Selecting cipher suites with socket.ssl

[Laura Creighton]

> Martin:
> I was mainly objecting the assertion "It's broken." I don't think that
> is true - it is mainly incomplete.

I think I should have said that instead, yes.  Apologies.

> > Besides, whatever he comes up with may be rejected on the grounds
> > that the code isn't good enough, or it adds the sort of complexity
> > you don't want in the standard library.  Modifying an existing
> > library shouldn't be by definition easier to do than adding a new
> > one -- foul code is foul code
> 
> I disagree that the current SSL support is foul code.

This is my day for misunderstandings.  I'm not saying that the SSL
is foul code.  I'm saying that if you write foul code, you shouldn't
think that you will have an easier time fitting it into an existing
library than submitting it as a new library for inclusion.

Laura Creighton