osdir.com


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[OT] master/slave debate in Python


On Tue, Sep 25, 2018 at 5:02 AM Thomas Jollans <tjol at tjol.eu> wrote:
>
> On 24/09/2018 14:52, Robin Becker wrote:
> > On 23/09/2018 15:45, Albert-Jan Roskam wrote:
> >> *sigh*. I'm with Hettinger on this.
> >>
> >> https://www.theregister.co.uk/2018/09/11/python_purges_master_and_slave_in_political_pogrom/
> >>
> >>
> > I am as well. Don't fix what is not broken. The semantics (in
> > programming) might not be an exact match, but people have been using
> > these sorts of terms for a long time without anyone objecting. This sort
> > of language control is just thought control of the wrong sort.
>
> Never mind the justification and the overblown coverage in publications
> like the Register - if you look at the patches actually merged (mostly
> in https://github.com/python/cpython/pull/9101/files) they all look like
> entirely reasonable changes making docs, docstrings and comments clearer.
>
That particular PR is mostly non-controversial (there's some that are
under dispute, and dealt with elsewhere, and I'm ignoring those).
"Master process" is only one possible usage model so "parent process"
is more accurate anyway; "master and client" is mismatched; in fact,
the only one I'd even slightly disagree with is "buildslaves", since
that's a technically accurate term (they ARE slaved to the master
buildbot process), and that one has been changed upstream to
"workers", so there's no issues there.

The trouble is that making changes like this with a view to
eliminating the words "master" and "slave" from all docs and comments
(rather than making them to improve clarity and accuracy) opens up the
leverage that SJWs need. "Hey, you changed that because we hate
slavery - now you'd better eliminate all references to 'black' because
we hate racism". So clear boundaries need to be set.

ChrisA