Final statement from Steering Council on politically-charged commit messages

[Chris Angelico]

On Tue, Aug 18, 2020 at 11:12 PM Richard Damon <Richard at damon-family.org> wrote:
>
> On 8/18/20 1:22 AM, justin walters wrote:
> > I for one don't want to see politics involved in PL development. However,
> > inclusivity isn't a political issue, it's a human rights issue.
> >
> > Do I agree with the PR, not exactly. However, I do think we as a community
> > should be accommodating to people
> > Whose use of the English language differs from the standard as long as the
> > meaning is clear.
> >
> > This thread reads like a bunch of old fuddy duddies complaining about
> > immigrants not speaking perfect English at a fast food restaurant.
> >
> > Feel free to ban me from the list if this doesn't meet your standards.
> >
> One comment on this. I think the focus on the words that have been used
> is very much a '1st world problem'. Most of the people actually
> suffering from the problems being discusses would very much LOVE to be
> in a position where the discussion of what words are the right way to
> say something was their biggest issue. The forces behind these issues
> very much love to see our focus go to debating our speech, as opposed to
> actually DOING something about the issue. This isn't an accusation that
> those bringing up the speech issues are part of the dark forces behind
> the problems, but maybe a call to them to think about what really is the
> important issue.

This was raised. And the commit message clearly associates a language
standard with "white supremacy", the notion that certain people are
inherently better than others. Is it a first-world problem to consider
that light-skinned people are better than others? (No you don't have
to answer that.)

Had the commit message simply said "specific standards of English are
unnecessary in a global project", there could have been a perfectly
reasonable debate as to whether it's better to say "Strunk & White" or
to say "please write clearly at all times". (And on that debate there
are MANY valid points, including whether simple rules restrict or
enhance freedom.) But it said that having a standard *at all* is proof
that we all believe that white people are better than others.

> This also doesn't mean that language doesn't matter. If our language
> makes a Human Rights issue seem to be 'normal', that is bad, and weakens
> our resolve against it. Sometimes though, the right use of a word can be
> powerful, and analogies are great teachers. For example, talking (to a
> techie) how in a master-slave network, the master node has very strong
> control over what the slave node does, and then comparing that to a
> person, asking how would it feel to be that 'slave node', maybe even
> needing to wait for your 'master' to ask before you went to the
> bathroom, or be considered to be 'malfunctioning'.

TBH I think that removing "slave" from technical vocabulary is an
oversimplification and overreaction. If you have people who are being
treated as slaves, that is a problem. If you have people who are being
treated as machines, that is also a problem. Do we need to abolish all
use of "computer" and "robot" and "drone" from our technical language
too? Or do we keep those terms around, and recognize that these are
forms of dominance that must not be used between one human and
another? Simply abolishing the words doesn't prevent the practice, and
it just makes them into "dirty words" that are hard to discuss in any
context.

The replacement of "master hard drive" and "slave hard drive" with
"primary" and "secondary" isn't itself a major issue, although I'm a
bit annoyed at the churn that comes with sweeping changes. But where
next? How many other terms need to be reviewed just to see if they
could be misapplied to people? Are we going to get a "Black Pixels
Matter" movement among LCD manufacturers?

And none of this justifies burying something in a commit message.

Which the Python Steering Council doesn't even want to declare was
wrong. They are quite happy for the behaviour to continue.

ChrisA