Stephen, Thanks for posting this most interesting topic. - until your posting I had missed the announcement from W3C. On Tue, 14 May 1996, Stephen Le Hunte wrote:
which seems to support some things that chemists might want to see. Perhaps the most important is support for style sheets and greek. I wonder what the scope will be for the chemical community creating such style sheets!
Transmission of greek and other characters is probably the least of our problems. There is a set of about 11 ISO entity sets (4 greeksets, technical, maths, etc) which are widely used in (say) the TeX community (most of these have direct translations into TeX). The problem is rendering the glyphs in whatever tool is used to process the info - browsers have lagged behind typsetters here, and most browsers only support ISO-Latin1. This announcement suggests that they may be going go to support much of the ISO entity set which would be very useful. (The glyphs will hvae to be mapped onto a variety of fonts and be scaleable - I suspect this may be patchy in the first instance).
Netscape supporting style sheets in late '96. I guess they had to announce it soon (couldn't be seen to being beaten off the mark by Microsoft) The public beta of Internet Explorer 3 (due this month) will support style sheets (according to the W3C style sheets spec - check http://www.w3.org/pub/WWW/TR/ for technical reports on style sheets,
'Stylesheet' is already an overloaded term. At present they concentrate on fonts and other layout but there is no fundamental reason why they couldn't (say) extend to deciding between whether molecules were inlined in the text or referred to by Roman numerals or whatever. Unfortunately there may be different approaches to stylesheets. There is an ISO standard (DSSSL) which I personally would like to follow since it's a complete programming/search language. (CoST - see below) - is based on DSSSL). Panorama (SoftQuad) already has its own style sheet approach which is different from the W3C one (I think). [...]
haven't already, (see above for W3C technical reports URL's). As for chemists creating style sheets - we'll still be at the relative disadvantage of not being an economically viable market for the likes of Microsoft/Netscape, or anybody on the W3C and so will get style sheet possibilities identical to everyone else. (i.e. for text presentation) Unless someone out there is willing to write a browser.
The technology is making this easier, in fact. The problem is that 'chemistry' covers almost every type of information known - text, maths, numbers, figures, , odours (!), programs, molecular biology and , of course, chemical structures. Therefore a chemical browser has to be a superset of all these disciplines. My own approach - Chemical Markup Language (http://www.venus.co.uk/omf/) is to divide a piece of chemical information into the different types of components. So for text, I pinch HTML (2.0 at present, but I'll upgrade when ever). For figures I will use mime types and/or CGM. Note that CML uses strict SGML so the whole thing is SGML processable. Maths will have to wait till a maths DTD is developed. Personally I am not keen on (say): P{\equals}nRT{\over}V since this is for typesetting. You cannot use it to automatically calculate something - indeed it's not clear even what the variables are. I'd prefer to see (arbitrary SGML-like approach): <MATHS> <VAR UNITS=pascal NAME=pressure>P</VAR> <COMPUTE> <VAR UNITS=mole NAME=moles>n<VAR> <MULTIPLY> <VAR UNITS="joule mole-1 kelvin-1" name="Gas constant">R</VAR> <MULTIPLY> <VAR UNITS="kelvin" name="temperature">T</VAR> <DIVIDE> <VAR UNITS="m-3" name="volume">V</VAR> </MATHS>
From this it's possible to *use* the equation in calculations - costwish (v.i.) can do this to a limited extent.
For this approach we need a generic browser that can deal with a multitude of disciplines. It has to be extensible, as more subdisciplines are added. I have built a prototype (costwish, based on JoeEnglish's SGML engine CoST)) which works on UNIX systems and displays a wide range of 'documents' from SWISSPROT sequences through MOPAC outputs to RSC publications. (See URL above for screenshots). I've released the core of costwish to the SGML community where the bugs are being ironed out and there is hope for a port to NT. My immediate plans are: - to release a revised version of CML (0.7) which will act as a development platform for applicatations (hopefully 1-2 weeks) - to develop costwish as a platform independent, secure adjunct to current browsers; this is moving rapidly and it's an area where we will appreciate input from the chemime community. The tool will be capable of very easy extension, so that if groups have specific applications they should be able to bolt them on very easily. P.
Peter Murray-Rust, Glaxo Research & Dev. (pmr1716@ggr.co.uk); (BioMOO: PeterMR) Birkbeck College, ubcg09q@cryst.bbk.ac.uk, CBMT/Daresbury mbglx@seqnet.dl.ac.uk http://www.cryst.bbk.ac.uk/PPS/index.html, http://www.dl.ac.uk/CBMT/HOME.html ----- chemweb: A list for Chemical Applications of the Internet. Archived as: http://www.ch.ic.ac.uk/hypermail/chemweb/ To unsubscribe, send to listserver@ic.ac.uk the following message; unsubscribe chemweb List coordinator, Henry Rzepa (rzepa@ic.ac.uk)