StaticTuple... naming, maintenance, ...

[Matthew D. Fuller]

On Mon, Oct 05, 2009 at 05:58:02PM -0500 I heard the voice of
John Arbash Meinel, and lo! it spake thus:
> 
> I would *really* like to switch to Cython 0.11+, as I have specific
> benefits. One could argue that we could try to be compatible, and
> people can compile using Pyrex, and just wouldn't get the memory and
> speed improvement of avoiding the GC...

I don't know anything about this particular case, and I'm not
qualified to say much about it.  So I'll just throw a few generic
and thoroughly unoriginal thoughts.

There are 2 categories of costs in retaining that sort of
compatibility.  One is the maintenance cost; how hard is it to have in
the first place, and maintain going forward.  The second is the
capability cost; how much does keeping working [without unreasonable
maintenance burden] with $OLD prevent you from getting the potential
benefits of $NEW?

Purely from the discussion, it _sounds like_ cost (2) here is small;
many of the gains can Just Happen without the code being different.
So cost (1) is probably the determining factor.  Not being the one
shouldering the cost, it's easy to say, but I'd as soon not have a
sudden gear-switching flag day unless the compat cost is fairly
significant.  I'd be fine with saying "Using pyrex will work
correctly, using cython will work better" either temporarily or
indefinitely, or if the costs are significant and growing, adding "and
at $FUTURE_TIME we'll stop trying to keep pyrex working".

For portability data, there's been a FreeBSD port of cython since
November, and the maintainer seems to be active in keeping it up to
date (currently at 0.11.3).  So that's one platform checked off for
it.


-- 
Matthew Fuller     (MF4839)   |  fullermd at over-yonder.net
Systems/Network Administrator |  http://www.over-yonder.net/~fullermd/
           On the Internet, nobody can hear you scream.