Post: Eat hot Google death, comment-spamming scum!
Good news: a small search engine company called Google – you may have heard of them – have implemented a system by which links tagged with the rel="nofollow" attribute don’t contribute to a site’s PageRank. This is a Good Thing, as PageRank-boosting is the main reason for comment spam. At least in theory, this should mean that comment spammers will have no more reason to post on blogs, although quite how long it will take them to figure this out and whether they’ll find some other reason remains to be seen.
Better news: a lot of people have signed up to implement this in their blog systems, most notably Matt of WordPress and Anil of Six Apart of Movable Type.
Better news than the first bit, but not quite as good as the second bit: Google, unlike some companies I could name, have (apparently) managed to implement this new feature without messing with the HTML specification, which explicitly allows for extra link types to be specified. It does, however, mention a profile attribute needing to be set in the <head> tag – I’m not sure whether this is something that needs to be added which points to some Google profile, but a test document I threw together without it seemed to get past the W3C validator. Something I’m going to have to look into at a later date, I feel, but at least it means people using nofollow will still be able to check their pages for validity.
In short, this is a very welcome development, and one I hope will help curb the glut of comment spam we seem to be seeing at the moment (especially for Movable Type users, causing some of them to move over to WordPress). In short, I’m hopeful – maybe we can kill off comment spam for good.
This is indeed a great idea, even if some cynics seem to suggest otherwise. To them, tackling a problem seems to suggest admitting defeat to the spammers. Perhaps blissful ignorance is better? 😉
But at the end of the day, Google and co. have achieved a standards complaint way of tackling an annoying problem (I’m fairly sure ‘nofollow’ already existed as a command for a ‘robots’ in meta tags etc.). As for profile attributes, you can have multiple profiles for the same document, so there’s nothing to stop you linking to Google’s profile and, say, XFN for the same document.
Kind Regards
I believe that although the HTML standards allow for free use of the rel=”” attribute there is actually something that determines the list of valid rel values. XFN values were added to this list, I expect that nofollow probably will be too at some point.
Ultimately, the idea is good. But it wont stop spam. Blog updating will be slow, people will try and find ways around it. Spam will continue for a while yet.
Oh, and the complaint most people have with “noindex” is that it’s not semantic. “rel” stands for “related”, so saying “related: next” makes semantic sense (it’s the next page in a sequence). Saying “related: noindex” means nothing. To be honest, they should have used “untrusted” as their value, since that would make semantic sense within the rel attribute. But hey.