August 19, 2003
Recovery Time

After a frustrating day of product testing and unsuccessful SSL tool bejiggerynation, I went to the meeting last night.

It was, truth be told, a lovely antidote for a frustrating day. The highlights:

Andy Lester is kicking off the Phalanx project. The goal of this project is to shore up and beef up the testing cases in anticipation of Perl 5.10 (Ponie). He's seeking volunteers, of course, for this reasonably massive effort - get in touch with him, or watch for more info. I've cited Andy's writing here before, but his work to promote testing is (as my frustration above attests) a worthy goal, not just for Perl, but for thinking about software and tool development in general.

Afterward, Rich Moss (of Moss Search Ltd gave a brief talk about positioning for Perl developers in the area; he's a recruiter largely for financial firms - though with an ear to other areas.

And most lengthily, Steven Lembark gave his presentation on some development he's been doing lately in the bioinformatics area. He's promised the slides online, but from my view the big message of his talk was to be creative and thorough in your algorithm planning - there being more than one way to do it can literally shave cycles if you think about the backside of a routine (in his case, using polar math (e.g. addition and table lookups) instead of square roots (expensive calculations).

It was in some respects reminiscent of a database optimization seminar I attended a few years ago, an all day affair where a man with a thick russian accent showed us dozens of tricks to do comparisons using math instead of strings. I should pull his book out sometime and see if it makes more sense today....

All in all, a good meeting. And next month is weblog-oriented - in conversation with Jason about his presentation to be, it should be very interesting....

Posted by esinclai at August 19, 2003 05:34 PM |