[Edu-sig] re: Elegant copy-by-value

[Kirby Urner]

At 07:24 AM 1/14/2003 -0500, Arthur wrote:
> >
> > hasn't he clarify it in http://www.python.org/sigs/edu-sig/ ?
> > That's what I read when I decided to join this forum.
>
>I see no reason to put any strictures on the list subject matter - beyond
>filtering spam.  Certainly the list does not suffer from an over abundance
>of traffic.
>
>And by question to Guido was quite specific.
>
>He might have answered me using no more bandwidth than he used with
>reasoning as to why he was not answering me.
>
>Why is formality necessary.

I think Guido was implicitly inviting you to join the higher ranks,
by continuing this thread among those who actually do mold the
language -- over on python-dev for example.

 From the Python website:

   Finally, <http://mail.python.org/mailman/listinfo/python-dev>python-dev 
is the heart of Python's development.
   Practically everyone with CVS write privileges is on
   python-dev, and first drafts of PEPs are often posted here
   for initial review and rewriting before their more public
   appearance on python-announce.

   Anyone can subscribe to python-dev, though your subscription
   will have to been approved, resulting in a delay of a day
   or two before you start receiving e-mail. The list address
   does accepts e-mail from non-members, and the archives are
   public.

There really is a structure to the Python community.  I think
it's up to the would-be contributor to pay some attention to it.
Trying to make Guido engage you here seems a kind of willful
inattention to

The pros and cons re moving copy and/or deepcopy into the
built-in core were already summarized on this list, rather
recently.  We also had a short thread on rational numbers
(a proposed new number type) -- in part because I was reporting
back from fresh experience with J, and one thing we *do* talk
about on edu-sig is comparisons with other languages (the
recent REBOL and longer-ago ISETL threads come to mind).[1]

I think edu-sig should only intersect language design and
development threads occasionally.  Ideally, I'd like to read
pithy synopses regarding the debates of the day -- but be
spared the blow by blow stuff.

If you do pursue your design concepts amidst the developer
community, I hope you will update us on your progress, or
lack of it, from time to time.

Kirby

[1]  One thing I'm working on now is trying to figure how to
pass J commands as strings back to jserver, using the Python
sockets library.  In principle, it's doable, and it'd be
cool to avail myself of some of J's powers, from within
Python.  I suppose Mathematica could be used the same way.
If anyone here has experience with this, I'm eager to learn
more.