RSS CAPTCHA Prototype
During the brief delay between when you click Post Comment and the form pops up, it's picking a random entry from the blogroll, pulling it down, and finding the first headline, just to be sure it's really there. After you make your post, it goes back out and checks again, to see if your input matches.
Interesting approach to attempting to catch a human behind the wheel. Reminds me a bit of old software protection schemes from ye olde days, though. Kind of like those ancient games that wanted to see "first sentence of second paragraph on page 39 of Flibbers of Froznats," a novel that came with the game materials that I had lost immediately. Kent's prototype goes and conveniently finds the book, though!
Hey - that gives me an idea. From now on - or at least as soon as I get a chance to implement it - I'll require all people trying to comment on my site to enter a phrase from a random page from one of my books. That'll be fun. :) And, of course, everyone around here already have copies of my books!
(No, this is not a statement of derision against Kent Brewster's prototype. This is me suddenly thinking of an annoying thing to do, which I'll never get around to doing.)
Archived Comments
No captcha mechanism will ever guard against human spammers. Ever since I put some simple anti-spambot meausres on my forum, 100% of the spam has been posted by people who are actually manually registering, copy-pasting a big chunk of links over, and clicking 'submit' (or at least they're close enough to humans that I can't tell the difference just by looking at the logs).
I've made it a tad less annoying, you guys. Come see, if you have a moment; thanks for the post!