On Friday 06 August 2010 17:15:46 Egbert Eich wrote:
On Thu, Aug 05, 2010 at 02:11:34PM +0530, Sankar P wrote:
On Wed, Aug 4, 2010 at 3:35 AM, Jos Poortvliet
wrote: On Tuesday 03 August 2010 14:34:13 Pavol Rusnak wrote:
On 08/03/2010 12:28 PM, Egbert Eich wrote:
- Had a longer phone discussion about the openSUSE strategy discussion, the sentiments of the SUSE Labs developers and how to overcome the disconnect of people at SUSE Labs (and other divisions inside SUSE not directly involved on openSUSE) and the openSUSE project.
I'm looking forward to a written summary of this discussion :-)
That would indeed be interesting. The SUSE labs developers are supposed to be community leaders, after all, and their input is/should therefor be valuable...
Right :) See below.
With so much noise in the project mailing list, I will be surprised if any of the kernel folks were even following the threads, instead of mark-all-as-read. I will also be eager to know what their opinion is.
First of all, interesting, bloody interesting read. And I understand the labs ppl very well - will discuss this with them in 2 weeks when I hope to be at their offices...
This is exactly the point and actully what's happening. It is one of the issues I brought up talking to AJ: there are simply too many communication channels (fora, MLs, IRC, ...) for those people to follow. Some kernel folks read kernel@ but this is a topic ML and sufficiently focussed and low volume to follow. Most people from the Labs are active in their own communities (kernel, gcc, samba as example) already while at the same time they working for Novell, where their duties cover a assignments totally unrelated to openSUSE. These two areas alone already consume a great amount of time. Thus participating in yet another community is nothing that seem to be very feasable.
AJ mentioned that to be informed about what's going on in the project people should at least read news@ and announce@ - still these are hardly lists where one can chime into an ongoing discussion. I also explained to AJ about my impression that several people I met there seem to feel a discomfort to voice their opinions on an opensuse ML being Novell employees at the same time. When I was talking to people at the Labs Conference and asked them why they did not bring up a certain topic on an openSUSE ML I heard more than once: 'Isn't this a discussion that should be had internally?'. (I strongly believe that this is a feeling that's shared by several other Novell employees working for OPS.) Indeed for someone working at a company like Novell for a while it may feel strange to discuss things with collegues in public: - Should one really openly 'flame' fellow Novell employees (AJ, Michael, coolo, the Boosters, ...) or decisions made by them? - What things can a Novell employee say in public? How much are the things one says influenced by internal knowledge that's not ment for public digestion?
I have this myself - I don't know where I stand being both a Novell employee and a community member. Something I want to discuss with ppl in person a bit anyway so I'll take it up with them too.
The main part of the conversation with AJ dealt with the feeling people had about the 'strategy discussion'. AJ was concerned that I had projected my own discomfort with it upon the people while I tried to assure him that this was not the case. I explained the sentiments: Most people in the Labs have a long history with a free software project. For those people a discussion about a strategy for the project seems to be rather strange thing: Free software projects live from the contributions by volunteers and thus are are meritocracies: the people who contribute get to determine the direction, the direction is determined by their contributions: people pick a topic, work on it, and only then present it and discuss it. From this discussion new aspects may arise, the contribution may get modified, extended etc and eventually (which however is not certain) a new feature will make it into a release. People who get to freely choose what they want to do are the most vibrant drivers of their topics. Therefore the direction of a free software project is never fully predetermied, rather it depends on: a. what topics are presently 'hot' on the street. b. who happens to be around to pick up such a topic and drive it. This understanding seems to be rather universally shared by all Labs people active in upstream projects. A project 'strategy' which predetermines from the outset in which general direction a project is going to evolve doesn't seem to fit into the picture. AJ explained to me one of the reasons for a 'strategy' was to answer the question 'why openSUSE?' ie to determine what distinquishes us from the others. I explained to AJ the sentiments expressed by some Labs developers: To them openSUSE is just the optimum between 'most current' and 'stable and easy to use' which to them is a sufficient distinquishing factor already. I reported to AJ also that I had invited all people who are not comfortable with a strategy for openSUSE to participate in the relevant discussions. This however again touches on what I mentioned above.
This to a large part summarizes the discussion I had with AJ. A perfect recipe how to overcome the disconnect none of us could come up with at the call.
The strategy part I can, I think, explain to them. I've been trying to get more strategic thinking within the KDE community for ages, it's finally working a bit (more due to aseigo than me, btw) and I have strong opinions on the matter. And I'm of course right ;-) (gosh I sometimes wish I was arrogant enough to actually say (and mean) that without a smiley) Will discuss this more at the nurnberg offices and try to blog about that again at some point in the near future.
Maybe one step would be to bring the SUSE Labs conference and the openSUSE conference together, an idea that has been tossed around for quite a while already.
+1000 Ties within Novell itself, within the community and between the two should be improved in general. Having face to face meetings is one of the most (if not THE most) effective ways to do it. And I'll walk the talk here, as far as I can - as soon as I think I should start being useful in this community and have some confidence when it comes to knowing what is going on I will try to organize more face-to-face meetings. Or rather, help ppl organize them for me of course ;-)
The change in structure of the Labs conference especially that would be required seemed to have been the major obstacle that kept it from happening so far. Now since there is about a year before the next Labs conference there should be sufficient time to make this happen.
Cheers, Egbert.
Now go drink beer, it's weekend. In case of French ppl, drink wine. And in case you don't drink alcohol, either start now or skip a night's sleep, you'll feel about as hazy as when being drunk. Grtz Jos (who skipped a night of sleep, so please forgive my ramblings)