4 ms·
Are you sure about that? I've seen a number of reports over the last few days that "master" in git is basically a holdover from BitKeeper which had master/slave
by gimboland 6y ago
Are you sure about that? I've seen a number of reports over the last few days that "master" in git is basically a holdover from BitKeeper which had master/slave repositories (and that in BitKeeper you use repos to "do" branching). I haven't looked into this in enough depth to convince myself either way, but I would do so before echoing your first sentence. Certainly a _lot_ of people are assuming it means "master copy" and has nothing to do with master/slave, but there do seem to be reasons to believe that's not actually historically true, i.e. a flawed assumption.
Edit: getting downvoted for pointing out something true; somebody's feeling uncomfortable...
- jayd16 6y agoYou're right that I do not know the history. It might not be historically true but I would argue that's irrelevant to the conversation. (even as non relevant the conversation is to begin with!) Current perception is what matters. Beyond that, seems like if it actually did refer to a master/slave architecture, it seems like there would be more reason to keep it describing what it is. Maybe the term invokes some violence but so does the kill command.
- karmakaze 6y agoI used to use 'master' and 'read-slave' database nodes. I'm happier with 'writer' and 'read-replica'. I would say this is definitely in context.