[XML-SIG] Spec for marking book reviews on the Web

Martin v. Loewis martin@loewis.home.cs.tu-berlin.de
Tue, 14 Aug 2001 21:53:45 +0200

> The goal is to finish the spec and then write a crawler that will find
> book reviews, assembling them all into one big database.  I'd be
> interested in getting suggestions on the spec, and pointers to any
> relevant standards that could be used to simplify it.

I'm not an expert on OPACs (online public access catalogs), but I
think a review should provide information similar to the book as you'd
find it in a library catalog.

There is plenty of precedent on this area already: ANSI Z39.50 is an
access protocol to such catalogs[1]. In Z39.50, books are described by
attribute sets. I'm not sure whether there are standard attribute sets
also, but it appears that there are atleast de-facto attribute sets,
like Bib-1 [2].

In fact, Book Review Digest offers a Z39.50 access to their database
[3], so you may want to look how they organize reviews.

There are apparently various approaches of representing Z39.50 data in
XML, e.g. using XER, RDF, SOAP; I'll leave have to leave that research
to you. It also appears that people are going the other direction as
well: allowing queries on XML files using a Z39.50 wire protocol.

As for appropriateness of this forum: I feel comfortable with such
discussions being held here - although people dealing with books by
profession are probably better to interview on this specific


[1] http://www.ariadne.ac.uk/issue21/z3950/
[2] http://lcweb.loc.gov/z3950/agency/defns/bib1.html
[3] http://mdusa.lib.umd.edu/cgi/list?action=dbz&db=BookReviewDigest