Re: Compiler Branch

[Andy Wingo <wingo@pobox.com>]

Hi Noah!

On Tue 13 Dec 2011 14:52, Noah Lavine <noah.b.lavine@gmail.com> writes:

>> Cool.  As a quick reaction, I have some doubts about this project.  But,
>> I guess a WIP branch would be a good thing to have, and it would make
>> the discussion more concrete.
>
> Probably so. But if you have time, what are your doubts? I would much
> rather talk about problems now than after I've implemented something
> that we'll have to change.

OK, I've taken a look at it now.  I apologize for my negativity, first
of all.  I'm not sure exactly where you're going with this, but it was
wrong for me to be a wet blanket about it.  So sorry about that!

Here is my calculus regarding this work, FWIW: it is low risk, and low
cost, since it doesn't actually affect Tree-IL or any other part of the
compiler.  The only cost it imposes is a coupling with Tree-IL, which is
easy enough to deal with on that side.  So that makes me much less
uneasy :-) And on the positive side, it has the potential to prove
interesting things, so at this point I'm neutral about it.  If it does
something useful, IMO we should throw it in, when you're ready for that.
Other people's opinions are welcome as well, of course.

Can you describe what the branch does, currently?  What do you see it
doing within three months or so?  How long do you intend to work on it?

What do you think about the tree-il differences in master relative to
stable-2.0?

Do you see this work as an optional pass, or a core part of the
compiler?  If the latter, what sort of algorithmic complexity are you
envisioning for this work?  (O(n) in size of program is ideal of
course.)

Cheers,

Andy
-- 
http://wingolog.org/