- From: James Clark <jjc@jclark.com>
- Date: Tue, 29 Apr 1997 09:38:27 +0700
- To: w3c-sgml-wg@w3.org
At 16:47 28/04/97 GMT, Peter Murray-Rust wrote: >If you think of a scientific publication, you can think of it having >components (graphs, tables, molecules, figures, citations, etc) and all >of these will vary from publisher to publisher in order and amount. >It's quite meaningful and possible to ask: > 'please extract all molecules from this paper' > 'extract all molecules which contain spectra' >(notice the word 'all'). In general we shall not know how many molecules >there are in the paper or what order they come in. I can easily imagine this as part of processing. What I am still having a hard time seeing is why this sort of thing needs to go in an extended link or a simple link in an XML document. If you want to do this sort of thing, what you need is a fully-fledged query language. If you try to use xpointers for this, you will continually be finding things you want to express, but can't. >This assumes that you know that you need to count 17 along. When you >have no idea of the structure of the document, other than the components >it may contain (but not the level they may occur at or their order), the >tools that are currently suggested are ideal. I think this is the heart of the issue. There seem to me to be two kinds of problem: a) I have a collection of specific documents, and I want to create links between those specific documents. b) I have an open-ended class of documents, and I want to describe processing that works on any document in the class. For (a), I think a cut-down version of xpointers does what's needed. For (b), which seems to be what you're doing, a cut-down version of xpointers won't be adequate, but neither, I would suggest, would an enhanced version be adequate: (b) is what DSSSL is mainly about, and for that you need a full query language. James
Received on Monday, 28 April 1997 22:52:50 UTC