Re: Diversity and offensive terminology in RFCs

[Ted Lemon <[e-mail]>]

On Sep 21, 2018, at 9:02 AM, Eliot Lear <[e-mail]> wrote:
> Although we don't actually have any specifics as to why or how the
> terminology posed a problem.  We have no understanding of the harm in
> this case.  In python one person chose to believe that another person
> really had a problem, and that was that, end of discussion.

Eliot, I've actually explained the context of what transpired to you on the hrpc mailing list, and I don't see what I said reflected in your response here.   It wasn't "one person."   And it wasn't just python—when this was raised in another open source project and taken to a vote, a majority were in favor of using different terms.   None of this rises to the level of statistical clarity, but when dealing with issues like this, that kind of data is difficult to come by.

Do I need to point out to you that quite frequently in situations with bad power dynamics, getting accurate reports from injured parties is very difficult, and there's a strong tendency for their reports to be disbelieved by the majority in power, but when the dam finally breaks, it becomes clear that there was a real problem, and that the majority, in ignoring that problem, created real and substantial harm.

If our standard for approaching this problem involves small p-values, I doubt that we will make any changes at all.   The question is, do we need to require that standard of evidence, or are anonymous reports from people who say they have experienced abuse as a result of the use of this terminology sufficient?