AUG
11
2005

KDE 4: Understanding the Buzz

With all the excitement surrounding KDE 4 development at the moment people are starting to ask why they have not seen any updates on what KDE 4 will look like. KDE 4 - Understanding the Buzz answers these increasingly common questions by explaining the current status of KDE 4 development and why the exciting work so far is only visible to developers. "Before any new features can be added to KDE and projects like Plasma can get underway, the porting of KDE to Qt 4 has to be completed."

Comments

Also, people need to realize that even the status of KDE4 since the article was written is still in huge flux. The kdelibs API is going to be changing and when that happens apps will be broken until they've been updated to reflect the new API.

This is a tremendous and complicated effort. It'll take time and patience on the part of library developers, application developers, third party developers, documenters, translaters, artists, and of course... our users. Stick with us though, we have every reason to believe it is going to be marvelous when it is done. :)


By manyoso at Thu, 2005/08/11 - 5:00am

Yes indeed, it is a HUGE job.

I can't help but wonder if this is the wisest course of action. That is, a professional software developer would advise against the plan. We are doing several things at once and this is not wise. If things work out correctly, we will probably have a great product, but if they don't, we could wind up with a large mess on our hands.

A more conservative road map would be to just do the port and release it as 4.0 and THEN start on radical changes for 4.1.

IAC, with the current plan, I expect to see delays so we should plan on a 3.6 release.


By James Richard Tyrer at Thu, 2005/08/11 - 5:00am

Not really. It's called unstable development, and it happens in every single project, commercial or otherwise. You get to something stable through iterative development and improvements.


By segedunum at Thu, 2005/08/11 - 5:00am

And the history of commercial development contains examples of new versions that went a bridge too far resulting in release delays.


By James Richard Tyrer at Thu, 2005/08/11 - 5:00am

> resulting in release delays

Since when are Open Source projects on a fixed release schedule? A delay in the KDE 4.0 release, so what?


By cl at Fri, 2005/08/12 - 5:00am

That would depend on whether it was a week, a month, or a year.

Wouldn't it?


By James Richard Tyrer at Fri, 2005/08/12 - 5:00am

No.


By Chris Howells at Fri, 2005/08/12 - 5:00am

Indeed, no. Its ready when its done... This is Free Software. There are no PHB's and marketing goals, only quality software...


By r_a_trip at Fri, 2005/08/12 - 5:00am

Well, GNOME is, as a matter of fact. I'm not saying that is a positive thing, just saying it is.


By Fruity Frog at Wed, 2005/08/17 - 5:00am

"And the history of commercial development contains examples of new versions that went a bridge too far resulting in release delays."

Is this supposed to tell us something new? Sometimes you just have to accept that something is complex and work with it.


By segedunum at Fri, 2005/08/12 - 5:00am

The aim of the large versions numbers (v3, v4) is to maintain binary and API compatibility through-out so that app developers can be guaranteed their exsting code once ported will continue working throughout the series (i.e. code written to v3.0 still runs unchanged on v3.4). To have a v4 and v4.1 with radically different API's is counter to this aim. What you propose is what in effect is happening, but just without the public release in-between. First a port to QT4, then change the API as required, then release v4.0.

John.


By odysseus at Thu, 2005/08/11 - 5:00am

> That is, a professional software developer would advise against the plan

And we are all just hobby software developers, right? I can assure you that there are more professional software developers working on KDE than you might think.

> A more conservative road map would be to just do the port and release it as 4.0 and THEN start on radical changes for 4.1.

a) And what should our user do with this great KDE 4.0? Doesn't make sense to release stuff that is just a port to a new library version.

b) We aren't doing several things at once. The plan is to first port KDE to Qt4 and then make the big API changes. Nothing else is happening in SVN. We just don't release the port to the public (see a).

> IAC, with the current plan, I expect to see delays so we should plan on a 3.6 release.

We don't have the man power for another KDE3 release. After KDE 3.5, everybody will be working on KDE4.


By Christian Loose at Fri, 2005/08/12 - 5:00am

> And we are all just hobby software developers, right?

Actually I don't know what people consider themselves to be. I don't hear anybody else saying that they studied EE/CS in college, perhaps because the culture here denigrates such an education.

> a) And what should our user do with this great KDE 4.0? Doesn't make sense to
> release stuff that is just a port to a new library version.

I notice that when people want to disagree with you that they seem to skip over the adjectives in the sentences. :-) I referred to *radical* changes. Not doing radical changes doesn't mean not doing any changes.

What the user would do with such a release would be to use it. There is supposed to be many improvements in Qt 4.0 and IIUC there are planed improvements to the libraries. This should be enough unless you advocate "featuritis" as the important thing in software development. Adding new features should not be the driver in KDE development now that it is mature software.

Perhaps all will be well. But, I keep reading about completely redoing this, completely redesigning that. I start to wonder how this is going to be done with the "limited manpower".


By James Richard Tyrer at Fri, 2005/08/12 - 5:00am

I would suppose that most of KDE's professional developers have studied IT or something near, in universities or "high schools".

And if EE is "eletrical engineer", then there is at least Eva who is one:
http://www.kde.org/areas/people/eva.html

http://people.kde.org/dirk.html tells me that Dirk Mueller has studied IT.

And probably there are more people having studied anything scientific.

Have a nice day!


By Nicolas Goutte at Fri, 2005/08/12 - 5:00am

See also:
http://www.kde.org/people/gallery.php
that confirms what I have thought and written.

Have a nice day!


By Nicolas Goutte at Fri, 2005/08/12 - 5:00am

Hmm, that page needs a little updating...


By Paul Eggleton at Sat, 2005/08/13 - 5:00am

While I don't disagree with you when you say that the current projected KDE 4.0 may be a little more than we can chew off, I do know that unless you aim high, you must fall short of the requirements..

While I haven't got a CS MA (I'm a linguist, comparative linguistics of the Sino-Tibetan languages of Nepal -- eastern Nepal -- to be precise), I've got about a decade of experience in the software industry. Aim high, damn the certifications and forget about cmm, that's the only way to produce something worthwile. Ours is a craft, not a profession.

And make sure your daily build can be released as-is.


By Boudewijn at Fri, 2005/08/12 - 5:00am

> KDE 4.0 may be a little more than we can chew off,

That is really my only point. that we should be careful that we don't:

'bite off more than we can chew'.

The history of software projects is littered with the remains of projects that did that. Yes, you are correct that we should aim high, but we need to be realistic regarding what we can accomplish for the first release of KDE-4.


By James Richard Tyrer at Sat, 2005/08/13 - 5:00am

I bet most changes which will be actually done will be rather evolutional and consequential for the developers who work on the stuff. It's just that the necessity of BC for the whole 3.x series helt those bigger evolutional changes back until the 4.0 release which allows breaking BC once again. So the huge amount of changes make the effort look "radical", but isn't if you actually look at the details. So calm down and stop worrying. =)


By ac at Sat, 2005/08/13 - 5:00am

If CS means "Computing Science" I studied 4 years at the university and before
4 years in school and 2 more years before as hobby. I'm sure (and know) that there are many others who have similar education in the KDE project.
The fact that I don't have a master degree and a doctorate is that I find it useless if you just want to be a developer. ;-) I better learn by experience. I'm sure that sometimes those who learned to code by themselves are better than some who studied at universities.


By Andras Mantia at Sat, 2005/08/13 - 5:00am

I'm glad to hear that there are professionally trained people working on the project. Perhaps we can form a group to do what ever.

The question of knowing how to code verses knowing software engineering is an old one that I won't start on here since it is way off topic except to say that much has been written about it.


By James Richard Tyrer at Sat, 2005/08/13 - 5:00am

This is not really workable. If we release 4.0, then the BC compatibility requirement comes in force which will then stop us from doing any "radical" changes. The model currently followed has been used before and has worked very well for us. Think about KDE2 -> KDE3 effort. KDE3 has been hugely successful so far.


By Nadeem Hasan at Fri, 2005/08/12 - 5:00am

I don't see why the API changes can't be made without completely redsigning the desktop and the panel. And the dozen or so other radical redesigns which I read about. Yes, the library API changes need to be made along with the port to Qt-4. But, every week, I read of more changes.

We need to remember that Murphy was an optimist and try to keep the project from growing too large to handle.


By James Richard Tyrer at Fri, 2005/08/12 - 5:00am

IMHO you should leave the decisions to those who do the work.


By cl at Fri, 2005/08/12 - 5:00am

A complete insult to the large numbers of highly skilled professional software devlopers that _are_ working on KDE and have done since it inception.

But I should have known, James Richard Tyrer is the expert softare developer that can save KDE from impending doom.


By Chris Howells at Fri, 2005/08/12 - 5:00am

The history I refer to is about work done by highly skilled professional software developers. The Lotus 123 release debacle comes to mind. They simply let the project grow beyond what they could accomplish and there was delay after delay and when they finally released it, it was unstable and buggy. Then there was the project to completely rewrite Windows for the Windows 2000 release. It was quietly abandoned because they found that despite having a PhD professional software developer running it that it was simply more than could be done.

I am not a professional software developer and I do not claim to be one. I am an electronic engineer. What I said was that a professional software developer would give you certain advice. I am not that person. The advice, however, is valid. The advice I would give as an engineer is to always remember Murphy's law.


By James Richard Tyrer at Fri, 2005/08/12 - 5:00am

> I am not a professional software developer and I do not claim to be one.

Well, I am a professional software developer working for a company that creates real estate management software (http://www.aareon.de). And I tell you, the development plan the KDE developers have for KDE4 is fine.

Pleased now?


By Christian Loose at Fri, 2005/08/12 - 5:00am

It isn't a question of being pleased now.

I will be pleased when the release is successfully completed. Till then, I will continue to be concerned that what is starting to sound like a total redesign of everything is simply too much to do for a single release.

Actually, I would be much less concerned if some of the developers shared my concerns instead of exhibiting what appears to be over confidence.


By James Richard Tyrer at Fri, 2005/08/12 - 5:00am

> Pleased now?
You remember me of a teacher I once had, who thought he was always right just because he had studied psychology.
Project leaders know how, or at least they think they do, to distrust software developer time estimations or at least multiply it with five or so.

Btw. is there a plan for this transition, saying which job is done by (or under supervision of) who?


By koos at Fri, 2005/08/12 - 5:00am

> You remember me of a teacher I once had, who thought he was always right just because he had studied psychology.

Then you misunderstood me.

I was replying to James point that professional software developers would advise the KDE developers to do the port differently. And I answered him that he is wrong because:

a) I'm a KDE developer
b) I'm a professional software developer
c) I would do the port exactly like it's done now.

I don't remember saying I was always right or that I studied anything.

> Btw. is there a plan for this transition, saying which job is done by (or under supervision of) who?

Don't think so.


By Christian Loose at Fri, 2005/08/12 - 5:00am

No that's what I meant to say. Since reasoning is a process of the mind, how can a psychologist be wrong?

> Don't think so
I find this a bit frustrating. I have time to work on KDE, but I really have no idea whom to contact or what areas to work on. Patches/questions sent to devel ml seems to either stay unanswered or interpreted as a personal insult of the author of that particular piece of code.
From the other side, I've seen someone porting code to Qt4 that I more or less maintain and completely breaking it. Never had any mail asking about this.
This is complete chaos IMHO, some list of the maintain->library should be available I think, no?


By koos at Sat, 2005/08/13 - 5:00am

"Patches/questions sent to devel ml seems to either stay unanswered"

Possible if the original authors/people familiar in the code aren't involved in KDE atm. And most developers don't read kde-devel due to the high traffic volume.

"or interpreted as a personal insult of the author of that particular piece of code."

Sorry, but you are completely wrong if you think that this is the opinion of authors on the whole. I don't remember ever experiencing it. Personally I am delighted that someone is using my program and then gone even further to try and fix something that they didn't like.


By Chris Howells at Sat, 2005/08/13 - 5:00am

"From the other side, I've seen someone porting code to Qt4 that I more or less maintain and completely breaking it"

Is this in KDE? Well, maybe that's because the initial aim of the porting was to get it compiling against Qt 4, not a completely fully working port to Qt 4 -- if we had that we could release KDE 4 now.

"Never had any mail asking about this."

So did you mail them to express your concerns?

"some list of the maintain->library should be available I think, no?"

There are a few lists of maintainers of various classes in kdelibs.


By Chris Howells at Sat, 2005/08/13 - 5:00am

For maintainers, much of kdelibs has MAINTAINERS files, and if that fails, there is always (c) notices. And it's probably best to use kde-core-devel and not kde-devel for library patches


By SadEagle at Sat, 2005/08/13 - 5:00am

I have to report the same experience with patches.

I have posted them to bug reports, I have posted them to kde-core-devel, etc.

Occasionally, people take it as an insult, but mostly I am ignored.

So, now I am posting stuff to KDE Apps/Look: KDE Improvements. But, that wouldn't be appropriate for everone.

If there are any other frustrated would be contributors that think they could put up with the old engineer, please contact me by private mail and I will see if we can start a project.

tyrerj@acm.org


By James Richard Tyrer at Sun, 2005/08/14 - 5:00am

> I have to report the same experience with patches.

Please start to understand: it's about you. First making yourself unpopular with your comments and then expecting to be received cordially doesn't work.


By Anonymous at Sun, 2005/08/14 - 5:00am

Interesting theory:

Better the bug goes unfixed than have someone that you don't like fix it.

And I thought that this was supposed to be a meritocracy. :-)


By James Richard Tyrer at Sun, 2005/08/14 - 5:00am

You have a SVN account for one and half month now - and did you make one single commit since then? No! You just keep bitching...


By Anonymous at Sun, 2005/08/14 - 5:00am

I deliberately waited till the 3.5 branch was split.

Then I have had a lot of problems with SVN. I thought that I had it working OK and was going to do a build that would be up to date and I got mysterious error messages:

svn: Delta source ended unexpectedly

I posted this to kde-devel: 08/10/05 21:30. Nobody had any cure. A Goggle search turned it up as a bug. So, three modules were dead. Other problems continue. I keep getting the wrong "admin" module. Am I just unlucky -- I never had any trouble with CVS.

So, I have a build started again and I hope that it goes OK. If so, I will be able to do some work.

However, you should not presume that just because I have not committed anything that I haven't done anything. I have icons and patches ready to commit. Some of the CrystalSVG ones were posted to KDE-Look.


By James Richard Tyrer at Sun, 2005/08/14 - 5:00am

Try to clean your local working copy with:

svn cleanup

(I do not know if it will work or not but when you get problems under SVN, it is a good try.)

Have a nice day!


By Nicolas Goutte at Sun, 2005/08/14 - 5:00am

> Please start to understand: it's about you

I second that; its impossible to have a discussion since the good old engineer always has to be right and never gives in one inch. Thats not what a discussion is suppost to be about.
Its faster to fix the problem yourself then to get into a discussion with JRT, is what I always say.


By Foobar at Tue, 2005/08/16 - 5:00am

Ah, so you now acknowledge that there are lots of highly professional software devlopers working on KDE. I guess that makes your idea of allowing the plan to be examined by professional software developers moot since they would presumably have already shouted on core-devel if there were problems.

K thanx.


By Chris Howells at Sat, 2005/08/13 - 5:00am

I never said that there were or weren't professional software developers working on KDE. It has NOTHING to do with my original posting. You appear to have, basically, misunderstood my original posting (perhaps with intent?).

I still have the same opinion. That a professional software developer would caution the KDE project against trying to make too many changes for the KDE-4 release.

Note that the meaning of 'professional' generally indicates a person that is paid for their services. And that is how I meant it.


By James Richard Tyrer at Sat, 2005/08/13 - 5:00am

"That a professional software developer would caution the KDE project against trying to make too many changes for the KDE-4 release."

None of the "professional software developers" working on KDE so far have, so maybe you should STFU.

"Note that the meaning of 'professional' generally indicates a person that is paid for their services"

I'm glad we're in agreement over something.


By ac at Sun, 2005/08/14 - 5:00am

Look AC, I am tired of this game.

I stated my opinion. I have explicitly said that it is my opinion.

Is it your position that I am not entitled to have my personal opinion?

Is it your position that I am not entitled to express my personal opinion?

This is FREE SOFTWARE as in FREE SPEECH! But, you appear to be against free speech. Note that, in case you don't understand, that free speech means that other people are going to express opinions and ideas that you don't agree with.

Unfortunately we see a lot of comments like yours both here and on KDE lists. You have nothing constructive to add. Your only statements are that what someone else said is wrong. You don't state why they are wrong. You don't offer a constructive rebuttal. Your remarks reduce down to a childish: 'No, your wrong'. What does that accomplish.

Nobody has to accept my opinion, but I am not going to change it just because some AC says that I am wrong. Especially when they take what I say out of context.

Now, you are still not attaching the same meaning that I am to "professional software developers". But, I am not saying that there aren't any working on the KDE project. I wouldn't say that because it is a negative hypothesis.


By James Richard Tyrer at Sun, 2005/08/14 - 5:00am

> Look AC, I am tired of this game.

Everyone is.

> I stated my opinion. I have explicitly said that it is my opinion.
> Is it your position that I am not entitled to have my personal opinion?
> Is it your position that I am not entitled to express my personal opinion?

The point is that no one really discussed your opinion because you managed once more to make a remark that was understood as offensive, in the very same posting.

Just how are the KDE devs to interpret what you said?

"a professional software developer would advise against the plan"

1. It could mean "one professional software developer would advise against the plan". Now that's far-fetched and certainly not what you meant, or is it?

2. It could mean "Any professional software developer would advise against the plan."

2a. It's a well-known fact that there are many professional software developers working on KDE using the definition of "professional" that has been established in this thread (some of whom have even spoken up in this thread and they said they didn't advise against the plan). So with this interpretation what you said must be false, this is basic logic. This was dead obvious right from the start to the people who replied to you so they discarded this interpretation.

2b. Or what you said was to imply that the people working on KDE are no professional software developers (as no one of them has advised against the plan). That's how your posting was understood as it's the only possible interpretation that is not obviously wrong. No wonder people feel attacked and took offense.

You stated later in the thread that 2b was not what you meant either. But then I've run out of possible (not-obviously-wrong) interpretations. And it seems I'm not the only one.

I'd advise against using such broad statements ("professional software developer would...", "engineers would...") in the future. You don't even speak for all engineers there are, let alone all professional software developers. Just skip this stuff and talk about the subject.


By cm at Sun, 2005/08/14 - 5:00am

> But then I've run out of possible (not-obviously-wrong) interpretations. And
> it seems I'm not the only one.

There is a term for people that interpert general and broad statements as applying to them (especial in a negative way): 'paranoid'. You, and others, misinterpreted what I said as some sort of attack on somebody. The statements made by various persons that in NO way addressed the question of whether this project -- which someone else characterized as a "HUGE job" -- is too much, seem to say more about the persons who made them than they do about me.

No, I didn't mean 1, 2, or 3. I was talking about a hypothetical person (this is a common rhetorical or literay device -- a subjunctive construct) and I meant someone that develops software commercially as their profession (the obvious definition). It should also be clear from context that I was talking about someone who was OUTSIDE of the project. Now, we may have a few "professional software developers" people working on the KDE project, but they are not working on KDE as their profession -- they are doing it as their hobby (and that includes me since I am a RETIRED engineer) and they certainly are not outside the project.

So lets just say that considering the (unbelievable) over sensitivity of some people that I should have said: 'an outside "commercial software developer" might advise against taking on such a "HUGE job" for a single release'. I am not saying that ALL such developers would say that, only that some would. If I had meant 'any' or 'all', I would have said that. However, I do agree with this hypothetical person. That is my point, and I wonder why nobody addressed the real point.

I plead guilty to not speaking precisely, but that is all. Why did I say it that way? Because I did not want to claim to be a "professional software developer". I did not intend any insult and the fact that others took offense is very questionable. It seem to be a game to me. One which I would prefer not to play anymore. When many people find the need to take everything personally, it makes rational discussion of any issue very difficult.


By James Richard Tyrer at Sun, 2005/08/14 - 5:00am

> You, and others, misinterpreted what I said as some sort of attack on somebody.

In fact I didn't. I was pretty sure you did not *want* to attack anyone personally. I just tried to explain why people *did* take offense by pointing out the different possible interpretations. Obviously your own was not included, but that should make you think.

> Now, we may have a few "professional software developers" people working on
> the KDE project, but they are not working on KDE as their profession -- they
> are doing it as their hobby

Wrong, not all of them. There are several people I'd call "professional software developers" paid by Trolltech, SuSE, Eric Laffon and others to work on KDE, and some even full time.

> If I had meant 'any' or 'all', I would have said that.

Maybe my command of the English language fails me here, but to me a statement like "a sw dev would do this and that" is as general as "a sheep produces whool" or "rain is wet". Leaving little or no alternatives.

What you said was understood as "If you were professional software developers you would see that this plan cannot work".

I know that's not what you wanted to say, at least not consciously.

> However, I do agree with this hypothetical person.

Why do you construct that hypothetical person at all? It does not give your point any more authority than you already have. I think most people here know about your engineering background by now. So why not let your opinion stand on its own instead of resorting to hypothetical sw devs or engineers. Otherwise people *will* point out that there are sw devs and engineers who disagree with you and your hypothetical expert.

> That is my point, and I wonder why nobody addressed the real point.

That's what I was trying to explain.

People may or may not be *over*-sensitive but it's not the first time this happens to you (and to few others as consistently I might add). I think you'd have a better chance to get your actual points discussed if you took that sensitivity into account.


By cm at Sun, 2005/08/14 - 5:00am

> Look AC, I am tired of this game.

And I'm tired of some stupid incompetent cantankerous old wanker who does not understand the free software culture, does not understand the open source culture, does not understand what drives KDE developers, does not understand how free software is written and released generally attempting to stick his oar in and making a mess of things.

If you can't understand how you've offended some people then there is nothing more to say, you are a completely lost cause.

Free speech allows you to say what you want. It does not force others to listen to what you are saying. Since I don't want to listen to the same old bollocks that you've been spouting on kde-devel for ages I'm kindly requesting that you go and find another project to troll.


By ac at Sun, 2005/08/14 - 5:00am

Admin, this post crossed a line I don't like to see crossed on the dot or anywhere in KDE.


By cm at Sun, 2005/08/14 - 5:00am

Pages