I've finally managed to get the last bits of the KDE 3.2 Alpha 1 (codenamed "Brokenboring") including KDevelop 3.0 Alpha 6
on the ftp server. The mirrors should soon pick it up.
There won't be any binary packages for this release because the KDE "Pi"
release is coming out soon. Everyone using Brokenboring is asked to
compile it with --enable-debug, so that we can get valuable feedback. There is a new unstable version of Konstruct also available to install it.
Other than that, feel free to play around with things, check if your bugs
are still there or if there are places where you can help. Check
the KDE 3.2 feature plan
for things to look for.
The code is quite rough in many places, but many of the developers
use it on a daily base and kdepim has improved so much that you can't live without it once you've tried it. :)
Thanks to everyone who convinced me that the code is good enough for
a release when I was tempted to drop the idea altogether. Just one final wish: check
for duplicates before you file bug reports and compile with debug information enabled
before reporting crashes -- you might find tons of bugs, but with
debugging information the chances we can fix them are much higher.
Comments
I've been using KDE cvs in a dialy basis (I'm not a developer), and all I can say is: Good work, people! It's fast (performance options), the context menus are being reorganized and the overall quality is pretty good for a pre-alpha release (I can do a lot of work here).
Problems: I am still experiencing rendering problems on konqueror. I wish that apple code merging on khtml could be made faster...
So, if you use debian unstable, give a try on orth's page at
http://hobbiton.opendoorsoftware.com/
For binary packages from CVS. It's worth a try!
Lucas
is there a possibility to install those without removing my kde 3.1.3 packages?
- kata >8^)
http://developer.kde.org/build/build2ver.html
> Problems: I am still experiencing rendering problems on konqueror. I wish that apple code merging on khtml could be made faster...
Merging apple's code isn't the holy gral and it isn't easy:
1. Both are moving targets which makes merging complicated.
2. Apple also makes mistakes or changes that don't make sense. That's why you have to check the patches thoroughly. Sometimes this can take up so much time that it makes more sense to program the functionality yourself.
3. KHTML developers and the Apple developers sometimes have different opinions on how things should be solved. Another reason why sometimes patches aren't merged.
4. Sometimes the developers didn't explicitly stated that a cvs commit was a merge from Safari.
5. etc...
Christian
-----
http://www.kde.org/apps/cervisia
6. Sometimes when it's stated that the code is merged from Safari, it's really a nearly full rewrite that's smaller, faster, and more correct.
For everyone here:
Let's live in peace :) We don't need to fight with gnome people, they are our friends. Both DE's are evolving and nowadays they are aimed to different audiences. We can learn a lot from them, as they can do from us. This is all about free software and free software is made with cooperation.
Lucas
+1 Insigthful
So GNOMEs are good choice :)
Hi!
How difficult would it be to use some programs from KDE 3.1
and some programs from of KDE 3.2 on the same system? I would like
to keep using kmail from 3.1, but might want to give konq
from 3.2 a try.
If I install kde3.2 into /usr/local/kde3.2 and not point
my PATH towards this directory, would I then be able to
use my old kmail by typing kmail and the new konq by typing
the full path to it? Of course i want kmail to use the kde3.1
libraries and never the new unstable libraries, and konq
to use the new kde3.2 libraries.
If not, any other nice way of doing this?
What happens to my configuration files in ~/.kde? Can i
fool a kde program into using an alternate directory for
configuration files, for example by setting some environment
variable?
(for what it's worth, I use slackware 9.1, which has kde
and all libs in /opt/kde)
In theory every KDE 3.1 program can be compiled against and can run with KDE 3.2. In practice KMail is one of the most stable programs in the KDE development versions, so don't hesitate to use it. Better don't mix different KDE installations at run-time, problems with configuration etc are immiment.
The best and easiest solution is to create a special user account for KDE 3.2. Then you can either start two sessions or you simply log into the other account, e.g. with 'ssh -X -l kde32user localhost' in Konsole and then start whatever program you want to try from the command line.
That would solve the configuration file problems.
What about the libraries? Will KDE 3.2 programs use the new
kde 3.2 libraries, and KDE 3.1 programs use the kde 3.1
libraries automatically?
I don't want for example kmail from kde 3.1 to use the
kde 3.2 libraries, since they are alpha-stable. Also,
I want to have konqueror from kde 3.2 use the kde 3.2 libraries
since they contain new stuff.
Well, you can have only one version of kdelibs-3.0 installed at a time. They are binary compatible, so old KDE-3.1 programs would still work with kdelibs-3.2. But it won't work the other way around. kmail from KDE-3.2 won't work with the old kdelibs, because new features have been added and are used by kmail and the other KDE-3.2 programs.
Result: If you want to use kdelibs-3.1, then you must stick to KDE-3.1 programs as well.
This is no problem. Have a look at http://developer.kde.org/build/build2ver.html .
Why on EARTH would you want to bother setting up an ssh connection to localhost with X forwarding?
Why not set xhost appropriately, and do:
su - kde32user
?
David
what about the amount of bugs?
kde 3.1 was released with many. ( 2000+ ? )
and their was talk about bringing it down in 3.2
has this been a succes?
Well, find out yourself:
http://bugs.kde.org/weekly-bug-summary.cgi
Actually, considering the size of the project, being released with 2000 bugs isn't that bad at all :)
But with 4884+ is...
More bug crushing (dualie delete) days before each alpha/beta/rc release?
But what about those 4468+ wishes and 362+ most wanted features?
Halving those might be a huge enought goal?
"Most wanted" because 2 people voted for it? Only 197 wishes are wanted by at least 3 people and only 73 by at least 5 people with the top item being wished by 73 people.
Sure sure, but as you might notice, I just took the numbers bugs.kde.org gave, why is that bad? Sure it might make it look bad by numbers, but that is what people see on it.
Forget my mention about them if that bugs so much. ;-)
Sure the item that matters is to get bugs fixed before other things start to get done on those wishes/feature lists... Nothing wrong with that as that is the most sane way to do them.
4,890 bugs in KDE, wow, never knew KDE was that buggy!
I think an acceptable release number would be under 2,000, man this new release schedule is messed up, I don't see how KDE developers can possibly get the project in the sub 2,000 area by December, I defintely hope it's delayed.
What strikes me even more is that KDE isn't beating the tide of bugs, each week, it seems that more bugs are added than actually getting fixed. I really hope that most are duplicates.
I think KDE needs a MUCH LONGER TESTING AND FREEZE PERIOD. In fact, KDE doesen't even have a freeze period anymore, now you can add new features even during an RC as long as it has approval. Ugh... not good.
Jesus Christ almost 5,000 bugs, don't tell me that that's a small number for a project KDE's size, because it isn't! And, I defintely do not want KDE to ship with 5,000 bugs.
It's an alpha, jackass.
Remember that nearly half of those bugs are with Konqueror, and most of those bugs are khtml/kjs bugs. A modern HTML/CSS/javascript rendering engine is very complex, and takes years to perfect.
Remember that a bug, can be like that:
1) After loggin in, kde erased all my data.
or...
2) When I was returning from gym, I crossed the street very quickly, and after that, but only after that, when started konqueror from clicking a URL in KMail, and an evil javascript resized the window, then, a font doesn't displays at 2px size, but 3px.
What I *mean*:
When you erase all duplicated bugs, feature requests, silly render bugs in konqueror, typos in GUI's messages, etc., you will find that only 15, or 20 bugs are really serious, and need to be fixed before an stable release.
(Sorry for my english, it's very very late, and I need to sleep)
You "just don't understand". Software has bugs, that is a fact. KDE-folks could use the next 10 years killing bugs, and the software would still have bugs! New ones would get reported all the time. If you want to wait 'till KDE has 0 bugs, you would be in for a LONG wait!
And those are reported bugs. Whether the bug is reported or not, they would still be bugs. Some piece of code might seem bug-free because there are no bug-reports, but reality would be vastly different. Thankfully KDE has active users that report bugs when they see them. That's why the number seems so high.
And how many of those bugs really apply? How many are from old versions? How many are from packaging-errors? How many are fixed in CVS but are simply not closed for one reason or the other? How many are user-errors?
It never ceases to amaze me how totally ignorant and uninformed individuals love to just spout their mouth and state their opinion on a subject that they know nothing about, and actually expect people to do it. If you knew anything about programming and about the nature of the bugs in KDE you would know that of the "almost 5000 bugs", many are duplicates, once in a blue moon or non-critical errors, and so on. And the number of bugs has has increased dramatically in the past week as an alpha was just released, thus its expected to be buggy, and because a lot more people will use an alpha than just normal head, the number of bugs found will obviously stay high for next few weeks.
And you also obviously have no idea about the release schedule. You can't just add features during the RC period. Even before the RC 1 release non-trivial CVS commits have to be aapproved by at least one other developer. Catch a clue, and stop trolling
Which version of Windows had 60,000 bugs when it shipped?
Also, like what you're doing here, many people spouted that statistic without understanding it. I think 70% of that figure were non-critical UI bugs, like certain widgets not getting erased properly, etc.
The fact that so many bugs are reported is the most important thing you can take home from that statistic. If that's still not good enough for you then you should try and fix a few. If you can't do C++, not all bugs require programming experience.
Yup, Windows 2000 shipped with 63,000 bugs according to an internal leaked memo from Microsoft.
http://www.cnn.com/2000/TECH/computing/02/17/windows.2000/
And Win2k has been the most stable version of Windows in a long time. WinXP and 98 were much less table out of the box, but later patches fixed that.
Ummm... Windows includes the GUI + operating system. KDE= GUI only. Don't compare apples to oranges. Or Windows to KDE.
Actually KDE = GUI+(some)Applications. Still useless, but more than the GUI.
Actually you compared Windows to KDE, I was just using Windows to illustrate my point; I had no intention to draw comparison. If you missed my point you should read my post again.
I think another thing most of you are forgetting is exactly who the bug applies to. For example, if there is a bug that only affects FreeBSD running on Athlon processors, then it's counted in the overall KDE bug count, not as a specific quirk. You'd be amazed how many bugs only affect about 5% of all users. A prime example would be a glitch in the rendering engine of Intel i815 chipsets that causes a some GUI's to be rendered incorrectly. Even though it only affects KDE (not Gtk apps) when ran on the i815 chipset, and it's actually a fault in the hardware, it's still listed as as open bug for the XFree86 project.
How the hell did any of you pass your english classes?
1. never said KDE should have 0 bugs, all I said that 3.2 should have under 2,000 bugs, in MY opinion jackass! I know it's an alpha and I didn't complaina bout the bugs in teh alpha all I said that by the time the release cycle is over and KDE 3.2 is released I defitnely do not want KDE to have 5,000 bugs. My biggest complaint is that KDE is not beating the tide and I can't imagine how they will make a release with even 2,500 bugs.
2. I know there are many duplicates and irrelevant bugs, but those usually get closed pretty quickly.
3. I know Windows was released with 63,000 bugs but in reality that is not as relevant. Microsoft has over 97 times the marketshare KDE has and it is bound to have many more known bugs. In addition, they have people who's actual job is testing the software and not just users and developers, so probably they're peole tend to find more relevant and just plain more bugs. Windows is also an ENTIRE OPERATING SYSTEM WITH A KERNEL DRIVERS, A DE, MEDIA PLAYER, MOVIE MAKER, CD BURNER, BROWSER, FILE MANAGER ETC. Therefore' it isn't fair to make this comparrison. I don't think a Linux desktop would really have a lower amount of bugs.
4. Also, I NEVER said you could just add new features in RCs, I mentioned clearly that teh feature sneed approval.
5. Please everyone here, please stop attacking me for a reason, I do have a clue and quite a good one, I've been involved in Linux for more than 5 years now.
The only valid point about releasing KDE 3.2 with 5,000 bugs is that many could be for oter architectures. Again, I nver said it was unacceptable for an alpha to have 5,000 bugs, I only said it is unacceptable to release it and I said this because KDE doe snot seem to be beating the tide of new bugs.
Also don't bother to make the stupid point saying "no developers manage to beat the tide" this is simply not true. For example the Galeon project has been able to beat the tide losing over 220 bugs and fixing 279 in just the past week.
Also, don't say that GNOME has more duplicates because they also use Bugzilla and there is no reason to believe this.
Also, remember that Galeon isn't even part of GNOME anymore.
-277
Remember, Galeon is not part of the GNOME project so nobody really uses it. This means it has a lot of bugs that have not been found.
Of course it is used. I know users who actually prefer Galeon as a browser in KDE.
i think most of them switched to MozillaFirebird.
Galeon has a bleak feature. NetCraft said that they are dying. With the recent introduction of Epiphany in GNOME 2.4, Galeon is all but dead corpse already.
Once again, most Konqueror bugs are related to khtml/kjs..
The galeon browser uses gecko, and does not maintain their own rendering engine.
Compare something more realistic like Mozilla. They had 213 new bugs for their "browser product" over the last week, and 34 bugs fixed/closed.
Galeon, is not part of GNOME but the poll on GNOMEDESKTOP.org clearly shows tat it is the prffered browser for GNOME still. It has many users, a dozen of new bugs reported each week and still actively developed. Galeon Developers only and their not even part of the GNOME project fix over 275 bugs each week, KDE with all of it's components many times won't fix this many, so don't say Galeon is not being heavily worked on or being used.
Also, the fact that Galeon uses Gecko or not is besides the point.
> Galeon Developers only and their not even part of the GNOME project
You can barely spell, yet you attack other people's English. Go away, troll.
You really showed how stupid you are on that one. Galeon is merely a "front-end" to Mozilla. They don't even have to write/maintain the most complex part - the rendering engine. So obviously they it will be shipped with a minimal ammount of bugs, seeing how it doesn't do *any* of the dirty work. If you want a much more fair comparision, compare it to Mozilla. It's an excellent browser, but only God knows how many bugs are in it...
> Please everyone here, please stop attacking me for a reason,
Fine, we'll attack you for no reason.
> I do have a clue and quite a good one, I've been involved in Linux for more than 5 years now.
Oh my. Looks like we've found ourselves a bona fide computer scientist here folks!
I can spell very well, it's just that unlike you I don't care so much about some message on a news item and why should I really, it won't affet my life in any big way.
I see you again have trouble READING and COMPREHENDING because I was not attacking anybody's english, I was attacking everybody's poor COMPREHENSION ability. Now, what does my spelling have to do with the issue at hand anyway. Nothing, stick to the topic and stop peronally attacking me. When you do this I consider that you fail to come up with anything to counter my arguement and so have decided to change the subject and launch personal attacks.
Again, I never said I was a computer scientist or anything like that, stop MINTERPRETING what I say again. All I said was that I am more than familiar with KDE, GNOME and most other important Linux projects, so don't think you're speaking to a newbie. I also, did have one year of C++, but that's about it.
Now, any of you care to stop making smart ass unrelated remarks and either post a comment about what I'm saying or don't post at all. Your lack of focus is one reason that makes it so hard for you to understand my post
BTW: I actually am in English Honors so, please stop making dumb assumtions.
affet -> affect
COMPREHENSION ability -> ability to comprehend (or comprehensional ability if you want, but that sounds ugly).
peronally -> personally
arguement -> argument
MINTERPRETING -> MISINTERPRETING
assumtions -> assumptions
Sorry, but I couldn't resist! ;-) And people have addressed the points you've made. I agree that some people were aggressive, but honestly, you brought it on yourself!
Don't you see what you're doing. Your, not posting on the topic as I just asked in hope that some people would actually want to discuss the topic. Really, this si very immature of all of you.
Ed, your post is a waste of space the DOt, it does not help or say anything new in any way. Correcting my spelling mistakes completely out of context is pointless and only proves that you must have low self esteem, to feel the need to do this when you clearly understand what my idea was and what each word should be. If at least, you rewrote my entire post with corrections, that might have been something.
You too have problem reading, the topic was never "Adi's spelling mistakes" it was about the number of bugs in KDE< bugs in software and the pace at which KDE is fixing them.
MAybe you'r smart, I don't know, but doing this IMo shows your a dumbass who can't stay on topic or come up with good arguements. It makes me feel you don't understand the idea or have nothing to say and this applies to almost everyone else in this thread.
Everyone had already answered your points, so there was no sense in duplicating that. Since you seemed not to notice, a summary might be: your original post was alarmist and defeatist, and no terribly grounded in reality.
What noone did was call you on your "i can spell very well" comment in a post riddled with spelling mistakes (and be fair, I did apologise in advance, and it was at least partly tongue in cheek!)
Look, you really should calm down a bit. I understand that you're upset you weren't taken more seriously, but get a grip! Go out and have a beer, find a woman, but a dictionary (sorry! just can't help myself it seems!) ;-)
Oh dear - that's embarassing ... I meant to say BUY a dictionary of course. ;-)
The point is that you are a liar and a troll, that all your points have already been adequately addressed except for your obviously silly knee-jerk emotional reactions.
You did not answer:
http://tinyurl.com/n9jf
and
http://tinyurl.com/n9jk
You simply went on about my spelling mistakes which was far from the topic. Also, YES I CAN SPELL VERY WELL, on the STAR test I was far above average achieving 99%, the fact that I type fast and do never go back to correct mistakes is entirely different. There is a significant difference between spelling well and being able to spell well.
Also how am I a troll anda liar? Please enlighten me and actually answer my posts instead of blabbering about how the sky is blue.