Proposal: default __init__

[Alex Martelli]

<fredrik at pythonware.com> wrote in message
news:8urf80$tsr$1 at nnrp1.deja.com...
> Andrew Dalke wrote:
> > >agreed.  I suggest writing a draft PEP and submitting it to
> > >python-dev (see PEP 0 for details).
> >
> > Okay.  Part of the reason for the length of my other posts is
> > to clarify my thoughts before writing such a PEP.
>
> maybe we need some kind of pre-PEP incubator forum?
>
> I'm not sure comp.lang.python is really working -- way too
> much "don't you dare changing *my* language" nonsense here
> these days... :-(

<smile> Yeah, and I'm doing most of that myself, I guess...
but I don't labor under the illusion that it's "my" language --
I know very well it's _GvR_'s! </smile>

Seriously -- isn't it better if objections to changes are
thoroughly expressed *before* the proposal moves to a list
where only a selected few can write?  If the objections ARE
indeed "nonsense", it should be reasonably easy to show them
up -- "no, your initbases idea won't work because..."; "no,
this PEP will not break mixin classes since..."; "general
objections to global variables are misplaced here, due to
the fact that...".  If valid (though maybe not decisive)
concerns are expressed, the PEP had better address them, no?

So far, it seems my 'nonsense' on this thread has managed
to change some people's misconceptions on how getattr works,
and prompted AMK to think the relevant part of his 'warts'
page (which you quoted) is in error and must be corrected.

Is it a sign of a discussion forum "not really working",
when some long-standing misconceptions are cleared up as
a result of the discussion...?

I know my somewhat-confrontational (OK, OK, flamey!-) style
is not to everybody's tastes -- but if I wasn't making
these points, somebody else had better be raising them,
since their validity (or otherwise), and their import for
the future of Python, should surely not depend on the
conversational graces of one or another person, should it?


Alex