Hey- So on Mahout, we decided per Rich's original thread on diversity@ on replacing whitelist/blacklist terminology, that if renaming our `master` branch to something else attracts just one more contributor it's worth it.[1]
Also on diversity@ Naomi first brought up that `master` as in master branch was actually a master record. (Daniel pointed this out). akm found a Twitter thread[2] which says basically what we were thinking (it's appear inclusive than be technically correct on word origin), however the thread also referenced an email thread from GNOME when they were having these discussions a couple of years ago[3] and the outcome was someone did research into something called BitKeeper (which was before my time) but that was where the first master branch existed (GNOME email cited the commit that created it) and from their docs: > In this section we are going to show how to interact with the master repository and how to deal with merging and conflicts. For this demo, we will need to create a small program which we will then push to the master repository. We are then going to modify the file on both the master and slave repository and then merge the work. So for our project- we had already decided to change the name of our master branch. But I keep seeing the (reasonable, and what I had originally thought as well) idea that "master" is for "master record", which is fairly demonstrably false. I was originally shown that tweet several days ago- and have since seen more and more of the open source community moving away from naming their main branch `master`, but as I see this continue to be a discussion, I just wanted to drop something I had seen. Apologies for not doing it sooner- (as you might have guessed) I've been spending more of my time in the streets this week than I have in code bases. As an aside- I don't think there should be ASF mandated guidance on this either, but I would like individual projects to make the decision for themselves from a place of what is the actual origin of "master" branch terminology. tg [1] https://issues.apache.org/jira/browse/INFRA-20403 [2] https://twitter.com/tobie/status/1270290278029631489 [3] https://mail.gnome.org/archives/desktop-devel-list/2019-May/msg00066.html On Sat, Jun 13, 2020 at 3:40 AM Dominik Psenner <dpsen...@gmail.com> wrote: > Context matters.. the context here is the branching model. Master in the > context of git repositories is git terminology, just like what default is > in the context of mercurial repositories or any other $default_branch in > the context of $dvcs. This has nothing to do with master/slave relationship > in the context of redundant systems or jenkins or what it meant around 1850 > in America. > > A project may choose the most appropriate branching model at will if this > is what the community wants and it does not violate against the code of > conduct. A short phrase in the project documentation like "master refers > to the default branch and marks the latest stable release" or "this project > uses gitflow branching model" may make this clear. Slave may be an > appropriate branch name, it depends on the context and its meaning if it > adheres to CoC or not. > > Warm regards, > Dominik > > Ps: Pizza may not be an edible good in the context of Terry Pratchett > novels. ;-) > -- > Sent from my phone. Typos are a kind gift to anyone who happens to find > them. > > On Thu, Jun 11, 2020, 07:11 Konstantin Boudnik <c...@apache.org> wrote: > >> Shall we make an addendum to CoC to reflect it as an official naming >> policy for our repos? As you know - they are up there on the github and >> might negatively reflect on our public image...? >> >> -- >> With regards, >> Cos >> >> On 2020-06-10 03:46, Daniel Gruno wrote: >> > On 09/06/2020 22.41, sebb wrote: >> >> It's obvious that master/slave is problematic. >> >> >> >> However is 'master' problematic when used on its own? >> >> >> >> e.g. master Git branch ? >> > >> > As mentioned elsewhere (on dev@diversity I think), the general >> consensus >> > is that 'git master' refers to a master record, not a master in the >> > sense of an actor. There is no 'slave' branch (well, not in most >> > repositories I've seen), so I would think that drawing a connection to >> > master/slave here is more of a willful determination to see problems >> > that don't exist. >> > >> > What is problematic to me is not the word 'master' or the word 'slave', >> > but rather the use of both within the same system. >> > >> >> >> >> Sebb >> >> >> > >> >
