Brad Fitzpatrick wrote:

Hmm.

It could be argued that the livejournal.js file in sgnodemapper *is* a special case for LiveJournal, along with special cases for Amazon, AOL, Blogspot, Facebook, ....

To be honest, I'm not entirely convinced about the sgnodemapper project. I can't help but think that a central catalogue of the URL normalization rules for every service in existance just won't scale, and that this needs to be a decentralized process. The simplest solution that springs to mind is this:

<link rel="canonical" href="sgn://livejournal.com/?ident=mart" />

Now of course processors of this would be required to ignore assersions that are in the "wrong" domain, and you might have to adjust the sgn syntax a bit to allow for this, but I think something like this would scale better in the long run. In the short term, the special cases in the library are a good bootstrapping solution for the sites which haven't adopted this yet. (all of them!)

I'm also not convinced that using URLencoded query string arguments is a good idea, but that's a discussion for another day. Is there a separate mailing list for sgnodemapper?