Abridged. Below.

On Tue, Jun 21, 2011 at 6:36 AM, Dale Schumacher
<[email protected]>wrote:

>
> Just to be clear.  The ? symbol represents "undefined".  The NIL
> symbol is semantically equivalent to (), the empty tuple/list.  And
> finally, the _ symbol represents the wildcard pattern with no implied
> binding.  See the Humus Overview
> (http://www.dalnefre.com/wp/humus/humus-overview/) for details.
>

Oh, heh! I stand corrected. This one is always sticky because there's been
so much variation in different languages. IIRC even Lisp and Scheme differ
slightly on this one, right?

This reminds me of one of the funniest things that I ever saw at work. This
architect, who's identity I will protect (I mean, nice smart ambitious guy
usually) wanted to roll his own cross platform RPC. I kicked and screamed
and pointed at things like THRIFT because I really didn't want to end up
owning another unnecessary proprietary RPC mechanism, but I lost that
argument and it got built and we deployed it.

You should have seen me nearly fall out of my chair laughing when someone
hit the first bug, which I had a feeling would be in there. I can't claim
that I found it myself (I was juggling a lot at the time,) but I was two
cubes away and heard the howling expletive emitted by the guy who did. I had
a plan to go looking for that problem. I was kinda robbed;)

Ruby and Perl (the two languages of central interest at the time) differ
slightly with regard to the truth value of the number 0. Ruby wants to say,
"yeah, zero is an object and it isn't nil." Meanwhile, Perl programmers
regularly use zero as a shorthand for falsehood.

CC FONC because it's a funny story about ill-advised use of language. The
poor Perl folks ended up having to write FALSE, or 'FALSE', if I remember
right (BCC'd someone who was there, maybe his recollection about that is
better than mine.) They were sooooooo mad. Fortunately, I wasn't working on
the Perl parts much, because if I had been, I probably would have led the
revolt.

-- 
Casey Ransberger
_______________________________________________
fonc mailing list
[email protected]
http://vpri.org/mailman/listinfo/fonc

Reply via email to