[Python-Dev] RELEASED Python 3.0 final

[Stephen J. Turnbull]

glyph at divmod.com writes:

 > But still, you can't honestly expect me to recommend 3.0 until someone 
 > has gotten at least a basic skeleton of Twisted up and running under it 
 > :).  My own attempts to do so have failed miserably, to the point where 
 > I can't even produce a useful bug report without a lot more work.

How about an issue in the Python tracker---or the Twisted one, with a
xref from the Python tracker to the Twisted tracker where the work
will be done---that says "Twisted wants to be ported but we don't have
enough developers, please help"?  Maybe with some encouraging
statement about how you can provide X amount of advice.

In general, maybe there should be some sort of (semi-)formal process
for proposing ports of libraries and coordinating work on them.  Even
just a focal point for where to make such requests, and a way to
saerch for them so you can find others with similar interests.

 > I don't think there's anything about the 3.0 language which
 > couldn't be supported in a VM that understood both 2 and 3.

Strings vs. bytes.<shudder>  It can't do both 2-style "bytes are text"
and 3-style "no way are bytes text" simultaneously AFAICS.

 > I also don't think 3.0 is perfect, and five years on, there will be
 > a temptation to make more "just this once" incompatible changes.
 > Of course, you've promised these changes won't be made, and *this*
 > set of design mistakes will be with us forever.

For values of "forever" approximating ten years.<wink>

 > It would be nice if there were a way for evolution to continue
 > without another reboot of the world.

Stephen J. Gould says not.<wink>

I think Java is a very different case from Python.  It is the product
of a language evolution that goes back to the early 1970s or so, and
the standardization effort was carefully shepherded by a powerful
company which provided resources to ensure that things went its way.

For that reason, I think it's a remarkable compliment to Python and to
Python 3 in particular that you consider Java an appropriate standard
of comparison for Python.

There's also the danger of stasis.  I think Lisp will never die, and
Common Lisp has done a good job of avoiding reboots.  But for
precisely that reason there continues to be a lively evolution of
seriously incompatible dialects, both Lisp-1 (Scheme) and Lisp-2.  I
see Python 3 as an attempt to bridle and ride this tiger, without
turning the rope into a noose and strangling the beast.

 > >If they're that easily convinced that Java is better they probably
 > >were a lost cause anyway, so I won't mourn their departure too much.
 > 
 > I really believe that *all* new users are fickle, if they don't have a 
 > mandate as to what they need to be learning.  Personally, I learned 
 > Python because of a memory leak in Swing.

Sure, but what Guido is saying, I think, is that as long as prominent
Python developers don't announce its funeral, the other things we
could do to encourage them are going to get lost in the noise of
inherent fickleness.  Which isn't just random, it depends on things
like availability of just the right library for one's app, etc.  But
there are too many of those to do them all, or even just to list them
up and try to prioritize them "objectively"---might as well be random.