Tom Metro wrote:
It would be a start, though I'm not so sure that Brainbench's web-based, open book tests are close enough to a certification to have the intended effect.
A self-testing program *need not* suck.
Here are four reasons why a home-grown testing program would be (could be) better than Brainbench, followed by an (improved) argument for a bottom-up approach to boosting Perl:
1. Paradigm. Multiple choice tests are for geography. *Writing Perl scripts* is the stock-in-trade of a Perl programmer, so why not test scripts?
2. Fun. Brainbench, like a sterile cinder-block high school classroom, is no fun. But when the test-writer and the test-taker are peers (in their love of Perl), the challenge can become a sporting endeavor. We love puzzles, right? The testing environment can be transformed from a student-vs-teacher to a sort of parlor game for geeks.
3. Inclusive. Wiki has its sandbox. Perl-test could have a sandlot. Bring your tennis shoes and an old worn-out copy of the Camel and step up to the plate. That is, take a test and/or write a test at no cost. No one with horn-rimmed glasses will be checking your student ID.
4. Quality. The Brainbench tests are limited in scope and change rarely, I would guess. (No 'demonstrable evidence' on this list, please :-| ). If a Perl self-testing program is to succeed, it would have to have the same kind of self-supporting energy that, for example, this list has. If it's fun and inclusive, interest will lead to quantity and quality. (Perhaps a user feedback system or loose-knit moderation could champion the best tests).
I'm conceding that this self-testing program could not substitute for a certification program. But I still believe it could be part of a strategy to create and sustain "buzz" about Perl from the bottom-up.
-Bogart
_______________________________________________ Boston-pm mailing list [email protected] http://mail.pm.org/mailman/listinfo/boston-pm
