3 ms·
> These sorts of terms are impossible to define As described in the article by a CoC critic, it used to be: "only whether their code was good" Not exactly a
by Jare 8y ago
> These sorts of terms are impossible to define
As described in the article by a CoC critic, it used to be:
"only whether their code was good"
Not exactly a shining example of clearly defined, objective, universal judgement. Of course, back then it could really "if Linus thinks their code is good", which is a lot more clear. But that time has passed.
The CoC tries to address the problem where people ARE in fact being excluded or lost by the Linux community, in less visible and obvious ways than a straight ban, for reasons equally less visible and obvious, but no less real.
You may or may not believe that this happens, that this is a problem, that the CoC helps or doesn't help with this problem, and even that the net effect of the CoC is positive or negative.
- bigbugbag 8y agoOne way or the other people are lost to the linux kernel development by what actually matters is which ones provide the best code. Is Sage Sharp contributing better code to the linux kernel than Theo Ts’o ? Would you replace all Ts'o contributions by Sharp's ?