Wednesday, September 27, 2006

Spam flatters

Spam gets smart, and not just technologically. I just got a comment which seemed to be both believable and nice, until I googled some of the keywords in it, and found that the exact same comment was all over the web, in some really strange places. Here's the comment, links removed:
****** ******* has left a new comment on your post "Let me screw up!":

I'm doing a paper about the merchandise liquidation and got this post. Its not where I was looking for but it is a good article for my Finance class... Very professional blog.
Of course I like to hear my blog is very professional. And yes, I know that Google can give random hits. Nice to hear that you can use what I write - what's not to love about this little comment? The spammer skillfully wields one of the more manipulative ways to address people, and makes it sound sincere and a little naive, as the sender positions him/herself below most people by being a student who uses your article/post in class.

But being suspicious, I checked if that search would actually lead to this blog. No, it doesn't. It does however lead to comments from the same person in several places.

For this blog I have activated several layers of blogger's protection options. First people need an account to post comments. Then they need to write a keyword which is not supposed to be copyable by bots. And last, I moderate comments. The first is fairly easy for a bot to deal with. The second was a good way to stop spammers, but either spammers are buying work time off chinese goldfarmers and having the spam done by hand, or there's now a way to translate the visual cue to letters without having it touched by human hands. Once those two things are dealt with, the spammer uses flattery and the randomness of net search tools to work on me, to have the human in the other end accept the comment. Clever, very clever.

(Another post analysing spam.)


Mike @ Vitia said...

Actually, the comment-spam strategy that has impressed me the most is the strong disagreement; the "I can't believe you said *: why are you so hostile?" thing. Got my attention right away.

Torill said...

Would be interesting to know what kind of spam works on which kind of people. Wouldn't a marketing researcher love to learn that?