Suggestion: make sequence and map interfaces more similar

[Antoon Pardon]

Op 31-03-16 om 12:36 schreef Steven D'Aprano:
> On Thu, 31 Mar 2016 06:52 pm, Antoon Pardon wrote:
>
>> it is your burden to argue that problem.
> No it isn't. I don't have to do a thing. All I need to do is sit back and
> wait as this discussion peters off into nothing. The burden isn't on me to
> justify the status quo. The burden is on those who want to make this change
> to justify the change, because if you don't, the status quo stays exactly
> the same.

Sure that is all very well if you stay out of the discussion, or limit
your contribution to mentioning that in your opinion that this is a
very low priority. I have no problem with that. But if you begin to
argue that the proposal has flaws and you argue against it then it
is your intellectual responsibility to support your arguments.

There is a difference between, (1) this proposal is flawed and (2)I
don't think this is important enough. Starting with the first and then
when pressed to support it, retreating to the second is not fair.

> And if you're brave enough to take to this Python-Ideas, let alone
> Python-Dev, the first question they'll ask is "What's your use-case?". And
> since you don't have one, this discussion will go nowhere.

So? That doesn't relieve you of your responsibility when you somehow
argue that the proposal is flawed, beyond there being no use case.

> Oh, there might be a few hundred posts on the topic, from people
> bike-shedding it, but unless you convince the core developers, all the
> chattering in the world won't get you one nanometre closer to changing the
> behaviour of lists and dicts.

I have no interest in convincing the core developers. That doesn't mean
I can't respond here to arguments against a proposal that are IMO bogus.
A bogus argument against a proposal doesn't become a good argument against
because there is no use case.

> So, Antoon, no, I don't have to justify a single thing. If you want this
> change, you have to justify why it should be done.

Indeed you don't have to argue against any proposal. You can just sit back
and ask for use cases and ask do be convinced this proposal is important
enough for the core developers to invest time in. But once you do argue against
a proposal, as you did here, you have loaded yourself with the burden to
support your argument. Because then your position is that even with an
import use case, this is still a bad idea. And if you can't support that
position it isn't fair IMO to then retreat to the "no use case" position
as if that somehow is a defence of your argued position.

-- 
Antoon.