Blog powered by Typepad

« Frank Jackson papers | Main | Zombie horoscope »

May 23, 2005


Simon Prosser

The main disadvantage of fine-grained categories in the 'people with online papers' section is that each person appears in only one category but many have papers in more than one category. The finer the grain, the more papers get missed; so there's a limit on how useful the categories can be as an aid to searching for papers on a particular topic.

I can't think of a solution that wouldn't involve huge amounts of work, though. In an ideal world each person could have several categories and it would be possible to search for people by category. But I imagine that requires a whole different level of technology, and an awful lot of work! At least having a list of topics after each name makes it possible to search manually.

Tennessee Leeuwenburg

The fineness of grade should be related to the number of papers. There should be about 30 papers in any given node - much more than that and it's too hard for a person to search exhaustively. What you need is a splitting method allowing you to break any node that is too large into multiple nodes, based on category.

A node should have up to 30 entries, any of which is either a node or a paper. i.e. 29 papers and one "node" of papers which is navigable. Or 30 "nodes" with 20-odd papers each.

Make it so! :)



My comfort level is somewhere around 40-45, I think. As Simon says, it's a trade-off, and I think aiming for 30 would introduce too much fine-graining (I'm already not wholly comfortable by the fine-graining introduced by the causation/laws and time sections). It looks like as things stand most of the sections are under 45, except for philosophy of language (around 65), metaphysics and epistemology (each around 50, even after the splits above), and social/political philosophy (also around 50). I just now split off a new meta-ethics section from the ethics section, which helps that one.

It might be nice to find a principled way to split the philosophy of language section (I already split off a "philosophy of linguistics section" a while back), but I don't currently see any really natural way to do that.

Ming Tan

Dave, the causation/laws category makes a lot of sense. The philosophers in this category are definitely engaging with each other's work. For example (off the top of my head), Handfield has a paper responding to the work of Lange, Eagle has a paper responding to the work of Johns, and Twardy has a paper addressing the work of Hiddleston.

Recent discussion over at "Certain Doubts" suggests that formal epistemology is a burgeoning area in its own right. So this category also makes a lot of sense.

Nor is there anything obviously wrong with the time and decision theory categories - except the need for a few more entries in both!

The comments to this entry are closed.