Thursday, August 25, 2005

Trackbacks and arXiv

Paul Ginsparg has introduced a new experimental service - trackbacks; most likely, Jacques Distler was the driving force behind this new feature. See Paul's explanation here. This service allows the visitors of arXiv.org, at least in principle, to see the list of blog articles that are describing a given paper. For example, look at the paper

You will see "1 trackback" at the bottom of the page. Actually, there are 2 trackbacks; their new system has some minor bugs so far, for example it cannot count properly. If you click at this "1 trackback", you will see two blog articles about the paper, namely Jacques' and mine.

Google is ahead in many things; unfortunately, their blogger.com that runs all these blogspot.com blogs (including this one) has some features missing. Trackback autodiscovery is one of them. I've tried to include an RDF code that has a chance to fix it, using some haloscan.com URLs, but it probably does not work. What does it mean? It means that if you like a blog article - and/or a discussion below it - about an arxiv.org paper, you may want to "ping" the arxiv.org manually. For example, you open

and fill out some of the entries:

It will only work if the blog article contains a link to the PS or PDF or abstract page of the article, and if the link has http://arxiv.org in it; for example, http://www.arxiv.org that often appears in my posts is probably enough for the operation to fail.

The process has been strikingly simplified for you. Below every article, you should find a web form where you only fill out the trackback URL and press "Submit" - the excerpt may be ignored if the trackback links to the arXiv...

Have fun.