To followup on my story, I did a search on one of the usernames and found an identical post made to a book forum on aol. The poster there used the same username as part of her email address. I sent an annoyed email asking why she was spamming and who was paying her.
And all of the spam posts to my forum stopped.
Didn’t get a reply email though. Rude. 
Now she knows your email address is active and will probably sell it.
She’s welcome to. I get hundreds of spam messages to that address, and they never make it past my filters. I’d be happy if spammers waste money to buy my address!
Maybe these are posters who just reeeeeeeeeaally loved the book.
Cuz if that book happened to be “The Cheese Lovers Cookbook and Guide: Over 150 Recipes with Instruction on How to Buy, Store and Serve All Your Favorite Cheeses” by Paula Lambert (Simon & Schuster, 2000), why a page-turner like that would just be plumb impossible to keep to one’s self. Believe me, I’ve tried! Classic beach reader if ever there was one. I hear Spielberg’s working on a screen adaptaion right now.
I can’t figure out why a recent silent spammer on my boards was pimping Imodium in their website url. Are Imodium sales down due to those compelling Pepto-Bismol diarrhea-dance ads and they are combating this by resorting to spamming?
Captchas are running up against a fundamental limit: the ability of a human to parse them, nevermind a bot. There’ve been a few captchas lately that I’ve been unable to read. I’m thinking that they’ll need to transition to an image-comprehension, rather than an image-recognition paradigm sooner or later. In other words, instead of a mangled text string, they’ll need to have a constructed scene, plus a question about that scene. It should be possible to assure an arbitrarily low false-negative rate by tweaking the multiple-choice answers.
If a bot can break that, then hand its writer a Turing Prize: they’ve successfully advanced the field of AI.
HORMEL?