Hello there,

this postings comes from a former "KDE lover" who just converted to xubuntu ... the intent is

a) to move a meta discussion in https://bugs.kde.org/show_bug.cgi?id=265051 to an more appropriate place

b) add the view of some more people

c) maybe maybe produce some effects to the better of KDE.

I dont want to repeat all the things that were said in the bug ticket, but maybe point out some of the important issues:

- my comment #25 ... it basically annoys ANY user who cares that tickets sit there for months without any visible indication that something happens.
Tickets have TWO important functions - "issue tracking to ensure things get fixed" AND "communication between those who find and those who fix bugs". And it didnt help to learn that "the developers dont care about votes".

- basically all the excellent comments from Clemens, esp. #39.

Is this just the narrow view of a few; or does KDE in deed have a quality problem? To many code for too few developers ... and mindsets that value new functions way too much over stability of existing BASE infrastructure?!

btw: I am not really happy that I had to move to xfce.

Nonetheless I am not the person that waits for the boat to be full of water before leaving.

For those of you who speak german: KWin maintainer Martin Gräßlin has posted a very interesting and (presumably intentional) provocative Blogpost whose english title would be "The bugtracker lie".

Die Bugtracker Lüge

The essence is: The KDE Bugtracker (At least the parts that are related to his work, e.g. KWin and Plasma) receives too few bugreports that have the quality the developers need, if information is requested it isn't delivered by the reporter or other people. Duplicates however are the thing that is most annoying and counterprodictive. That results in the developers not noticing the few bugreports that actually are worth the database space they take up and bugs not being fixed.

Systems that should indicate if a bugreport is good (voting) are not effective or misused (as stated in the comments) and the KDE bugsquad is trying hard but cannot keep up with the amount of bugreports that keep pouring in.

Interesting point, hefeweiz3n.

But I wonder how other projects get around this problem as it must be universal...

toad wrote:Interesting point, hefeweiz3n.

But I wonder how other projects get around this problem as it must be universal...

I guess by ignoring bugs. Promising that they will be fixed in the next version of the tool.

The post from Martin is quite interesting. I agree that too many bad tickets are a big problem; and maybe it would be a good thing to actually restrict the ability to open up bugs. Something like a rating system; where one has to collect some credentials before he can submit bugs.

But on the other hand Martin is mentioning, that he thinks that the user wishes are completely out-of-sync with the development view; therefore he ignores them.

Final point: i do think this is also a question on "strategy". About the values that are "core" to the development team. A common understanding: is it more important that we create some new application in KDE-style; or is it more important to focus on other priorities?

If the bug-tracking system is useless it is only too understandable that it is ignored. I as a user find it cumbersome to say the least and am not surprised that duplicates are the norm. Add to that the bewildering choice of division and subdivisions, versions and sub-versions and the mess is complete.

My problem: I cannot think of a system that works and wonder at the same time how other projects cope.

Final point: i do think this is also a question on "strategy". About the values that are "core" to the development team. A common understanding: is it more important that we create some new application in KDE-style; or is it more important to focus on other priorities?

That is something for the devs to decide. If every Tom, **** and Harry thought they are God's gift to KDE and told them what to do I'm pretty sure that KDE would not exist.

Also, the devs have to get a certain amount of satisfaction from their work or I suppose they'd bugger off (who wouldn't).

Strategic planning is afaik decided upon during the KDE gathering and what happens after that is a rough implementation of an overall strategic goal.

edgue wrote:[...]
a) to move a meta discussion in https://bugs.kde.org/show_bug.cgi?id=265051 to an more appropriate place

nice

edgue wrote:Tickets have TWO important functions - "issue tracking to ensure things get fixed" AND "communication between those who find and those who fix bugs". And it didnt help to learn that "the developers dont care about votes".

I kinda agree there. Why kinda? See below.

edgue wrote:[...]
- basically all the excellent comments from Clemens, esp. #39.

I'd say that especially comment 39 is a real bad one. And writing that you find that one excellent rather underlines that you don't really understand what a bug tracker is for.
A bug tracker is not a place for soap operas.
Telling that you switched to something else or wrote on one of the dev-ml adds absolutely nothing of information to the bug report.
In contrast it degrades the bug report even more.
Making it harder to fix it!

There is no need of 100 people to say they also have a bug, use voting for that. There is no need for people mentioning other bugs in detail there, create a new bug and if you think that it might be related mention the bug number.

What is a bug report for you might ask at this point?
Well you write what problem you have and then answer questions. That is what it boils down to. If no questions are asked you can mention what versions are affected, e.g. still in 4.6.3.

So looking at that report there are numerous comments that add nothing but instead make the report more useless. Look at the report yourself. How long it is by now. You can even try to find the relevant information.

edgue wrote:Is this just the narrow view of a few; or does KDE in deed have a quality problem? To many code for too few developers ... and mindsets that value new functions way too much over stability of existing BASE infrastructure?!

Well you are one of the problems yourself, as I have pointed out. Commenting is about adding value to a report, your post #25 does not do that. What you write in 25 should either be discussed in a forum or a mailing list. I don't know which ml would be suitable though. And in total I think roughly two thirds or even more of your comments on that report -- judging from the number of lines -- are totally useless.

So if you want to improve something, then please write good bug reports and keep soap operas off the tracker.

edgue wrote:Nonetheless I am not the person that waits for the boat to be full of water before leaving.

No, instead -- as it looks like -- you fill in more water ...

And before I forget it, ml are certainly the wrong place to advertise one's pet bug. If that would work it would only end up in mls being as destroyed and painful to work with as bko already is in some areas.

Yeah it sucks getting no replies and I was pis*ed myself some months ago when not getting needed feedback on some review requests, though you have to keep in mind that people aren't here to answer whatever you ask.

First not everyone could answer and second not everyone who can has the correct hardware, has the time or has the motivation. Having to read report like these with so much cr*p in them does not add motivation as you might have guessed ...

Before I forget it.
What do you think should be the solution?

E.g. take the following scenario with two displays, A with 1000x500 px and B with 1200x500 px.
You use A and your panel has a size of 1000x50. Now you switch to B and I suppose you want the panel to have a new size of 1200x50. So its height or width -- depending if the panel is on the bottom/top or left/right -- should be assumed as fixed. [1]

Now what should happen in the following case?
You are on B and set the panel to 1100x50 and then go to A.

• Should it be downsized to the whole width 1000x50 which in turn could lead to 1200x50 when moving back to B or
• should it keep the aspect ratio of the width, i.e. 1000/1200 * 1100 = 916,67.

Another way would be to store the settings for each display and to then use that. If nothing had been stored the system would try to find a sensible default. The problem there is that if the user changed something for say display A, makeing it use the whole width they would have to do the same for display B.

As you see there are different ways to react upon and you can be sure if one of these is chosen there will be reports propagating the other. Adding options for that would in turn make the ui a lot more complicated, as the initial problem would have to be described etc.

[1] I am quite sure there are people around who would disagree.

So, lets try to answer the questions raised:

* Starting with the easiest one: I agree that there is more than one way to do it; a first guess might be: make sure that it works as it did before KDE 4.6; and that KDE works the same way as any decent graphical desktop. Especially when talking about one display all the time - make sure I dont have to go in and maximize that manually. I agree, a KDE solution might require some preferences somewhere to make that configurable ... but hey, a minimal auto-maximize was there with 4.5; wasnt it?

* About the "bad comment #39": well, his statement is simply expressing my underlying concern: there is a energetic user who contributed massively by testing & writing up bug tickets; and none of that had any visible effect for him. He felt ignored anywhere he turned to.

* But, ah right, that is misuse of the bugtracker. And you are right. But: already the first entry that opened up that ticket described in clear words the problem that MANY users affect. What happened? Nothing. Other users came in; reported that they have the same problem. What happened? Nothing. Instead, KDE 4.61 and 4.62 came out ... and didnt resolve the problem. Then at some point, users get angry. That is when things start to deteriorate.

It is pretty simple: when a bug tracker isnt used to track bugs resp. the solutions to bugs ... it turns into something else. Namely a medium for affected users. Is that a good thing? Probably not. But to be expected.
And "easy to fix". A simple comment early on: "this is on my radar, I will work on it" would be good enough. As pointed out before: I dont mind when a developer says "Sorry, it will take me 3 months to fix it" ... if one just KNOWS that someone is intending to fix it then.

In other words: most users DO understand what belongs into bug tickets. And as long as the system really functions as a "bug tracker" (namely: it tracks the progress of bugs and their resolution) they will try there best to be productive. But - when it turns out that the bug tracker is just a huge pile of users problems that nobody seems to care about ... it is no longer a "bug tracker" for those who write problems. So dont be too surprised if people start chatting instead of just silently sitting there and hoping that some future release x.y.z solves more problems that it brings in.

edgue wrote:* Starting with the easiest one: I agree that there is more than one way to do it; a first guess might be: make sure that it works as it did before KDE 4.6; and that KDE works the same way as any decent graphical desktop. Especially when talking about one display all the time - make sure I dont have to go in and maximize that manually. I agree, a KDE solution might require some preferences somewhere to make that configurable ... but hey, a minimal auto-maximize was there with 4.5; wasnt it?

I have no clue if something was in 4.5, as I am not affected by this bug.
And no you totally misunderstood me. I am against a configuration option here as it would only complicate things.

edgue wrote:* About the "bad comment #39": well, his statement is simply expressing my underlying concern: there is a energetic user who contributed massively by testing & writing up bug tickets; and none of that had any visible effect for him. He felt ignored anywhere he turned to.

I miss to see where the author of #39 contributed anything other than marking one duplicate and a still in ... to this report.
His first comment should rather be a wish, but ok. Then he goes on arguing how he perceives things are getting worse also #13 and #21 are classic ...
500+ words on ot stuff, prejudices, ...
Great contribution.

* But, ah right, that is misuse of the bugtracker. And you are right. But: already the first entry that opened up that ticket described in clear words the problem that MANY users affect. What happened? Nothing. Other users came in; reported that they have the same problem. What happened? Nothing. Instead, KDE 4.61 and 4.62 came out ... and didnt resolve the problem. Then at some point, users get angry. That is when things start to deteriorate.

At some point? Less than one month!

It is pretty simple: when a bug tracker isnt used to track bugs resp. the solutions to bugs ... it turns into something else. Namely a medium for affected users.

I fail to see how bko failed to _track_ this bug.
What I see are some users failing to use bko correctly and some still keep adding ot stuff to that report...

And "easy to fix". A simple comment early on: "this is on my radar, I will work on it" would be good enough. As pointed out before: I dont mind when a developer says "Sorry, it will take me 3 months to fix it" ... if one just KNOWS that someone is intending to fix it then.

Well yeah would be nice. Yet with the count of bugs this is not really realistic. Also I never write that something is on my rader on any report.
Why? Simply because I don't know when I have the time to look at it and if I might be able to fix it.
Further if you look how many bugs have been reported on plasma -- many of these duplicates, reports with not enough information, already fixed ... -- it is a little adventurous to expect people taking over one's main bug, even if they don't know when they'll have time to look into it.

In other words: most users DO understand what belongs into bug tickets. And as long as the system really functions as a "bug tracker" (namely: it tracks the progress of bugs and their resolution) they will try there best to be productive. But - when it turns out that the bug tracker is just a huge pile of users problems that nobody seems to care about ... it is no longer a "bug tracker" for those who write problems. So dont be too surprised if people start chatting instead of just silently sitting there and hoping that some future release x.y.z solves more problems that it brings in.

Aha. You are totally wrong as had been pointed out in the blog entry by Martin. Many reports are duplicates which the user was informed of. Further many reports lack information.
I asked on many reports for more information and got no reply.
Also great is if more and more stuff is put into a bug report or a wish. Yeah these specific users know how to use bko ... not.

edgue wrote:* About the "bad comment #39": well, his statement is simply expressing my underlying concern: there is a energetic user who contributed massively by testing & writing up bug tickets; and none of that had any visible effect for him. He felt ignored anywhere he turned to.

I took the time to read the comment in question. It's not a good comment at all, in fact it's off topic and particularly bad. People who post in bug trackers should be well aware of KDE's Code of Conduct, which should not govern just this forum but any developer-user interaction Comment 39 is a clear violation of that code.

einar wrote:I took the time to read the comment in question. It's not a good comment at all, in fact it's off topic and particularly bad.

I think the point is: that guy operated within the KDE code of conduct for a quite a while. In exactly the way developers are looking to users: he tested early releases and provided feedback. And what I get from reading his statements is ... it didnt make any difference.

But alright, if the audience agrees that the "bad thing" here is to violate the KDE code of conduct; and not the fact that the KDE community lost a person who was willing to support the project; what can I say then.

mat69 wrote:I asked on many reports for more information and got no reply.
Also great is if more and more stuff is put into a bug report or a wish. Yeah these specific users know how to use bko ... not.

Then something needs to change dramatically.

If there is no feedback in 10 days .. close the ticket.
Have some counter that tracks user ids.
Same guy having two tickets closed on him?
Dont allow him to write up another ticket.

Yeah, that one might imply a more rigid user management.

The point I am trying to make is: I was sitting in front of my computer for more than two months; and every other login I had to maximize the damn panel. I checked the data base; checked for duplicates and even commented on possible duplicates to inform the users ... and what did happen?

Nada. Zip. Nothing.

I vote for the bug; to be told that the KDE developers dont care about the voting system. Why are the "votes" still in place if that is so?

Long story short: I think you folks have a serious problem. And the problem is not only rooted in the fact that some KDE users get impatient and frustrated at some point.

edgue wrote:But alright, if the audience agrees that the "bad thing" here is to violate the KDE code of conduct

The Code of Conduct was made precisely to prevent these kinds of outbursts, even if the reasons were understandable, and because some overly frustrated people went too far with their "criticism".

I'm sorry, but saying "I'm switching" does not help anyone: neither the reporter, nor the developer or the users.

einar wrote:saying "I'm switching" does not help anyone: neither the reporter, nor the developer or the users.

Well, it does free the air waves

I always welcome it when people use it 'cos in the end that is what they want to do and it is probably better for all concerned. My opinion: good riddance.

Of course you're right - it is better if such individuals kept their own egos in check and realised that the world will actually continue turning without them.

toad wrote:Of course you're right - it is better if such individuals kept their own egos in check and realised that the world will actually continue turning without them.

There is some truth in your statement, but on the other hand ... the original bug is now "resolved" because it was closed as "invalid". Of course, that reaction is perfectly grown up, mature and ego-free.

Luckily it didn't take me long to find at least one duplicate that addresses the same problem (guess what; open for months; no sign of activity) ... so maybe, some day, some release this problem will be fixed.

Reminds on the thing I so much like with gnome, the CADT model: http://www.jwz.org/doc/cadt.html

Final thought on my way out ... a guy from marketing once told me: one disappointed customer can easily outweigh 10 to 20 happy customers ... but well, that are user problems. Obviously there are too many users anyway; how else could one explain that there are so many open tickets that the developers cant handle them any more.