The current state of the art in comment spam

Write, geek! gets a fair amount of spam replies. This sur­prised me at first, when it began hap­pen­ing almost imme­di­ate­ly after the blog was set up and con­tent was post­ed. I should have known bet­ter; there’s almost no cost to spam­mers in spam­ming even unpop­u­lar blogs, so why would they make an excep­tion for mine?

I’m using the Akismet plu­g­in for Word­Press, so it’s not like any of these com­ments actu­al­ly make it to my blog. In fact, I’d nev­er even have to see them, if not for the fact that I reg­u­lar­ly clean these com­ments out of my spam fold­er by hand. I do this part­ly to ensure that noth­ing legit­i­mate gets fil­tered incor­rect­ly (which hap­pens some­times) and part­ly because I like to sort of keep tabs on the cur­rent ‘state of the art’ in spamming.

The cur­rent state of the art in spam­ming is this: the com­ments are get­ting bet­ter. No longer are com­ments jam-packed with dozens of links com­mon­place (one par­tic­u­lar default Word­Press set­ting prob­a­bly made those almost 100% inef­fec­tive), but they’ve been large­ly replaced with com­ments that mas­quer­ade as… actu­al comments!

The idea of noise dis­guised as sig­nal is noth­ing new if you’ve used e‑mail in the last 15 years, but that the noise is get­ting bet­ter (read: more dif­fi­cult for humans to detect) is some­what sur­pris­ing. Of course, these com­ments are no match for a large, dis­trib­uted sys­tem like Akismet, which all-knowingly sees what’s being post­ed to prob­a­bly mil­lions of blogs, but the well-disguised, large­ly pseudo-flattering com­ments are prob­a­bly now designed to get human blog authors to click the “Not Spam” but­ton, free­ing them the com­ments the spam box so that they can do their SEO-based dirty work.

Of course, gen­tle read­ers, I’m far too smart to fall for that, but not so blind­ed by my hatred for spam to be unable to appre­ci­ate a well-crafted work of author­ship, like this one I just found:

Sure, it’s not per­fect, but some­one out there put some mod­icum of thought into it, which is the least you could ask of the author of a work that’s going to be dis­trib­uted on a mas­sive scale.

Plus, it’s a lot bet­ter than this anti-gem I also just found:

Can you get more unin­ten­tion­al­ly self-referential than that? (No, you can­not… and yes, that was a challenge.)