You mean every body in at least our current develop team should/will have a 'committer' permission? I agree with you that CTR is ideal model when our project has 'enough' eyeballs and tests. But for now, I'm still worrying about the quality.
Most of us, including me, have a really bad old habit: Push too much. If everybody can slow down a little bit, and review the code many times himself, I think we can switch to CTR. I'd like to hear from others, any thoughts about this? BTW:There are tons of abbreviation-with-capital-letter need to study... that's interesting. 😄 On Sun, Mar 12, 2017 at 10:50 PM, Niclas Hedhman <[email protected]> wrote: > Yes, that is what people coming from a corporate background normally think. > In reality, you now have (potentially) every user watching and approving > what you are doing, and the number of eyeballs on commits are typically a > function of the size of user base. This aspect of "open development" can > seem a bit scary once it strikes home, but as the user base grows, every > committer tend to be more and more careful. Good testcases will bring you > more safety against "low-level bugs" than RTC. > > I hope it is reconsidered, as RTC is rather rare in ASF for 'develop' > branches. > > If you have ever heard about the "VETO" in ASF, then that is the "counter > balance" of CTR when a committer is "going rogue", i.e. committing "bad > stuff" for whatever reason and possibly ruining other people's work. VETO > should be rare in a healthy community, a bit like a emergency break on a > train. Read more at; https://www.apache.org/foundation/voting.html > > > Cheers > Niclas > > On Sun, Mar 12, 2017 at 6:23 PM, 大诗胸 <[email protected]> wrote: > > > Here is the reason why we decided do RTC: > > We wanna make sure every modification is reviewed by another people. And > > avoid commit any feature without discuss with teammate. We use this to > > minimize the low-level bug. We do realize RTC will slow down the > > development，but we think it will be worth doing. > > > > > > > > On Sat, Mar 11, 2017 at 8:41 AM, Niclas Hedhman <[email protected]> > > wrote: > > > > > Out of curiosity, why is "pull request model", normally called RTC > > (Review > > > Then Commit) in ASF legacy terminology? RTC has traditionally been used > > for > > > mature branches, say maintenance release branches (2.1.1, 2.1.2, 2.1.3) > > but > > > for 'development' branch or the new upcoming release, the dominant > model > > is > > > CTR (Commit Then Review). > > > > > > We don't disallow RTC, but generally encourage to start as CTR, as we > > have > > > found that faster development is possible in CTR. > > > > > > What are your thoughts? Is this simply a left over from a > > corporate/github > > > environment, or has this been actively decided upon? > > > > > > > > > Cheers > > > Niclas > > > > > > On Fri, Mar 10, 2017 at 5:42 PM, 大诗胸 <[email protected]> wrote: > > > > > > > New contribute page: https://github.com/apache/ > incubator-weex/pull/88 > > > > > > > > > > > > On Thu, Mar 9, 2017 at 11:33 PM, Jinjiang Zhao <[email protected]> > > > > wrote: > > > > > > > > > Guys > > > > > > > > > > I have written an email to list the things we are > working-in-progress > > > > and updated > > > > > the issue template of alibaba/weex repo to announce that we will > stop > > > > > replying new issues any more from the end of this week. > > > > > > > > > > Additionally, I could build a page named "how to contribute" on our > > > > > website tomorrow. And then add the url to the issue template above. > > > > > > > > > > Thanks. > > > > > > > > > > Jinjiang > > > > > > > > > > On Mar 06, 2017, at 11:59 AM, Jinjiang Zhao <[email protected]> > > > wrote: > > > > > > > > > > Yes as you suspect, some of these PRs are from discussion in other > > ways > > > > > and some of them are from old pieces of work. 50% - 50% I think. > But > > > yes > > > > > it's not OK of course. > > > > > > > > > > Guys I have some suggestions: > > > > > * First we should put more old information here, just like what > plan > > we > > > > > have made or how many issues we are doing with. > > > > > * Migrate old alibaba/weex issues to apache gradually. > > > > > * Stop replying new issues in alibaba/weex (at least at a > appropriate > > > > time > > > > > if not from now on). > > > > > * Encourage Chinese users moving here MORE. > > > > > * Try to guide more users abroad go to Apache directly (through > > > Twitter, > > > > > HackerNews ... I think). > > > > > * We should pay more attention here. > > > > > > > > > > Thanks. > > > > > > > > > > Jinjiang > > > > > > > > > > On Mar 03, 2017, at 06:29 PM, Niclas Hedhman <[email protected]> > > > wrote: > > > > > > > > > > Gang, > > > > > I see a massive amount of work (well done) being done and pull > > > requested > > > > > via GitHub. > > > > > > > > > > Are these old pieces of work that are pushed now, or is it new > work? > > If > > > > the > > > > > latter, how does people know what to work on? > > > > > > > > > > Perhaps it is simply totally independent pieces, driven by personal > > > > > interest, BUT I SUSPECT that this is agreed upon at work, on some > > side > > > > > channel or otherwise. And that is NOT OK. > > > > > > > > > > Apache projects are expected to have discussions on features and > > fixes > > > on > > > > > the mailing list. As the saying goes around here; "If it didn't > > happen > > > on > > > > > the mailing list, it didn't happen!" > > > > > > > > > > Additionally, all communication in GitHub Pull Requests could > > disappear > > > > > tomorrow. The project's history would be gone as a result. No > records > > > of > > > > > anything, other than the commit messages. > > > > > > > > > > So, everyone, I know this is difficult, especially since most of > you > > > are > > > > > Chinese, probably with limited English fluency. But we need to get > > the > > > > > discussion going HERE... not at work, not on WeChat, not on Slack, > > not > > > on > > > > > GitHub or any other medium... HERE. > > > > > > > > > > > > > > > Thanks for listening > > > > > -- > > > > > Niclas Hedhman, Software Developer > > > > > http://polygene.apache.org <http://zest.apache.org> - New Energy > for > > > > Java > > > > > > > > > > > > > > > > > > > > > > -- > > > > sospartan > > > > Phone:13588488290 > > > > HangZhou > > > > > > > > > > > > > > > > -- > > > Niclas Hedhman, Software Developer > > > http://polygene.apache.org <http://zest.apache.org> - New Energy for > > Java > > > > > > > > > > > -- > > sospartan > > Phone:13588488290 > > HangZhou > > > > > > -- > Niclas Hedhman, Software Developer > http://polygene.apache.org <http://zest.apache.org> - New Energy for Java > -- sospartan Phone:13588488290 HangZhou
