And for my next trick...

I've got about 100 papers in my BibTeX database, and it continually grows. Many of the papers, given that I work in a field, have common authors; some also have common books (e.g. papers from the same proceedings).

Unfortunately, each entry is standalone; only the bibliographic entries themselves have IDs (e.g. linn:semantic-reliability). It would be firstly an enormous saving of effort if I could link bibliographic information in a graph-like way (e.g. link papers to proceedings rather than duplicating text, automatically implying a year of publication), and secondly it would be great to be able to process this data (e.g. finding similar papers based on author, or all papers in a year, or exploiting CiteSeer's 'number of times cited' information).

So, for my next trick: a bibliographic ontology, and a tool which will import BibTeX, allow sensible editing, and spit out BibTeX for backwards-compatibility. It should also deal with and output OWL, because I am researching the Semantic Web, after all.

Any thoughts, research community?

Postscript: now wouldn't it be great to integrate this with FOAF, so authors are directly integrated with the other knowledge you have about them? Paper recommendation, Semantic Web style.

Posted at 2003-12-16 04:06:56 by Richard • Link to And for my next tr…

Additions to Address Book Project

Witness abedit.py, which is a quick script hacked out using Pashua to allow you to edit your Address Book, including the hidden fields you normally can't get to: Middle Name, Maiden Name, Title, Suffix, Department, etc.

Install Pashua from the link above, then visit the Projects section to grab the files.

Posted at 2003-12-16 02:06:27 by Richard • Link to Additions to Addre…