Suggestion: make sequence and map interfaces more similar

[Random832]

On Thu, Mar 31, 2016, at 09:17, Mark Lawrence via Python-list wrote:
> On 31/03/2016 14:08, Antoon Pardon wrote:
> > Op 31-03-16 om 13:57 schreef Chris Angelico:
> >> Okay. I'll put a slightly different position: Prove that your proposal
> >> is worth discussing by actually giving us an example that we can
> >> discuss. So far, this thread has had nothing but toy examples (and
> >> bogoexamples that prove nothing beyond that the author knows how to
> >> mess with Python - fun, but not a strong argument on either side).
> >> Give us some real meat to work with, instead of these drips of
> >> tantalizing blood.
> >
> > What a strange request. Whether or not something is worth discussing
> > is a personal judgement. So there can be no proof of such a thing.
> > I would say: judge for yourself and act accordingly.
> 
> Drivel.  This is comp.lang.python, where "Practicality beats purity" 
> every time, not comp.theoretical.claptrap.

So can we discuss how a unified method to get a set of all valid
subscripts (and/or subscript-value pairs) on an object would be a useful
thing to have without getting bogged down in theoretical claptrap about
the meaning of the mapping contract?