Probabilistic search - psearch.tar.gz

[pragma]

In article <e0u6na$1mdo$1 at digitaldaemon.com>, sundaresh says...
>
>Not many experts here walter.I am dissapointed.
>

Well, as you say, "the proof [...] is in the eating".  You, my good man, have
merely had a taste of our wonderful community so far! :)

In all seriousness, I think Wang's request for proof/data was the prime example
of an enlightened response.  I'd be suprised if others lurking here weren't
thinking the same thing and just didn't bother to post redundantly.  

My $0.02:
As this group is still participating in the formulative steps of the D langauge,
we get a lot of "pie in the sky" ideas around here.  If I may be so bold as to
speak for the group: we've all learned to approach things here with a dose of
skepticism (as professionals do) before embracing them completely.  The
unfortunate side effect of this behavior is that perfectly good new ideas still
have to fight for some degree of recognition thanks to this de-facto acceptance
process.  

And of course Walter's opinion trumps everyone's when it comes to what actually
gets into D. ;)

Also, I've noticed that people here actually read posts rather than skip over
the longer ones and hunt for attachments.  So posting just a .gz along with a
one-liner citing how awesome your algorithm probably didn't help.  It really
wouldn't hurt your case at all if you compose a small abstract about the
algorithm (how it works, what its based on) and supplement that with some timing
and performance data.  At least that way, nobody has to even touch a compiler
unless they feel the need to validate your claims. :)


- EricAnderton at yahoo