There has been a lot of noise recently about Google’s proposed “solution” to comment spam – the rel=”nofollow” attribute for links, which makes search engines ignore them for the purposes of ranking pages by their associations.
It seems to me that there are actually two distinct problems here.
- Google’s problem. “All these comment spammers are messing up our search results. We’re not making as much money as before!”
- Joe Blogger’s problem. “All these comment spammers are polluting my blog, and I have to spend ages deleting it all!”
Now, which problem does rel=”nofollow” actually solve? Well, it certainly solves Google’s problem – which is not surprising, given that they suggested it. The Googlebot can say “Ah, this is a blog. Does it have some links with rel=”nofollow”? If it does, I can therefore trust all the other links on the page. If it doesn’t, I won’t trust any of them.”
However, to solve Joe Blogger’s problem, spammers would have to say “Hmm. I’m going to rewrite my extremely efficient spamming engine to check each blog to see if there are links there that have rel=”nofollow” and, if there are, not bother to spam it”. But they won’t. They’ll just spam you anyway.
The only way rel=”nofollow” will ever help Joe Blogger is if so many blogs use it that blogspamming becomes entirely pointless – and then the blogspammer will just stop outright. And, given the number of old and abandoned blogs littering the web, that time is some way off.
I suppose it might have more immediate effect on spammers who target particular centralised communities, like LiveJournal. If all LiveJournal links suddenly acquire rel=”nofollow”, then the spammer may hang up his script. But for Movable Type users like me, where every installation is different, this won’t happen. I’ll be getting my daily dose of online poker and phentermine (what is that, anyway?) for some time yet.