……
What trickery? KDE 4 has achieved its goals for a .0 release. They may not be your goals, but that’s entirely beside the point, and that’s what you don’t get here.
The development process in a software product is that which starts with the idea on the back of a beermat (so to speak) and which ends with the designation “1.0”, after which development continues, but not in a similar fashion as the process preceding it. The post-1.0 process is bugfixing, refinement, those things, while the process preceding it is getting the feature set implemented. In other words, 1.0 is feature complete, but not necessarily stable. When the project wants to add features, or do a major revision, the process starts over again, moving to “2.0” (which in itself is a “1.0”). You can still add new features for a 1.0.1 release if you want to.
This is a universally accepted pattern of software development, which applies to all matters of software, open or proprietary. It is of course also highly idealised, and many deviate from this norm.
Haiku is an example of a project following this process quite closely. On the back of the beermat, they wrote their ideas and goals down: feature parity with r5, as well as cross compatibility. This is what they have been working on now for a long time, releasing snapshots. They also set more attainable goals, like self-hosting before it goes alpha. Haiku will only be released as r1 as soon as the feature set is similar to that of r5, and compatibility is achieved. Before that, we’ll have alphas (bug ridden, not feature complete), betas (feature complete, but unstable), and release candidates (stable and feature complete, released to find bugs only normal user scenarios can track).
THAT is the commonly accepted model of development, whether YOU try to redefine it or not.
That is what we people do. We set up a paradigm, a reference, we standardise, so that we can easily tell things apart, so that we can see things for what they are. The KDE guys are the ones redefining the concept of a stable, 1.0 release, which is fine – but don’t go crying in the corner when the rest of the world calls you on it for redefining commonly accepted definitions.
Edited 2008-01-22 17:48 UTC
Not everyone agrees with or uses this model.
Edited 2008-01-22 18:03 UTC
Which is great. But then, don’t use the same designations as the model you disagree with. It’s confusing, and the responses here agree with me. A 1.0 release carries with it certain expectations on feature completeness and stability, and the KDE guys cannot fulfill those expectations, and hence, people are disgruntled.
If I develop a scientific model about comparing nations’ cultures (like the Hofstede model), and use the exact same names and designations as Hofstede did for his model, then people are going to be confused, and they will ask me to use different terms. Which makes sense.
Get it now?
Edited 2008-01-22 18:16 UTC
KDE isn’t at 1.0 though and having your version being a X.0 means nothing, imho.
People? Not really. The only ones I see complaining are some “power users” (God I hate that term) and some site editors.
I dont see many complaints from Joe User.
Yes, but I dont agree.
Edited 2008-01-23 03:54 UTC
But Thom, you are arguing against release early, release often. I’m in no way affiliated with KDE, but I did work (on my owns small bit of) a very popular and rather big Free Software project. Release early, release often does mean spitting out a new version for mass consumption, even if it has bugs. Good enough, is good enough.
Being open source doesn’t excuse you for bugs, missing features, or feature regression. What you’re missing in your comparison is the role of distributions in Linux vs Apple or Microsoft. With Linux, the ‘Monolith Product’ is the distro release— Individual projects, if they want to survive AND flourish, must release early and often. (With the only exception being the most mature of utilities. Things like venerable, command line apps that do one very specific thing.)
Projects that don’t release die. It can be a lingering death, but it’s dead non the less. That’s the real value add of having distros making their own stable releases– Even if you disagree with their choices, you get an ‘official call’ on if ‘good enough’ is ‘good’.
You don’t pay license fees of any description, so you have no comeback as you do with other software
Sounds like you’re advocating people never use FOSS for critical apps. Plausible deniability. If there’s never any valid way to criticize the programming, how can I rely on it to work? What is my course of action if I’m unhappy with the product?
It sounds like you’re essentially saying rely on only Microsoft and Apple for mission critical.
No, that’s not what I’m saying, and it’s nice you only quoted one part of that.
However, what I am saying is that, certainly in the open source software world more than any other, many people have started using it for all sorts of serious and sometimes critical purposes because people have grown confident enough that they can start using it. This confidence might come from using it personally, it might come from the fact that your distro now ships a particular version, or it may be that you have a commercial company like Red Hat or Novell backing it all up. That’s how open source software has become as prominent an well used as it has over the past few years.
The point is, the open source software that people use regularly for all these tasks did not miraculously become usable and of good enough quality when a .0 release was made.
You do what everyone has done for years – when it becomes good enough then you start using it. In a nutshell, you stop worrying.
You use the formula that has worked well for open source usage over the years and you wait until it meets your needs, you wait until your distribution ships it, or if you’re still not satisfied, you file bugs and write code. If you’re still not satisfied then you simply stop worrying about it. Judging from the comment response around here many people are worried, so we’re stuck with it.
When someone talks about ‘products’ and ‘mission critical’ I’m usually pretty sure that they’re not going to be able to comprehend this.
This is my last post in this thread, because I’m not much interested in arguing, which is what this has become, and because I’m not much interested in your continued warped perspective. You no longer are diuscussing, you’re lecturing, and it’s boring. It’s clear you’re more interested in a battle than a discusison, so I’ll be ending it here.
The fact is, the “trust” you quote, the very thing that would make someone use a FOSS product, is cashed in once you release something that is not ready and prove that users ought not trust your releases. At that point, the incentive to trust the team not to give you unfinished code is removed.
You might say that the trust I should have is in a distro, not KDE directly, but we both know that is far-fetched, as people regularly download KDE and Gnome, which is why it’s for download on their sites. Distros are not their only customers, and it’s a poor assumption to believe all users know that their current release is not supposed to be used by the general public. Imagine if Microsoft said “Vista is not really ready. Go ahead and buy XP, and your OEM will start distributing it when ready.” That doesn’t mean Vista was ready, but releasing it as is has caused them incredible bad press and cost them loads of customers, many of whom flat out won’t use Vista as a result. Surely, KDE doesn’t want people saying “We have no plan to every upgrade to KDE4.”
All in all, this has cast me as anti-KDE, which I am most decidedly not. I just think that tagging this release 4.0 was a mistake, and that’s all. It’s a 4.0 developer’s preview, or even 3.9.7, but it’s not worthy of the name 4.0, and KDE is paying for that mistake internet-wide. It doesn’t mean KDE4 won’t eventually be awesome, but it does mean that people who were surprised by its incompleteness will not be very likely to try it again until there’s a much wider push from distributions.
I have no idea what that’s supposed to mean, but ya gotta love a load of arrogance to cap off your reply. It makes you so credible! Well done!
Fair enough, but you’re the one who I seem to remember wading into this discussion like a bull in a china shop, telling us how disgraceful this was for a .0 release, they’re deceiving their users, people are naive, there is a clearly defined definition of a .0 release in your head, pathetic excuses etc.
If you can’t discuss what’s been said and can’t argue sensibly in return, I can understand that. You didn’t need to tell me that.
Many open source releases in many projects have not been good enough, by many peoples’ definition of ‘good enough’, for many, many releases. As soon as something becomes good enough, it has the features they want and they can use it you’d be surprised how quickly they forget about previous releases. In this case, the vast majority of users will probably not see KDE 4.0, as their distros will not install it by default.
However, again, you’re putting your definition of ‘not ready’ on it. It really depends on the requirements of the software.
Once you get to something people want with the features they want, it’s amazing how any past releases seem to melt away.
Do they? Not any of the users I know. If I want to check out something then I usually go through the SVN procedure, or install packages in my distro. I simply cannot see the vast majority of users going through even those processes. People going through the trial of downloading or installing from SVN can see what they’re going through.
That’s quite funny, because that’s exactly what has happened in a lot of cases.
People might not use Vista until SP1 or SP2, or even the next version of Windows (many just skipped 2000 straight to XP as well), but trust me, once it stabilises enough people will be using it. Once it does what people want and they run into fewer problems, few will care really.
Upgrading to KDE 4 will be a long process for many and won’t happen for some time yet.
Well, you’re putting your own requirements expectations on KDE 4, and trust me, post 4.1 in .2, .3, .4 and so on, if and when KDE 4 does all the cool things that they’re working on and people look over someone’s shoulder at what they’re using and think “I have to have that!” this mistake you say they’re paying for will melt into oblivion. Seen it happen before, will do again.
That’s how most people will get it. Those checking out KDE 4.0 are checking it out, and those whining about it are doing so here. It really doesn’t mean that much.
Don’t take my word for it. Just look at your comment history to ably demonstrate that you don’t get open source development like this. People always talk about some kind of finished ‘product’, or something that is ‘gold’, or something you can use for ‘mission critical’ stuff, but the fact is that you use it when it becomes good enough. When it becomes good enough, the options, features and general quality is far more diverse and better than a lot of proprietary offerings. It’s the getting there that is difficult to swallow for some people.
I’ve read all of this chain and you really need to calm down, segedunum you’re coming off way too aggressive. Noone said anything was disgraceful. I suggest you calm down and walk away from the computer for 10 minutes.
I think we, the user community, largely do accept 4.0 for what it is. Conversely KDE, and other projects that take bold moves, need to accept that if they release a project with numerous obvious issues the community will call them out on it. The fact that the main way a user interacts with the project is incomplete and unstable is known by both the developers and the early users. It would be hypocritical to praise the underlying architecture when it gives little benefit yet due to the flawed facade.
Sure it has potential, but the community is giving an honest appraisal of the release the KDE did. Segphault was fairly forgiving in his analysis IMHO. He attempted to show the potential while at the same time did not deny the faults. Based on the various articles and attached discussion the response has largely been “cool stuff, let us know when you plan to release something we can use.” It does not need feature parity with 3.x, heck it does not even need to be stable. But right now 4.0 offers nothing more than a frustrating test bed for concepts.
I have high hopes that plasma will get sorted out in the next month or so, but till then you can not fault the community for telling the truth.
I thought the intention was to stimulate more user feedback, rather than to”be celebrated and rewarded”.
Now they have the feedback they asked for, and I think it would be wrong to reject it.
People aren’t critizising KDE4 for trying to reach for new heights or daring to set out in bold new directions. Everything I’ve read by just about everybody hails them for their choices and vision. Nobody is saying that KDE4 should have set their goals lower. Personally I’m thrilled by the potential KDE4 is showing as are most people.
The only thing they are being critised on is how they handled the release of KDE 4.0. That’s it. Stop trying to make the ‘fight’ about something it isn’t.
Hell the only reason most people even bother to critizes is that they care passionatle about how the sausage is made and want to see it made better in the future. Don’t make the mistake of thinking that those who critisize the way the realse of KDE4 was handled are ‘opponents’ of KDE or in any way reject what they are trying to achieve. While I’ve been critical to KDE4, I’m also a huge fan and think it’s the coolest thing that has happened in Open Source software for a long time.
I think you’ve nailed probably the most important point, and one that is consistently drowned out in the whining over the semantics of versioning.
The dynamics of the OSS ecosphere are changing, and not for the better. There seems to be this assumption now that despite being free in every sense of the word, OSS applications should be held to the same standard as proprietary applications.
Critics will counter, “Well why shouldn’t I? I use software to solve a problem, I don’t care if it’s proprietary or OSS”. That deflects from the issue that proprietary software has an development infrastructure funded by license revenue, whereas OSS relies on a development infrastructure funded by the collective investment in time and effort by those that wish to utilize the software.
But we don’t have that any more. Any armchair quarterbacking that this should have been tagged as something different utterly misses the mark that this project was developed transparent, in the open, and with development code available along the way. The devs and distros, particularly openSUSE, went to great lengths to make testing packages available for non-developmental users. Why? For testing, feedback and bug reporting.
And what happened? A flurry of responses along the lines of “I’m not touching it until it’s released officially”. So anyone claiming that releasing it as a “developer release” or “technical preview” or any such nonsense overlooks the fact that they would invariably have been ignored by the userbase at large.
Anybody complaining that the devs somehow let the community down by tagging this release and throwing out the public is sidestepping the issue that they let the devs down by not making an investment of their own precious time to help with the development process.
I doubt anybody that was seriously following the development to this point was shocked or surprised at the “finished” product, because expectations had been set appropriately. Most of the initial negative feedback seemed to be based on nothing at all other than screenshots, and the evolving into a herd mentality of decrying the actions of the KDE team, as if they actually earned a position to complain.
The sad thing is that if the devs had simply left the aging infrastructure, which dates back to 2.0, and instead focused their efforts superficially on wobbly windows and other such shiny baubles, they would have been lauded for making a cutting edge release that would no doubt be yet another Vista-killer(tm) and herald the arrival of the Year-of-the-linux-desktop. I suppose the one benefit is that most of the complainers will happily jump back on the bandwagon now that the foundation is in place and people will start to have more and more shiny baubles for 4.1.
I have no doubts that KDE4 will certainly grow to new heights, and this current tempest in a teacup will blow over and be considered nothing more than short term pain towards long term gain, but the disappointing thing is that maturization of OSS seems to gain a wider userbase without a proportionately increasing base of users willing to commit time and effort to support it. That doesn’t bode well for the future.
Heavy sigh.
For those of you who don’t follow American politics, the major narrative right now is about how “change” happens. Some think it comes from shrewd work behind the scenes, some from a spirited fight against the opposition, and others from rallying people around a compelling vision.
In my view, the engine behind the growth of free software has been the latter, an idealistic and optimistic belief that if we work together in a transparent and inclusive community, then we can turn the conventional wisdom of the software industry on its head and challenge the entrenched interest in the status quo.
The strength of our community and its ability to produce fundamental change (like KDE4) lies in its ability to engage and empower the user in the development and maturation process.
If we reject this role, if we don’t care to see how the sausage is made, if we don’t want to have anything to do with immature software, if we choose to become passive consumers of finished products …
Fine words, Butters. Perhaps you should try politics, too
But I suspect you’re being extremely optimistic: only a relatively tiny percentage of users is ever going to have the inclination and skills to get involved in seeing how the sausage is made. And as a platform grows and draws in more users who aren’t technically inclined, so that percentage will shrink even further. What’s important is that those who are interested and who do all the work aren’t sat on either by a dumb majority clamouring “I want, I demand, the moon is made of cheese and you will cut me a slice” or by the kind of closed-source corporate outlook that turns consumers into passive victims.
What I haven’t heard in this thread is gratitude. Perhaps the unhappy tendency would prefer that KDE didn’t exist or, shock horror, that it was pay-for-only on Windows only? Perhaps they’d prefer not to have any opportunity at all to look inside and test stuff? In the circs, complaining about a name seems pretty perverse.
We are very lucky to have KDE, imho, and to judge from his presentation on that video, KDE is very fortunate to have Aaron S and his colleagues. There are some grounds for saying that desktop Linux plateau’d and began to stagnate a little 2-3 years ago now. There’s been no big break-out in user figures. Certainly much of the growth in non-Windows stuff since then has been taken by Apple, and Microsoft is shaping up for a post-Vista, Windows 7 world very quickly. KDE 4 is much more important than some folks seem to realize. If desktop Linux doesn’t get this very right it could be in real trouble as an alternative that delivers what PC users will by then have come to expect.
Indeed!
I would prefer a weekly column by him here on OSAlert. His comments are always so insightful and well written.
I’m glad they ditched dcop in favour of dbus. Finally better interoperabiltiy between kde and gnome applications.
I think a lot of the comments in the Art discussion are unfair when comparing KDE 4.0 quality to Windows. The KDE team 1) doesn’t have the funds to test as extensively as the Windows team should be, and 2) they told us all along what the quality was going to be of this release was and why.
If you didn’t hear their disclaimers you must have had your head up your as… OR be one of those ppl who only have KDE installed by their distro, which haven’t started shipping this and thus are totally uneffected by this.
Remember, the Linux 2.6 kernel had to have a few releases before it was really 2.6 quality.
Geez people, all this about a FRICKIN NUMBER!!! Really, it’s not like they’ve fooled us or misled us, they just changed the number… which frankly, is something I’m going to say they have the right to do.
I think that’s silly. Funds or not, the 4.0.0 indicates to a user that this is the fourth major revision and it’s ready for release. KDE is using their production release as a massive beta test because they couldn’t get the feedback otherwise; they are deceiving their users and hiding behind the “we told everyone it’s not ready for production use” excuse.
I wholeheartedly agree. These are just misleading excuses. You can’t claim that it is unfair because they do not have the funds to beta test. I find this quite ironic because one of the claims to promote OSS is that the community at large can work through bugs quicker and more efficiently than closed source. So which is it? Is OSS model just bad, or maybe KDE4 was just released way too soon (me thinks the latter than the former).
And yes, that silly number IS important. By implying that a revision number is too be ignorant simply means that for most all software such as KDE should be ignored period. The whole reason and logic behind it is to let users know what to use, what to test, and what to avoid. This is error number two on their part. You simply can NOT use the argument that makes an assumption that people should just know. It simply does not and can not work that way. Keep in mind for every one person that spends their life reading about this, there may be 2-3 that will be unaware. Not everyone has the time or interest to follow everything in this world. Fact remains there will be people that will stumble upon 4.0 thinking it is a final/ready to use GUI when in fact it is a far from prime time release.
If anything the excuses that are being thrown out are worse than the actual release.
I wholeheartedly agree. These are just misleading excuses. You can’t claim that it is unfair because they do not have the funds to beta test. I find this quite ironic because one of the claims to promote OSS is that the community at large can work through bugs quicker and more efficiently than closed source.
Quick! What’s the difference between KDE 4.0 and Mac OS X 10.0? … The KDE developers were honest with what it was?
Not being able to afford large scale beta testing doesn’t adversely affect the speed in which bugs are resolved, it affects the number going into the first release.
The truth is large scale OSS products get the majority of their testing done in .0 releases, every single one.
Things aren’t much different in the proprietary world. The difference being that you generally get a honest assessment of the state/quality of the product in OSS.
That’s a nonsense argument, and you know it. Just because Apple does it, it’s ok for the KDE team to do the same? The Apple team sues bloggers and employs wild DRM schemes, does that mean the KDE team can do so too?
Come on, you can do better than that.
The point is everyone does it, and it’s something that can’t and shouldn’t be avoided — except the dishonesty.
Hang on a moment Thom…Microsoft does it…Apple does it…hell, most games are released these days with more bugs than what lives in my backyard (and I’m pretty aware of what bugs live in my backyard, since I’m a macrophotographer). It’s the defacto norm, whether we like it or not.
There are several issues here – people are far less inclined to download and install a beta/rc release, than a final release. That’s a well known fact, and if you look at download numbers, I think you will find that it proves my comment. Less users = less potential bugs found. Furthermore, most of the people that did download it probably never reported any bugs. They simply downloaded it out of curiousity.
KDE and many other open source projects generally have very little funding, and not enough developers to go around. Corporations like Microsoft and Apple have tonnes of money, and tonnes of talented developers. So, what’s their excuse for releasing dodgy products?
As to betas, I had a mate download XP SP3 RC for me and post the CD to me (I’m on dialup, so wasn’t worth the effort for me). I installed it and wow…what a p.o.s. Boot times went from 90 seconds to over 10 minutes on my system. The system went from relatively fast, to damn well super sluggish. I’ve now uninstalled it (another 20 something minutes spend doing that, with another reboot taking 10 minutes for things to load and repair, etc) and cannot recommend it at all. Sure, it’s a beta, but hell, if Microsoft can’t get a late release RC right…
The real truth of the matter is that KDE was honest with its users, indicating that KDE 4 was going to be a backbones release, nothing more and nothing less. Do you, and others, want KDE to stagnate like E17? Ten years in beta whilst they perfect things for a perfect 4.0 release would be a total joke, and would kill the project.
Furthermore, some distributions comments that they won’t support KDE long term release are major BS. 3.5 is supportable, without issue. Said distributions (I won’t name them, cos we all know who I’m talking about, and we all know my immense dislike for said distributions for their modus operandi) are just using that as a front to cease distributing KDE in favour of their Gnome favouritism rant. Period.
Dave
Well not really. Yes it sucks that .0 releases have relatively poor quality, but I haven’t seen any large scale software project (for consumer use) that does it differently. OS X 10.0, Vista, Gnome 2.0, etc They all had major problems, similar to KDE 4.0. So unless you have a magical way to develop bug-free software without large scale user testing on a realistic budget, there isn’t really much of an alternative. If you have one the software engineering community would love to hear about it.
In this case, a good way is to look at the top three or five most used applications/functions, and ask if they have even 60% of the functionality from the previous version. If there’s even a question about it, then no, it’s not ready for release. As much as I’m a little tired of snipings at the kde devs, come on now. It doesn’t take a giant usability team to figure out that the panel is just a toy or that fish is messed up.
I’m sorry, but that’s not true. Vista was arguably buggy and only recently has begun to perform well after all the patches, but already on the day of its release was finished, it was feature complete, and it was not only stable enough but rock-solid. And that’s without having into consideration that Vista is a project way more complex than KDE is.
What do you keep dredging up this ridiculous counter-argument for? I’m more than a bit worried that you agree that Apple uses this method, because they are selling a paying product that should be ready, with all the alphas, betas and technology previews used in advance to shake down the software. Clearly, with OS X 10.0 and Vista this method hasn’t worked – even after years in development.
Given that, how do you expect an open source desktop environment, taken in a new technical direction as KDE 4 is, to follow this broken model? The open source community has always existed on the principle that you released early and released often and you chucked out point releases for people to try out until everybody got to a point where they were confident and they all said “It works for me!” This is how open source software has always been used and how many people now use it for all sorts of purposes. Distributions are generally part of that layering process as they don’t package up stuff to use that they know has some problems to sort out and some development to go. That’s pointed out and people know about it.
As Butters has said above, if you’re not going to be a part of the solution then get out of the way.
Come again?
For the millionth time: nobody is arguing against the concept of release early, release often. Please, show my any comment where I said that this concept is flawed, or that KDE 4.0.0 should not have been released.
All I (and a whole boatload of other people) am arguing is that calling this release .0 – meaning, relatively stable, feature complete – is a mistake. It should have been called what it is: a developer release, just lik BeOS did back in the day. They released several develop releases to attract interest from developers, and they succeeded at that pretty well – for such a new operating system, BeOS already had a fairly large pool of software available for it.
The KDE developers contradict themselves majorly with this release: they say it’s not ready for normal users, yet they say they want to release it as 4.0.0 to gain bug reports from normal users, because without the 4.0.0 tag, they would not use and test it. So, which is it? Do they want users to test this, or do they want users to skip this version and dive into KDE 4 with 4.1?
Confusion abound, which could all have been solved by simply naming this thing KDE 4 Developer Release 1. The interest among the technically inclined crowd (which is what they want, right? The technically inclined? Seeing they discourage non-technical uses from using it) would not have been any less, and I, and many others, would still have installed and tested it.
For the ten billionth time, .0 is a release number, nothing more, and you do not get to decide what the requirements of that particular release are. You do not get to decide the requirements that makes a piece of software that you are not involved in ‘feature complete’. There is every indication that it will match your requirements and ideas at a given point, but when that happens we’ll know about it, won’t we? The more widespread usage of open source software these days has worked on that principle, not on release numbers.
Yep, and we all know how successful those are at ironing out bugs and strange corner cases before a .0 release. As far as the developers are concerned, getting a developer release, as you call it, was the aim for .0. Get it?
4.0 meets the developers’ requirements for a stable set of development libraries and a stable known starting point. Users who can will try it out, test it and report back, but like everything in the open source world, it will be widely used when enough people think that it’s good enough. KDE is no different to any other open source project, and no different to Gnome 2.0 as it was. Few would have thought that was ‘good enough’ at the time.
You define users as if they can all be thrown into one pot. They can’t. The line between users and developers is pretty blurred in the open source world. Users who are aware and want to try this out can, and those who just use a distribution won’t see it until it’s good enough. The usage of open source world has always worked on that principle, as it does for the people who are waiting for Vista SP1, SP2 or even the next version of Windows: “Is it good enough? Does my OEM or distributor think it’s good enough? Will my software run? Then we’ll use it.”
1. Experience of Vista development previews, betas and release candidates (certainly when you look at some of the issues that came out post-launch), of previous KDE alphas, betas and release candidates and Gnome release candidates and even point releases, has shown this clearly not to be the case. It is highly doubtful that this Ars review would ever have been published if it was an alpha, beta or RC.
2. People simply assume that a .0 release in any project is good enough for ‘ordinary users’ to come in and criticise, and that all the changes are apparent to the user. That isn’t the case. There may be technical goals and requirements in a release that have no bearing for end users.
3. You can’t underestimate the value of coming from a known starting point, with known bugs and known feature deficiencies at any given time. You just simply don’t get that from alphas, betas and RCs.
if the feature list for 4.0 is complete and there are no show stopper bugs then you release. Release early, release often.
And of course there are differences between a free software project and a commercial one. No one is asking you to pay money for KDE4 and noone is making the distros package it and make it the default DE. The only people who can even get at it are those technical enough and interested enough to know what they are getting into.
That’s what svn access, alphas, and betas are for. You are *severely* twisting a CatB catch phrase to suit your purpose by using it in that distorted way. (And when did that become the Bible, anyway?)
Edited 2008-01-22 00:29 UTC
the definition of ‘show stopper’ is something that (hold your breath) stops the show. If alpha and beta testing reveal no more of these, then it is time for release.
How on earth did I twist the phrase ‘release early release often’ when we’re discussing something that people are complaining about being released [i]early[/]??
I don’t thing anyone is claiming the Cathedral and the Bazaar is any kind of bible, bit, you know, its about developing free software and kde is free software, so hey, not that far out of left field I’d have thought.
So, in summary, a free software project, in concert with the opinions expressed in a book about free software, released a piece of feature complete software, mainly so that application developers have an ABI stable base to develop against and people are complaining like the developers individually came round to their houses, formatted their systems and put a gun to their heads and made them install KDE4.0
As far as I’m aware, there is not a single distro with KDE4.0 as anything more than a preview package at present.
Programmers have a stable development environment to work against. Power users get to run the latest and greatest and start filling out bug reports, making suggestions and contributing however they feel. Distros get something that they can start packaging, port their unique apps to. Technically unsophisticated users don’t even need to know it exists. And people with nothing better to do, who get something for nothing sit around and complain about other people’s hard work on the internet. Which category do you fill?
Please address how SVN availability, alphas, betas, and developer previews do not fulfill the function of releasing early and often? You completely sidestep that obvious point and imply that the CatB mantra means that they should “go gold” early and often, which is, in my opinion, the best recipe for credibility loss. And current events, if you’ve been watching, *do* support that opinion. KDE *has* lost credibility. Not due to the state of KDE4 at this time, which is not bad, all things considered. But because of the poor judgement shown by the project leaders, *and their unwillingness to face up to it*. The KDE community seems to have gone into full defense mode over it, willing to argue the most nonsensical positions rather than admit that they released an alpha as a gold release.
Sure, DE enthusiasts can just shrug that off. But those who are in a position of making decisions on deployments of desktops in business *must* take notice of the level of responsibility demonstrated in the *actions* taken by the project leaders. It’s our *job* to take notice of things that affect the future of the deployments for which we are responsible. And the coolness of plasmoids, etc. should not be allowed to cloud our judgement.
Edited 2008-01-22 12:45 UTC
Have you actually tried developing on a desktop out of SVN? I’ve been developing against KDE4 for the last month and a half and having a moving target (which SVN, alpha/betas and dev previews all are) is a pain in the backside. Since .0 I’ve had a stable API/ABI and a functional desktop that lets me progress with development rather than catching up on continuous system changes.
With who? The devs like it, the distros appreciate having a stabilized build target and most user won’t see it until the distros are ready. So who does that that leave? a handful of people with enough clue to build from SVN, install from dev repos or run a live CD but not enough of a clue to realize what it is they’re talking about when they run off to the web to complain.
all of this is nothing more than your opinion, and going by both what you’ve said here and in other posts, in my opinion that isn’t worth much. What exactly have you contributed? Any examples of your judgment floating around that would give weight to these continuous attacks on the works and choices of others?
If you have enough brains / responsibility to make these assessments and you’re looking further up stream than what the distros support then you ought to have enough sense to not go with .0 release of a major project re-write. All this talk of bad judgment is just total BS. No one is forced to use KDE4. No distro is pushing it onto their users. It isn’t the default desktop anywhere. Just because you have a different ideal as to what makes a release .0 ready compared to someone else sure as heck doesn’t make you right.
Observations. I’ve made observations. If those observations make some people uncomfortable, then so be it. I’m certainly not against KDE in general, or KDE4 in particular, or I would not be going through the pain of actually running KDE4 on my main desktop box.
You still have not provided a credible answer for why a very buggy codebase with large areas of missing functionality needs to “go gold” before it can be developed against. I do development using the SVN branch of Django all the time. The API is not completely frozen, but the developers are *very* responsible even with respect to the SVN. API changes are minimal, and very well documented. But if the developers of an API take a more cavalier approach, I can understand that you might have trouble with their SVN branch. Considering what the KDE guys apparently consider “gold release” standards, I can imagine that their SVN would be a nightmare. But there is no reason that *has* to be. And certainly no reason that reasonably stable betas and Developer Previews could not be released.
for example?
I have. And then the following example you provide:
which, if you are honestly offering this up as a comparison, indicates you’re not really qualified to even try and make the ‘observations’ that others may find ‘uncomfortable’ (bah). KDE is one of the biggest FOSS code bases around. A web framework like Django doesn’t even have an ABI to worry about protecting for a multi-year development cycle. And wow, how do they manage to control the API of such a monster? I’m not trying to run their work down, but I do web based development for a day job, I’ve written more than one framework myself and believe me its a task light years apart from what the KDE guys have achieved.
at this point our conversation is getting repetitive and unless you demonstrate some deeper understanding of the realities of large scale software development then I don’t think we have a lot left to say to each other on this topic. Take my withdrawal however you like…
Well, if you are quite finished launching personal attacks upon me, then I would say that it is logical for you to sign off. Though I’m not sure why you think one should have to be experienced in developing large scale C++ projects to have an opinion upon whether it is reasonable for a DE project to release a pile of very buggy code, missing large sections of functionality, including, but certainly not limited to, allowing the user to move, hide, or change the size of, the panel, and call it a gold gold release. The best answer you have given me so far, once the personal attacks have been filtered out, is that if they had called it a developer preview, you would not be able to develop against it because the api would be too unstable. But since they are trying to pawn it off as a gold release, you can. I guess that’s just something they teach you in “Developing Large Scale C++ projects 101”. It makes no sense to me.
I guess I’m just used to OSS projects, even large ones and C++ ones, having some pride in, and concern about, the quality of their gold releases. But of course, not being a KDE developer, I’m not entitled to feel that the quality of the release is poor, simply based upon my hands on experience with it.
I’m already kicking myself for hitting reply…
Site one. The worst I did was call into question your judgement and portray your public efforts in a negative light. You do of course see the hypocrisy in complaining about that don’t you?
Its more than just language specifics, though of course understanding those is important. Its about introducing a large scale rewrite of very large, publicly developed project to all of the downstream people who need access to it in a stable way. The least numerous but noisiest of these people being the ‘DE enthusiasts’.
And unfortunately yes, the specifics of development of this type of project are important and if you don’t know enough about them then maybe you should restrain from commenting so voluminously and so vehemntly, especially in the face of repeated correction.
Oh heavens, stuck with not being able to change the panel size? How long was it from 2.0 until GNOME could right click on a menu and edit it? Where were the drivers in Vista? In any software project of this scale you will never get immediate feature parity with such a mature, feature complete code base as kde3.5. This is a testament to the progress made in 3.5 not a slight against 4.
For someone so free with the vitriol when they’re dishing it out to others you get remarkably thin skin when a little of it comes back your way.
I’m sorry that that was all you managed to pick up and that it makes no sense to you. Maybe someone else can do a better job of walking you through it, but I honestly think this is a ‘lead a horse to water but can’t make him drink’ case.
again, call into question the professional pride of others, cast dispersions on their character and work. Ok to dish it out but quick to complain when you receive it.
You are more than welcome to feel any way you like about it, but in light of what has been explained repeatedly about who the .0’s intended audience is, when you come out in public, ignore what has been said numerous time and then continuously cast dispersions about the quality of something you don’t seem well suited to judge, a perception that is hard not to make reading your posts, then don’t be surprised to be called on it.
For it’s intended audience, 4.0 is wonderful. Its a delight to develop on and was delivered not a moment to soon. Sorry you don’t feel that way, but don’t expect a warm reception when you go out trying to rain on others parades.
Let me make absolutely certain that I’m clear on this point:
KDE is intended to be a DE usable by regular users. But you believe that to be qualified to judge whether it should have been released for general consumption, one must possess a deep technical knowledge and have hands on experience with the development of large projects. Correct? And you don’t think that there is anything nonsensical about that position?
something you don’t seem to mind when you’re making the points but back away from when other request the same of you… but I digress…
Yup. And if a regular user grabs a distro and picks KDE as the DE they’ll get a powerful, configurable, feature rich desktop. KDE 3.5.
Where is it available for general consumption??? From source? from dev builds in a few distros? This is not an operating system that comes pre-installed on computers from best buy and will be in the hands of joe uers hours after its tagged. You keep trying to phrase the discussion in terms of commercial software when that just doesn’t apply to a community developed project that will by and large be deployed to the users through 3rd parties. If you go to the KDE web site and go to the download section http://kde.org/download/ you’ll see a list of distros that package dev builds and source code. How is this ‘generally available’? Without the (repeated) statements of the developers the very nature of what you have to do to get 4.0 alone should indicate to you that it’s not available for general consumption yet. But it is feature complete, there are no show stoppers and there is an audience for it, so it was time to release.
your original assumptions were flawed so your conclusion above is invalid.
Actually, he was pointing out that you were trying to claim credibility by comparing apples to oranges, as it were.
The issue of buggy code is questionable. I’ve been using KDE4 as my standard desktop since it was tagged, and I’ve yet to run into blockers that would force me to revert back to KDE3 or any other desktop. It works. Any apps that are missing in KDE4 are still very accessible in their KDE3 form.
Yes, I’m a bit irked that I can’t move the panel to the top of my monitor as I’m used to in KDE3, or XP for that matter, but holy f&AS($*(# s*&#$%)(#$ *#$(#%&S, with all of the work that has gone into KDE4, it’s actually galling that people would consider that a blocker.
Missing large sections of functionality? I don’t remember seeing the document where KDE4 committed to having specific functionality for 4.0.0. Yet when Gnome 2.0 removed large sections of functionality, it was praised as an advance in usability. Give me a f* break.
And the mere fact that you keep referring to it as a gold release hints to your proprietary software bias. Gold releases go to the plants for mastering. It implies a “nothing can change beyond this point” mentality. This is OSS. The parameters are different.
The API’s were frozen back during the testing cycle. That’s more than your own pet project can account for. Everything done since was bug-fixing and polishing.
You’ve got a chip on your shoulder for whatever reason, that much is clear from your posts.
At the end of the day, KDE 4.0.0 is a working DE. It might not do all the things you or everybody else wants it to do, but that’s a small sacrifice for the benefits the community expects to see down the road.
KDE 3.5.x still exists, so it’s not as if the devs are trying to pull some underhanded maneuver to force people to migrate. 3.5.9 will be coming out around the same time frame as 4.0.1.
Nobody is forced to use 4.0.0. It’s there and it’s asking for attention, but’s not demanding it. Which is good, because it’s becoming clear that the cake-and-eat-it crowd wants things handed to them right-now-this-minute, without making an effort to contribute.
You’re somewhat atyipcal of that approach. You keep mentioning all these “clients” that you manage and support with linux and OSS as if that’s supposed to be some carrot for the devs to follow, but at the end of the day, have you really contributed back to the projects you’re using, or are you simply leveraging them to supplement your income?
If you had participated in the development cycle, even with only the articulate and intelligent feedback that I must say you are very capable of providing (that’s a compliment, not a sarcastic dig, FTR), then you at least have a bone to pick. But instead, you’re grandstanding, and it helps neither the community or the developers.
Give a little, gain a lot.
That seems to be the perception that some people have. Possibly because my observation has a fairly narrow scope, but people seem to want to interpret it with a very broad scope, and the discussion has gone way off topic. My basic observation, based upon watching the reactions to KDE 4.0.0 and using it myself, when I can, is this:
—
It would have been a better strategy, and more optimal for KDE4’s credibility, to have called what we know as “KDE 4.0.0” a “KDE4 Developer Preview” instead. And it would not have hampered development.
—
That’s the important bit. But here is some further elaboration. It would have set the proper expectation for reviewers. KDE fans would not have had to scramble to claim that nobody said it was really reading yet and that reviewers were judging it unfairly. It would have avoided a lot of the comparisons to the quality of Vista when it came out. And it would not have stolen the thunder of the release which is planned for July.
Calling it a Developer Preview would not have changed the current level of stability of the API or ABI.
And it would have been more accurate. I’m a stickler for calling a spade a spade, instead of calling it a golden trident with a host of qualifications afterwards.
A little more patience would have paid off with better PR both now and in July, which is something at which FOSS projects, in general, could stand to be more skillful. As I have said before, there is nothing worrisome about the state of development of KDE4 at this point in time. It was the presentation that was unwise.
You don’t understand development. Having open repository access, and releasing alphas and betas is one thing, but getting a catch-all in place for lots of bugs, feedback and getting enough people to actually really test something is quite a different matter.
There is also a point to tagging something, setting it in stone and releasing – you are working progressively from known points and know issues and bugs that existed at a point in time.
Because that’s the way it works, and it’s why people have talked an awful lot about development methods such as Extreme Programming and alike and why there has been so much scrutiny over the years as to why most software projects fail to meet expectations – ever, no matter what the release number.
Welcome to 2008. I mean seriously. People don’t understand this?!
You are openly admitting that the 4.0.0 lebeling of the current code base was an intentional ruse to trick people who would not have upgraded to it had it been called an alpha, beta, or developer preview, into moving to it?
Well, I’m surprised. But I suppose that direct admission is progress, of sorts.
Considering the way that KDE 4.0 is being treated by most distributions, I don’t agree. As if this point about KDE 4.0 not being production ready hasn’t been made a thousand times on a thousand sites, it is not as if any mainstream distributions are going to try to provide KDE 4.0 as if it is a production-ready environment either. At least, none that I know of.
Frankly, I think it’s more like one out of twenty people, at most, don’t know that 4.0 isn’t production-ready and installs it by mistake.
This is a non-issue.
Yes, it’s important but not the way you people think. A version number is just an arbitrary indicator that it’s a new version and that it’s different from the previous. Many people are just making way too much noise about the fact that it’s a “major” version change. Well, guess what? It doesn’t matter if it’s “major”, it’s just an indicator that X is newer than Y. I went from OpenBSD 3.9 to 4.0 and nothing groundbreaking happened. I guess I should have been disapointed because 4.0 was just better and that it didn’t blow my mind?
Who do you think the community are you ejit? Do you really believe that KDE fashions its own in-house community and test team out of……nothing? If you aren’t a part of it, get lost.
Yes, that’s exactly what it’s for. It’s not there for you to measure your requirements up to a project that you had no hand in developing and that has no obligation to you personally or financially. You use it when it’s good enough for you and matches your requirements. You had a point somewhere in the middle of all of that?
There are dozens of clueless ejits who went off and downloaded CTPs of Longhorn and complained when it didn’t work properly, or when it hosed their system despite warnings. Some people just won’t read anything. Go figure. The majority get their installations through OEMs and distributors.
Who gets to decide that it’s not a prime time release? It’s certainly not you I can guarantee you that. It’s the developers, who had certain goals and requirements (stable development library interfaces etc.) for .0 before they moved on. If those goals and requirements differ to yours, well tough luck.
OEMs do not install CTPs or betas of Windows for end users, and distributors will not package KDE 4 as their default desktop in any form for some time yet. If people stumble across a CTP or beta of Windows, or they stumble across any piece of software that isn’t good enough for them despite a lot of commentary on the subject, well, tough luck.
People will stumble across a lot of things. C’est la vie.
Excuses for what? That it doesn’t match up to what your own idea of what the requirements for a project are?
How exactly are they deceiving users when they tell them before and after release that it’s a de facto beta / developer release?
1. To lead into error; to cause to believe what is false, or
disbelieve what is true; to impose upon; to mislead; to
cheat; to disappoint; to delude; to insnare.
[1913 Webster]
Don’t be so naive. Everyone knows there’s an agreed definition of “release.” When the KDE team suddenly up and decides they are going to redefine what a .0 release means and release beta quality software in their general release builds, and then pretend that their entire userbase is somehow anxiously reading their website, it reeks of excuse. This is NOT ready for primetime, and should not be called 4.0, and it’s nothing short of silly to assume people will read the directions or check on this stuff. It’s deceptive at worst, merely irresponsible at best.
I’m sure KDE4 will be great when it’s ready, but it is NOT ready for users right now, and users who go to download things and they will download the most current release, which is 4.0.
To believe they can successfully redefine what a .0 release means to everyone by posting some byline on a website is truly naive.
To believe a .0 release is anything more than a glorified beta intended to get things going is naive.
You can’t fault them for the exaggerated ignorance of others.
Edited 2008-01-21 21:58 UTC
A beta can be expected to be reasonably feature complete. KDE 4.0.0, which I am running along side my usual desktop, is actually alpha quality software. Crashing, which it does frequently, is OK for a beta. Large area’s of missing functionality, in addition to crashing, are not.
The state of development of KDE4 at this point in time is not worrisome. The attitude of the project devs in releasing this as a .0 release and then persisting in their defense of doing so *is* worrisome to me as administrator of several dozen corporate desktops.
Edited 2008-01-21 22:20 UTC
O.K., .0 releases are typically far from perfect. However, the KDE devs have pushed this argument beyond any known limits.
However unstable and buggy .0 releases may be, they should at least be reasonably feature complete.
I feel that this is a major source of much of the frustration. How can I be expected to use KDE 4 and report back, when there is hardly anything you can do with it: No mail client, kopete does not work for me at all, no calendar, no address book, still not the promised webkit based browser etc. pp. I sure can bear with a fair number of crashes, however, I would really like to be rewarded with some functionality for doing so.
Edited 2008-01-21 22:38 UTC
Could you please stop posting your repetitive flamebaits? The KDE team uses the exact same definition of a dot-Oh release as GNOME did for their 2.0 release, as Apache did for their 2.0 release, as X.org did for their 7.0 release, as Apple did for their Mac OS X 10.0 release, as the Linux team did for their 2.4.0 release, as Netscape did for their 6.0 release, as Microsoft does with basically every Windows release, and so on.
Your countless repetitions don’t change the fact that similar releases were made over and over again.
We all understand that you try to boost page hits with your flamebaits, but it gets old. Really. Just stop it. Please.
What-ever. This is neither flamebait, nor is it influenced by anything other than the same willingness ANYONE has to comment on this site. In fact, I’d argue the reverse, when idiots come here and attempt to nullify our staffs’ comments by suggesting it’s solely to “boost page hits,” I think it’s lowest and lamest form of trollery. It’s clear you would prefer to avoid the issue rather than participate. If you just want to have your own groupthink opinion continuously spit back at you, I invite you to go to Digg.
I posted a comment, and responded to the responses. That’s hardly the flood of posting you’re suggesting in your post.
KDE4 is not ready for general consumption and the version number is therefore misleading. That much is clear. But to suggest that is okay because others have done it that way is an absolutely pathetic excuse.
It’s easy being an armchair critic. So please, enlighten us how this could be handled better. And no, waiting another 6,12,18 months is not an option so don’t bother reiterating that stupid idea. Everything needs testing, and developers can’t labour away at unreleased software forever.
Sure it is. You just don’t like it. That’s doesn’t make it “not an option.”
They can and they should, at least until it’s actually ready for release. I hope you NEVER said anything negative about Vista, after all, they can’t work on it forever!
Let the world know: Leos says “don’t bother testing your software or waiting until it’s ready to release, just ship it according to a calendar and let the users find the bugs.”
I can’t even tell you how ridiculous this argument is. If this were Microsoft (and it was), they would be skewered alive, but since it’s an open source project, people are so forgiving. Maybe the groundwork of KDE4 is incredible, but what exists today simple doesn’t show the user something finished, and that IS cause for complaint.
Not a realistic option. Look at E17. That’s how far that option gets you. I’d rather have KDE 4.0 Buggy than KDE 4.0 Duke Nukem Forever.
Hahaha. So you’ve never done any open source work before I guess. Working on software without a release for that long is not fun, and eventually developers will just leave.
I said plenty of negative things about Vista. Just like I said plenty about KDE 4.0. Both are buggy and definitely .0 releases, but that doesn’t mean it was wrong to release them.
My own software I test thoroughly. But only because I can. It’s small, and I can test it myself before delivering it to the customer. With 4 million lines of code, you need the help of users. So unless you have an example of a software project of that size that managed to release without serious bugs, you’re full of it.
You’ve just destroyed everything you’ve written right there. No, they couldn’t work on it forever, which negates the alternative approach you say KDE should be taking.
Bloody hell. Who do you think does the real testing in the open source (and even the proprietary) world you fricking ejit?! At least open source projects and their development model admits that.
Probably because you’re stupid.
Because in the open source world in particular, development is done in the open and the ‘community’ makes a collective decision about when it is good enough. That’s the way it has always been open source software, and the way it will always be.
I mean, you couldn’t just download a 2.0 release of Bacula and assume that you could start just using it to back up all your critical files, unless you and others tried it out and were confident enough in its ability.
How many people who are simply users do you think will see KDE 4.0 as their default desktop when they log in in the morning?
OK- What happens in the broader ecology? The 3rd party apps all start another feature cycle against 3.5. Porting doesn’t start. You still don’t get a complete KDE 6-12 months down the road. What happens in KDE’s SVN? The maintainers of the polished apps, since thay can’t hack, move on. The developers of the less complete bits are hacking and bugfixing in a in an empty house as the first group developers is only paying marginal attention, or have moved on. Worse yet, you get this nasty case of incest– Without a release you’re preaching to an ever shrinking group of folks.
Truth– The KDE Developers pushed the release back twice. They assessed the gains of sitting on KDE 4.0.o for longer and reached a consensus that it wasn’t ‘worth it’.
Well, there are a fair number of people doing skewering IMHO– So it’s hardly all flowers and puppies. As far as the ‘not finished’ goes, KDE 3.5 isn’t ‘finished’ either. 3.5.9 ships next month. Software is never ‘finished’. Now, if you mean, “Yarr, there be bugs and features have gone overboard!” I’m right with you.
Now, with that being said– The difference between Virgin Vista or OSX 10.0 is that with KDE you’ll see a lot of bugs quashed on the Jan 30th release. (With thanks to the broader testing 4.0.0 gave it.) You’ll see panel ‘regressions / missing bits’ get filled in over 4.0.x, and you have the whole KDE ecosystem based around KDE 4 by the end of the summer. That’s pretty darn quick turn around, no?
I hear what you’re saying. By this logic though, why not release six months ago? Why not release 3 months ago?
Let me sum up everything I’ve said in one sentence: if you have to tell *real users* “this isn’t really the finished package,” you shouldn’t be tagging it as a general release.
That’s all I’m saying, and I stand by it.
OK– it boils down to ‘good enough’ vs ‘as good as’. Since this is a KDE thread we’ll use the Dolphin file manager. It didn’t spring free form into existence. If the developer said, “I’m not releasing until it’s as good as Konqueror.” there would be no Dolphin. Back in the KDE 3 days it was released as a basic file manager- Some people saw it, liked it, helped out and gave feedback. Another release, same thing. And again. And again. Now it’s in KDE 4.
If the developer waited until it was ‘as good as’ it would probably be bitrotting on a hard drive half ported to KDE 4. For active projects the benchmark HAS TO BE ‘good enough’. Because ‘good enough’ is needed to get to the point of ‘great’. That’s the real value add of distro’s stable release… Even if you disagree with their choices, they have the role of the arbiter on the meta level. On the project level ‘good enough’ is ‘good enough’.
For example, you might be an informed choice about the risk vs reward of installing any given bleeding edge piece of software… But try doing that for every package on a Linux box. Say, “I Heart My $DISTRO”. So if unsure, wait for your distro– Otherwise a projects call on what’s ‘good enough’. And KDE 4.0.0 meets that bar, at least here.
No one is arguing about Dolphin. If you want to continue this thread, please consider using a private message, this thread is way too deep.
Do you think your insults somehow make you credible? You’d probably have to ban yourself, because insulting other posters is clearly against this site’s very own rules.
Please understand that your flaming has absolutely no point, because the 4.0 release has already been made. You can’t change the past. 4.0 is out since a while, 4.1 will be released in July.
Get over your ego. You are not the KDE leader and with your attitude you most likely will never become an integral part of the KDE team. Thus nobody from the KDE team will ever listen to you and your rants against the KDE release policy.
This is simply not true. Netscape 6.0 was, as we all remember a special case, as Netscape was in a quite desparate situation at that time. However, GNOME 2.0 was of course far from perfect when it was released, it was unstable, it was buggy and so on, but it was relatively feature complete. (o.k. it took several months more for Evolution to get ported, however, Evolution officially wasn’t part of the desktop at that time).
GNOME 2.0 did not ship with a mostly dysfunctional panel. What they released was unquestionably more than just a shell without a roof.
The same is of course true for MacOSX and Windows Vista. Imaging MS would have released the final without a mail client and without an address book, postponing both to SP 1. Get the difference?
Edited 2008-01-22 09:57 UTC
The release was made and its shape was far worse than KDE 4.0.
GNOME 2.0 had almost no GUI features for modifying settings.
Mac OS X 10.0 was not even able to record CDs or play DVDs. You couldn’t even use Classic apps to get this features, because hardware access is required for them.
Mac users had to boot into Mac OS 9 to be able to do that. OTOH it’s no problem at all to mix KDE 4.0 and KDE 3.5 apps.
You don’t get the difference. A X11 desktop is heterogen by nature with a mix of toolkits being the norm rather the exception. It’s no problem at all to run Okular, Evolution and K3B side by side.
You also don’t get that KDE 4.0 is not an operating system, but a desktop environment. A Linux distribution is an operating system and it’s the distributor’s responsibility to offer a “complete experience”. With mail client and so on.
BTW: Vista is the first Windows release with a discrete address book.
Of course, you can do the same on Windows and MacOSX. No need to run safari or mail.app, noone forces you to use outlook express and windows address book.
And of course, you don’t need KDE4 for doing so. If you just want to run a bunch of unrelated applications, you can do so in IceWM or twm.
Edited 2008-01-22 11:44 UTC
There are any number of philosophies on when to release software and all kinds of factors play a part in choosing which to pick.
KDE went with the traditional FOSS approach of release early/release often. They have their reasons for this approach. It may not mesh with what you want, but then you (I assume) aren’t writing it.
No one’s arguing the release. We’re arguing whether it should be tagged 4.0.
If its not tagged 4.0, people won’t test it. This is true at least for myself. I installed it on 3 machines here when it hit 4.0. I wouldn’t have bothered with a beta or developer preview or technology preview. Sure it has lots of issues, but at least I tested it.
Congratulations. You just did.
And your point is?
I’m saying that there is evidence to support the idea that labeling this anything but a release would have resulted in far fewer testers, making it take longer to get better.
No there isn’t. Releases are defined by requirements, and only developers get to decide when software has met the requirements they wanted and that a release is warranted – not you. This is much more the case in the open source world since development is done in the open.
Wonderful high-horse sentiments, but you don’t get to decide what is beta quality software and what isn’t, and you don’t get to decide what the requirements of KDE 4.0 are, and were. If you were paying licenses for this then you might have some comeback (and if you were you’d be pointed to a service pack or the latest version), but you simply do what everyone else has done with open source (and even proprietary) software for years – you ask “Is it good enough?”, and if not, you don’t use it. If it isn’t, presumably your distribution will not make it its default desktop, and you have no problem.
Welcome to open source development sweetheart. As Butters says above, if you aren’t going to be a part of the development process and the solution then sod off. Otherwise, you’re the inside layer of the onion that only gets to see KDE 4 when the outer layers have been stripped away and enthusiasts and especially distributors say “This is good enough and it is now our default desktop when you install KDE.”
Who downloads what? The users who download KDE and try it out know what they’re getting into (as with any piece of open source software), and it’s beyond the reach of most people who use KDE. The vast majority will pick it up when distributors think it’s OK to package it in earnest, and when system administrators think that it’s a good move for their users (an extra layer in the onion).
Once again, you don’t get to decide what goes into a .0 release. As an OSAlert contributor, if you don’t understand how this works then, quite frankly, you really shouldn’t be posting articles, comments and writing about this stuff.
If Microsoft or Apple decides to have a more open and inclusive development model with users, and they release regular point releases, and they point out that the .0 or odd numbered version numbers are for testing and feedback only, and many people download it, install it and there are some bugs and missing functionality, you honestly believe that they have cause for complaint?!
Sorry, but you’re so stupid it isn’t even funny. Become an ‘ordinary user, and quit worrying.
Users don’t install KDE, they install a distro. Users dont really care if it’s called 4.0, 4.1 or 9 & 1/2 weeks.
Sorry, it’s not deceptive. If they had told everyone this would be an awesome, mindblowing release and they changed it at the last moment, yes, that would have been. However, they’ve been saying from the start it’s not a full-fledged update but mainly a framework update.
Thankfully most ordinary users aren’t stupid enough to read too much into version numbers.
Plasma has drawn a lot of criticism for its vision: which is to completely change the way we interact with our desktop. There’s been all kinds of negativity because KDE 4.0.0 hasn’t met that vision with the very first release. If a new cell phone company springs up with a vision to “make the world’s most durable cell phones with the best battery life”, but releases a fairly average phone as their first product, it’s not a failure. Their vision isn’t the first step to take – it’s that point off on the horizon you keep working towards. Same thing with KDE4 – and it was a welcome relief to see Ars acknowledge that Plasma, as it is now, is a framework, the first step towards realizing the vision for KDE4.
So *show me*. So far all I have seen is a bunch of talk about how fantastic, revolutionary, and all around totally, mind-blowingly ***COOL*** KDE4 is *going to* be. That and a very unimpressive and lackluster 4.0.0 which is a big step backward from 3.5.
Now, maybe KDE4 is going to be all these wonderful things. But *show me* don’t tell me. I’ve been using KDE 4.0.0 since it came out. (Dropping back to my usual desktop when I need to get real work done.) I watched Aaron’s presentation. It all *sounds* great in his talk. But aside from watching KStars and Marble over and over and over getting a bit tedious, the only new feature that he was able to demonstrate in the 1 hour and 20 minute talk, other than a prerecorded video of the desktop effects, which are pretty old hat under Linux desktops now, was tagging a media file from Dolphin.
Show me.
Edited 2008-01-21 20:25 UTC
The point of view that you show here, by saying “show me”, is that of a *consumer*, which illustrates perfectly this post of butters:
http://osnews.com/permalink?297105
Please do yourself a favor and don’t restrict yourself to being a mere consumer — that’s good enough for Apple/Microsoft users, but Free Software users are supposed to behave differently. The whole point of Free Software is to blur the disctinction between producers and consumers, instead everybody is an active “user”.
This is no more than empty rhethorics. The posting you responded to basically says that the reality check he undertook showed that most of the bold promised made by the KDE devs haven’t materialised yet.
Stating that has little to do with consumerism, it is basically a statement of self-evident facts.
I would add that there are two issues. One is that simply more time is needed to port or re-write the missing bits.
However, the second one is that in my view at least, it is still a complete mystery how some of the bolder promises with regard to changing and improving the user experience are going to be fulfilled. Heck, what is plasma going to mean for the user except for the ability to clutter your desktop with dozens of applets? What exactly is revolutionary about this?
So, I feel that on the conceptual side, i.e. regarding user interface design and usability, there still a huge void which ought to be filled. Else, KDE4 will remain attractive mainly for hackers but for users.
What you say would make perfect sense if Plasma was indeed going to fail to fulfill its promises. But it’s in fact going to succeed, thought that is not visible in 4.0, and the great idea that will make it successful can be simply explained:
The basic idea of Plasma is to make a framework that is general and abstract enough that none of the traditional elements of a desktop (panel, background, desktop icons…) are hardcoded. Then in KDE 4.0, for lack of time all what they could do with that framework was a basic panel etc, and in KDE 4.1 all what they will do will be feature-parity with 3.5 so I guess you will be disappointed again, but the underlying framework will allow just about any other kind of desktop concept. So at this point we should start seeing interesting third-parties plasmoids reinventing the desktop.
In other words: nobody knows right now how to reinvent the desktop. But with plasma, anybody will be able to play with ideas with a few dozens of lines of python code, whereas with the KDE 3.5 architecture each component of the desktop was hardcoded as thousands of lines of C++ code.
There are other benefits of plasma such as integration in apps (amarok 2.0) and suitability for other form factors (media centers) but that’s not my point.
The Ars article did explain the potential of plasma as a framework and it should be clear to everybody that this is a great idea.
And it has never been tried before. Former implementations of desktop widgets were just that — desktop widgets. Plasma allows to reinvent not just desktop widgets but every component of the desktop.
Technically, this is only made possible by the QGraphicsView introduced in Qt 4.2 and will benefit a lot from the huge improvements in Qt 4.4 (to be released in 2 monthes, will be used by KDE 4.1).
So what you are saying is: We have the technology, Concepts are still missing (“nobody knows how to reinvent the desktop”) but will eventually appear out of nowhere.
I think it is a misconception to treat UI concepts as a minor issue. I believe that they are at least as important as the underlying technology and I doubt seriously that without a comprehsensive concept of what kind of user experience KDE is meant to deliver, anything beyond simple desktop gadgets will come out of plasma.
KDE 4.0 already has a taskbar and several menues implemented with Plasma which I would consider being beyond “simple desktop gadgets”. You also have form factors and containments and plasmoids which (in theory, and some already partly implemented) adjust themselves to the containment they are associated with. As Aaron mentioned in his keynote this will make it possible to develop to develop GUIs for a wide range of different devices i(e.g. PDAs) with the same base libraries. So there is a concept albeit a very technical one. Maybe it’s just not flashy enough for the “next gen” Linux users…
We all know that the taskbar in its present state is barely usable. In fact in its present state, it is little more than a simple gadget.
The menu (which a recent article on osnews.com referred to as an “abomination”) is not an original idea of the KDE 4 developers but a re-implementation SuSE’s kickoff. Having it re-implemented in plasma does not at all improve its usability over the original program.
Improving the user’s experience would take much more than cool technology: It would require cool and thought-through concepts of UI design. This is, what I’m actually thinking of when I read the pledge “reinvent the desktop”.
I am surprised that this seems to be a minor topic for many KDE folks.
What I saw was an impressively well-designed framework. I think they are showing us the start of something big. But, this takes a lot of work; they already have thousands of volunteer programmers working and they’ll get there.
Design is important. You don’t tack on integration between applications later and get good results, you design it in from the start. You don’t tack on support for hot swapping peripherals and sound and networking for a whole environment. If you want to share the libraries between all these applications and not have redundancy, it’s best they be well-designed from the beginning, and that’s what the KDE folks are doing.
Don’t just throw down the gauntlet to a lot of very hard-working people in the middle of a huge project. Pick up your C++ compiler and get cracking on the code.
Do it. Now.
I’m pretty sure such would be percieved a failure and would be heavily damaging for the reputation of the respective company. I believe that the public is pretty much mercyless in this regard.
Which does not rule out that you always have a second chance to compensate your earlier failure. Look at the history of Mozilla, whose reputation also reached pretty damaging lows around the release of Netscape 6..
Edited 2008-01-22 09:22 UTC
Personally, I would have called it something like “KDE 4 Developer Preview 1”, but that is just a semantic issue. The real problem is that the KDE devs are making a big deal of this release at all. Why is there a 3 day “KDE 4.0 Release Event” for what is essentially a developer preview?
The real problem is that the KDE devs are making a big deal of this release at all. Why is there a 3 day “KDE 4.0 Release Event” for what is essentially a developer preview?
You are trolling right? Oh well, I will bite.
Do you think all the code for KDE 4.0.0 wrote itself?
The developers have just passed a major milestone. No, KDE 4.0.0 is not a developer preview, because the devs say it isn’t. Now, let them celebrate their success or they will run out of enthusiasm and burn out.
BTW great work KDE devs, have some fun and I look forward to your future work.
What’s the point in discussing what it should have been called at this point….It’s out now and there’s no going back =)
So what are developers thinking?
See even if it worked, the new direction of KDE sounds like a step in a direction I don’t care to go. There were promises in Vista, and despite being told once you use it you will see, well I haven’t. I don’t like most of what was changed. Same in Leopard for a few issues. And now KDE seems to either be following in some of the same direction or develops all over are thinking some of the same things.
I can believe the developers in all of these see some benefit to the framework, as that is what they all claim is being improved. So far, what they thought users wanted seems different than what users want. It might come to pass, it might pay dividends down the road. Cannot see it so far.
And yes, their excuses ring hollow. If this had been called a Release candidate it has so many problems I wouldn’t agree it even deserved that name. Been trying to use it, but have to keep turning it off. Just not functional enough to mess around with very much.
I know this has to bug these people some. They are putting in many man hours for no compensation. To release it and have people whine about it like I am doing has to rub them the wrong way. I think they should have gotten it further along before calling it sort of done myself to avoid that. I know I have been waiting with anticipation to try the first release. Didn’t use any until this as I thought it was a release. I mean it is or it aint. i didn’t expect perfection, I didn’t expect it to be without issues. I did expect more than 4.0 in fact is.
Frankly, I’m not surprised about the bugs in KDE 4.0.
KDE 4.0 introduces a bunch of brand new technologies that haven’t had any time to bake yet. The KDE 3.x series has about six years of baking behind it.
Vista was late by 2-3 years and still had lots of problems that are only now getting addressed in the service pack.
Mac OS X 10.0 was similar in terms of “roughness”.
I think KDE 4.1 or 4.2 will be the first reliable releases.
Edited 2008-01-21 23:08 UTC
I read the warnings and I expected bugs and rough spots. I have been using Linux for a long time now and am no stranger to running beta software on my desktop.
I wanted to use KDE 4.0, helping out if I could with testing and patching. I still want to use KDE4; it shows a lot of promise, especially the underlying APIs. But even with the warnings I expected KDE 4.0 to be a usable, if buggy, desktop environment. Put simply 4.0 is not usable as a desktop, missing so many features that even basic tasks become a chore. To even use 4.0 at all for prolonged periods of time I have found myself keeping a terminal open (which I usually do anyway) for launching apps and logging out when plasma completely crashes or stops redrawing, taking the panel and desktop with it. That simply is not step forward as a usable desktop no matter how charitable you want to be.
Hopefully sometime soon plasma and kwin will stabilize to the point where I can use it for more than an hour or two at a time. For now though I will just use the KDE4 apps that are (somewhat) stable. The KDE team have done a great deal of work and they deserve praise, but plasma is not ready for use.
Edited 2008-01-22 00:24 UTC
Does anyone here know about Resolution Independence in KDE4 and QT4? If I buy a 200dpi+ resolution monitor in the future is KDE4 ready for that monitor? If not would anyone like to do an article about that issue with regards to KDE?
Plasma is largely resolution independent and Qt 4.4 (to be released in a couple of months) will make this even better as it allows widgets in a QGraphicsView (and every object in a QGraphicsView is fully scalable, can be rotated, sheered etc etc while still allowing for input). A fully scalable interface can be build upon that, and TT might do that already for Qt 4.5 on which KDE 4.2 will be build.
wonder if we’ll get KDE5 LCD,
60+ comments and only 3 or 4 related to the actual article lol
I agree with the author about most observations. I have tested KDE 4.0 on kubuntu and I’m quite disappointed. Meanwhile, I’m explained over and over by Aaron Seigo and other smart people that KDE 4.0 is not KDE 4. I’m also told that underlying technologies are very promising.
Well, about that last point, since I’m an optimistic person I want to believe it is true.
Now, I just want to say this: in my humble opinion, users, I mean ‘real’ users (you know the so-called Joe) don’t give a dam about that childish war 4.0/4 and the debate about alpha, beta, RC, developer releases is nowhere among their priorities. All they want is something that work and possibly something appealing and looking nice.
Actually KDE 4.0 fails in both ways. Right now, it is a technology preview and even that point is not quite a success because as such it also fails to show off all the gorgeous future KDE 4 or 4.0 is supposed to bring out.
I have no doubt, KDE will get better starting from that present state. It’s just a pity it’s not ready for users right now.
My two cents about the GUI: it looks terrible. No consistency, no elegance, eveything look bold, most desktop and interface ‘elements’ are oversized, wrong default font, too much use of bold fonts, mix of big and tiny icons in the panel is awful. I stop here, I could go on and on for a very long time.
Of course, it’s just MY opinion but please, please, ask independant GUI designers, fresh and new people, people not working already for KDE, ask them what they think and what is good or not. Just ask them. There are tons of students in design schools you could ask for that.
And don’t give me the ‘you can change everything if you don’t like it’. Sure it’s nice to be able to do so BUT most users don’t go further than changing the background so KDE needs a strong and nice default look. It’s essential. The technology may sound nice but if it’s wrapped into the wrong package most people won’t come to it.
Well, I think Oxygen looks fresh, beautiful, and elegant. The article writer seems to like it too.
Still, take a look at the Bespin theme that the article mentions. Maybe you’ll like it more than Oxygen. One of the many usability enhancements in KDE4 is that downloading new themes is now trivially easy.
http://kde-look.org/content/show.php/Bespin?content=63928
Oops, that is not correct. I was thinking of the wallpaper downloader that is now integrated to the Plasma desktop. Hopefully the other aspects of desktop look get soon similar integration.
I do know Bespin and found it superior to Oxygen.
However I would not call bespin ‘trivially easy to download’ and install if I may add since Joe User should go through SVN and compile the whole thing according to bespin homepage: http://cloudcity.sourceforge.net/download.php
Of course, bespin is still in beta stage. Like most kde4 themes by the way. I’m sure there will be plenty of themes in a ‘I hope’ near future. But anyway that wasn’t my point. My point was I feel they have made wrong choices about the whole default look and UI, not only the theme, but the bold font everywhere, the default font, the size of that thing supposed to replace the kicker, the border surrounding icons with three tiny icons which make desktop icons look like aircraft carriers, the gigantic clock, the taskbar with both a huge icon and text which looks completely lost aside the icon, etc.
And again I would be happy to hear from independant and non linked to any open source projects, graphic or UI designers about KDE 4.0 interface because I feel many points are deeply wrong.
KDE4 has too many bugs! I can’t imagine how the developers could release this to the mass…
I have used KDE4.0.0 for at more then a week, and I have found very few bugs most are with plasma, and most should already have been fixed for KDE4.0.1.
What bugs are you talking about? Since you feel you can claim that KDE4.0.0 is a mess? Have you even tried KDE4.0.0 or are you just trolling?
Edited 2008-01-23 19:13 UTC
Tell me, have you reported these bugs to http://bugs.kde.org or are you just too busy complaining about them?
– Gilboa
Edited 2008-01-24 14:18 UTC