I think I ended up replying only to Ted last night. ---------- Forwarded message ---------- From: Swapnil Daingade <swapnil.daing...@gmail.com> Date: Wed, Jun 7, 2017 at 1:07 AM Subject: Re: Is Apache Myriad dead? To: Ted Dunning <ted.dunn...@gmail.com>
Please see inline On Mon, Jun 5, 2017 at 11:28 PM, Ted Dunning <ted.dunn...@gmail.com> wrote: > > > On Tue, Jun 6, 2017 at 12:56 AM, Swapnil Daingade < > swapnil.daing...@gmail.com> wrote: > >> >> The problem is that there is essentially no real community that is >> happening. >> >> retiring doesn't help that >> > > The core problem here is lack of a viable PMC. A PMC has to have 3 active > members at any given point. Typically this requires about 8 live members. > Myriad is wildly short of that and thus will have serious problems doing > any releases. > Swapnil: I don't think the existing committers are any less passionate about Myriad that they were earlier. The real question for me is what features do we work on next. Will said he had ideas to propose. We would all like to hear his (or anyone else’s) ideas than this talk of retiring. I don't think it will be difficult to regroup. > > >> >> >> None of the engineers previously working on this will be working on >> this now. And that sort of situation isn't going to change. >> >> Events at MapR contributed to this situation. MapR scaled back its >> involvement in Myriad and all its committers left. >> > > Well, that is one way to look at it. > > On the other hand, if you actually were involved in the situations, you > would know that none of the committers left because they didn't get to work > on Myriad as part of their day jobs, nor did any of them feel enough > attachment to work after hours (as I do on my projects), nor did any of > them continue with the project after leaving for a new startup. > Swapnil: I worked for MapR. I am a committer and I left after MapR shuffled resources away from Myriad. I am not sure how you are making these conclusions. "nor did any of them feel enough attachment to work after hours (as I do on my projects), nor did any of them continue with the project after leaving for a new startup" "None of the engineers previously working on this will be working on this now. And that sort of situation isn't going to change" > >> MapR is of course free to take its own decisions. But it sounds like >> there is interest in working on Myriad, just not under the ASF umbrella. >> I feel without ASF, one company will have too much control on Myriad. >> > > The ASF is moving to retire Myriad because it can't make the cut as a > viable project. No company will have control over the Apache version of the > project at that point because the project is nothing to control. > > Swapnil: I am sorry but its confusing to me weather you are speaking on behalf of MapR or ASF. > The desire to try to reboot the project outside of Apache has almost > everything to do with the fact that Apache processes and the lack of active > contributors means that nothing can happen. It isn't an end run around > Apache constraints for the purpose of control, it is an attempt to keep the > project alive at all. > > Swapnil: You are trying to reboot a project that hasn't been shutdown yet. Why not contribute now, as part of Apache? Why insist on retiring. You are telling me that somewhere inside MapR (or elsewhere) there are people who are super excited about working on Myriad and somehow being part of ASF is hindering them from contributing. I want to know who these people are? If they are so passionate about Myriad, why haven't they engaged with the community already and made any contributions? Its a little hard for me to believe that retiring from ASF will automagically fix everything and we will have a thriving community. If we has problems with the Apache overhead, can we discuss this in a different thread separate from the retiring discussion. I feel our community, however small, has rewarded committerships based on meritocracy. Anyone is free to come, contribute and gain influence. But lets first start with the contributions. > >> Ted, you yourself warned us against this >> http://www.zdnet.com/article/hadoop-veteran-ted-dunning-when >> -open-source-is-anything-but-open/ >> > > Read the article. I warned about projects like Ambari. One company has all > of the PMC. > > At this point, the situation with Myriad is almost the opposite. > Swapnil: yes, the situation is not like the closed open source projects that you talked about and I would like to keep it that way. Do you think the situation will remain the same if the code is moved under github.com/mapr. Wouldn't MapR have control to decide who commits and who doesn't? > > >> >> >>That means that it will always be a distraction to get committers >> qualified as PMC so that they can approve releases and it will never really >> be possible to exit from incubation. >> >> I suggest we start with the contributions first. >> > > Can you name the 3-5 active PMC members who will vet the next release? > > Swapnil: yes, shouldn't be a problem. I am one Regards Swapnil > >> >> >> On Mon, Jun 5, 2017 at 2:53 PM, Ted Dunning <ted.dunn...@gmail.com> >> wrote: >> >>> >>> On Mon, Jun 5, 2017 at 10:15 PM, Swapnil Daingade < >>> swapnil.daing...@gmail.com> wrote: >>> >>>> In that case I suggest we not retire >>>> >>>> >> "Darin - yes we've done more planning internally, and we do plan on >>>> having some engineers spend some time on this project, doing some (minor) >>>> maintenance for our customers." >>>> >>> >>> The problem is that there is essentially no real community that is >>> happening. >>> >>> None of the engineers previously working on this will be working on this >>> now. And that sort of situation isn't going to change. >>> >>> That means that it will always be a distraction to get committers >>> qualified as PMC so that they can approve releases and it will never really >>> be possible to exit from incubation. >>> >>> Outside of the Apache limits, we can have a much more flexible structure >>> of who can commit. We don't plan to limit who can commit. In fact, we will >>> probably make it more open than an Apache project normally is. >>> >>> >> >
