Felix-Nicolai Müller <fnmueller@opensuse.org> writes:
Andreas Jaeger schrieb: | Reading Felix-Nicolai's blog post, I'd like to get your input on the | following statement from him: Sure thing | | I believe that openSUSE is supposed to be a community based and | community driven product. While many devs work for Novell and do great | work, openSUSE does also depend on its many users. So they should get a | hear. It is not OK to only raise this issue on a factory mailinglist - a | mailinglist a regular user will never see. | | * factory was setup as list for future development of the openSUSE | distribution. Why is that not working? Where would you expect to | have general discussion about the direction of the distribution? It is not working because this (and issues like this one) is not discussed in spots of the internet that are reachable by the regular users. Most don't subscribe to the list for future openSUSE development, but go to the forums. This would have been a good place to ask what users expect and want. It is perfectly fine to discuss this on this or the -kde ML amongst the devs. But don't expect this ML to reflect the broad scope of users that make openSUSE what it is - one of the most liked Distros in the world with many "support branchens" and spreading the good word about Novell too. There is no way you can ay at the end (doing it this way) that you asked the users.
You never will be able to reach everybody, even if you go to the forums. Should we really vote for every decisions? Or just for critical ones and who defines critical? An option would be if somebody does a vote in a neutral way (like Larry started) and then represents the outcome - but do you think everybody on this list moving to the forums is feasible?
The rather weak response on my blog either shows that the interest in KDE4 is rather low, that people feel it has been all decided anyways, that a blog is not the right form to reach the regular user or that it has not been publicized enough. I don't know which of these reasons is true (yet).
Neither do I. Your blog is aggregated on planetsuse and therefore should reach quite some folks...
However, how the KDE4 discussion developed indicates that there is a bigger and more general problem we will have to deal with asap: lack of communication and miscommunication. Let me make this perfectly clear: This is OSS, so the attitude of the KDE devs that everyone has to either like it or shut up and live with it (which is my personal experience when I hear things like that there are "just some community members beating up again on Beineri" (which is not true, I would rather say his nervous costume is rather thin)) is not debatable, at least if you believe that the mainly by Novell paid KDE devs really are OSS like (I am not saying here that this is my impression, but this is what some might be inclined to debate). However, if they do what they want and "ram KDE4 down the users' throat" they must be able to take the heat. And this is what happened. It was silently decided in 11.0 to only push KDE4. It was (as for as I experienced it) by user intervention that KDE3 was included. Where would openSUSE be today if that had not happened? Would the customers be happier or less happy? I somehow doubt taking away choice would have made them happier. The KDE devs and any developer group (as in OSS) can do what they want, but they also have a responsibillity to take a stand in front of the user for their action. Either that or the users decide to do what they want and execute their freedom of choice and use a different product (remember: "rammed down ..."). And do not expect users (who are all contributers - even if it just for the cause to spread Linux and / or openSUSE) to help products they do not understand or like. Even regular users can have a major impact factor. It is geruilla marketing. Now, communication has already improved by the last painful KDE4 discussion for 11.0. At least people on certain MLs are now asked what they want. However, this does not serve the "normal" user. Maybe it would be wise to split the discussion between devs and testers (who live on MLs) and the users (who live in forums). That way devs could be
It's not as black and white as you say - there are users on mailing lists and developers on the forum as well...
presented with an overview of what regular users want - just like the summary I tried to compile on my blog - and discuss all that internally (but still openly and transparently to the interested public) on a less heated ML. That way, even if users complain later on, you can tell them that you !really! gave them the opportunity to have a say. This is still not the case now.
As I said, the process has already be improved, but even when opening the thread on KDE4 for 11.1 it was not made clear how the process to reach those options given was achieved. I now know what the devs think to be legit and possible but I do not know who said what when and why. This points directly to the general problem and what I describe as missing transparency. I want to know the true reasons, which I ultimately only can if I can get a transcript of the discussion how
True reasons? There's nothing to hide: Check how many bugs the kde-maintainers@suse.de list has assigned, see the work for KDE3 and KDE4, assume a fixed number of Novell engineers working on KDE - and then it's an evaluation what to do: Support both KDE 3 and KDE 4? Or spend all energy on KDE 4? Or be the only current distribution that works only on KDE3? Currently 40 % of our KDE users are KDE 4 users, so going forward KDE 4 is a valid option and I personally would rather have a perfect KDE 4 than a less perfect KDE 4 and KDE 3 support. You will argue - and you're right - that even if we drop KDE 3, KDE 4 will not become perfect but long term that's the bet we want to do.
those options were reached. To give you a better example: I filed a bug report against the non-alphabetical ordering of DEs in 11.0 installer. It was closed with the comment "This was well thought through by us". Sadly the bug was removed from bugzilla (at least I can't find it no more), but yaloki remembers the bug including the response. When I get such a response for making myself a fool for creating bug reports I feel entitled to know the anonymous mass of "us" and the definition of "well thought through". So basically: Who said what when and where. In order to understand why devs or any group wants something and sees certain choices a short summary of why the think so and what they think is not enough. One needs to see the process how the decision was reached. It would help all groups involved to at least have the opportunity to take that chance.
Could you explain in a practical way how to do it with decisions like this one? There are team and project meetings where these things are discussed and decided in general.
Sometimes I am under the impression that how things are done are decided in a 15-30 minute talk in some office. I would be delighted to be proven wrong.
Who is a contributer in OSS? Everybody! While having more devs would be nice, many many people contribute by pushing OSS, the idea, helping friends, people on IRC or in forums. Some pay, some spread the word, some code. Measuring codelines is not what can be used to measure contribution. The answer "what do you do? Why don't you do it yourself, it is OSS" is not a generally appropriate answer. Such an answer cuts many people short. Without customers and people pushing OSS, devs would be nothing (except to be happy for themselves, but then again, they would probably be lacking input from other devs). No one dev alone can cope with something like a bigger than little OSS project.
Are you actively involved with an Open Source Project besides openSUSE? I'm participating in some and I often see and hear "code wins" - and unfortunately only seldom the users are asked.
| | * openSUSE is community based and community driven - but we have Novell | engineers doing engineering work with limited time and resources. | | Community means not only discussion but also contribution - and I've | heard many requests about KDE3 and KDE4 but so far only saw Carlos | stepping up and saying "KDE3 is so important for me that I create a | KDE3 LiveCD for openSUSE 11.0". Discussion can be very helpful contribution. | | It's good to get input and help with prioritization but how can we get | more people working on KDE to maintain KDE3 and to improve KDE4? KDE3 will die. I don't expect anyone to step up to maintain KDE3 as those who could due to knowledge and time and money will probably wisely invest the effort in KDE4. I expect those who don't find KDE4 to be attractive to change DE. For instance: I am in the phase of changing to XFCE. Just as a personal comment: From a technical standpoint I could step up and help maintain KDE3. But I am not willing to invest precious time in a deemed dead project (as this is always heavily emphasized by current maintainers) as well as I can help better in other areas, like usability (a straw openSUSE or Novell folks just would have to pick). And this is where transparency is of matter again. I find it really hard to produce useful input on usability issues because I am not supplied with the intel I need. E.g. the UX group made a questionnaire and some research on the installer and while making public that the openSUSE 11.0 installer is the best ever and the easiest and so forth the actual numbers (and I would need the setup of the actual experiment) can only be seen if you have access to the internal Novell network. How much BS, traps and effort do I have to go through in order to be _able_ to contribute? And how can it be expected from me to contribute
There shouldn't be any BS and traps and little effort - but unfortunately in some areas it's still too much effort.
in all areas (maintain KDE3, talk to the community and cope with usabillity)? How much can and do you want to ask for? It's not like this is my entire life or if I'd get money out of this to live from.
Anybody can only contribute as much time as they have - you can do a lot with an odd hour here and there IMO.
I hope this answers you questions. I hope you can answer mine.
Not fair, you wrote much more than I did ;) You ask many questions - some of them seem to me rhetoric. If you like to have an answer to ones that I missed, please ask clearly again, Andreas -- Andreas Jaeger, Director Platform / openSUSE, aj@suse.de SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg) Maxfeldstr. 5, 90409 Nürnberg, Germany GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126