> Our development model has changed over time. While this was > true in the early days, we now contribute our changes early > to the mainline kernel. I agree that while things are in the > "what do you think" state, we mostly use the more convenient > communication "go to Martin's office and talk it over" over > the less convenient one "mailing list".
Not having the privilege of direct access to Martin's office (although the few times I've been able to speak with him on the phone, he's had good suggestions and the discussion has been valuable), I can't argue whether it's more convenient or not. It's not an option for the rest of us, so I would argue that it really doesn't scale as a method of working together. > That's exactly what should happen. But it is the > responsibility of the author do drive that discussion. I did > not see that happen in this case, and it is exactly the same > what I need to do here in Böblingen to get my code accepted by Martin. I think I would politely disagree. If someone posts something and asks for comments in the appropriate public forum, that is IMHO sufficient pursuit. If others have difficulties with the approach or comments, I'd observe that it's the responsibility of the person with the question to contact the author and discuss -- preferably in public if the issue is sufficiently important -- why it's not the best solution. Most of us are mature enough to at least hear someone out if they have a differing opinion on a solution. Maybe I'm still a bit of an idealist, but ultimately, though, in my mind, running code wins. We can always refine a solution - and we will - with the peer review process, but at the end of the day, there's the fundamental difference between something that works, and kibitzers. A particular solution may not be perfect, but something that works is (usually) better than nothing, and the better technical solution will be determined by field use and feedback, not be desk checking. It's those darn edge cases where we have problems. 8-) > Martin is not maintainer of this architecture because he is > IBMer. He is maintainer, because he does do a good job on > reviewing other people's work and giving constructive advices > and because he is really dedicated enough to work overtime > and do-what-it-takes to get to the best possible soloution. > Your statement heavily implies that Martin does make a > difference between contributions from IBM and contributions > from elsewhere. I don't see that: I frequently do discussion > with him about ideas and their implementation, and I hear > "don't do that" or "solve it differently" approximatly once a week. > I think it is unfair to demand a different maintainer after > failing to drive the discussion of subject driver towards integration. I don't want to seem to disparage Martin -- as you say, he's doing a nasty, thankless job extremely well, and if I seemed to imply otherwise, then please accept my apologies. What I do see is that the process of his job *is* somewhat impeded by some of the restrictions he is bound by by virtue of him being an IBM employee. Not being able to look at and comment on proposed implementations freely is a problem. Not being able to work with the consolidated repository because IBM Legal can't figure out how to build the right Chinese walls to keep his job safe is a problem. Being the "only" pipe into the arch/s390 and arch/s390x trees is eventually going to be a problem in terms of succession and workload. As he *is* the only gatekeeper for this architecture, this seems like something that we -- as a whole community -- should address as a whole. One option on the table is to share the responsibility a bit -- it's gotta be a major hassle for Martin's free time to be the only guy doing this. If the option of using a external organization like SHARE or WAVV would solve some of the legal problems, then it's worth the discussion. At least IBM Legal has some history with those organizations, and has a past model of how to deal with them. We're not starting from scratch there. For me, that's the business of what we're doing here in this exchange -- there is a percieved problem present. This discussion is supposed to explore options to solving the problem. It's not about Martin failing to do something, it's about solving the problem, however perceived or real. I happen to think that a non-IBM voice would be helpful in partially solving this problem; you have a different perspective. Let's discuss it and find some middle ground. -- db ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
