Adam Hodgkin wrote;

One of these Portals needs to do a deal with CAS or Reed Elsevier and start offering specialised chemistry content searching function. Which will it be? AltaVista, Excite, InfoSeek, Yahoo, etc?

What technology will they use? XML, simple text searching with metatags or some database derived technology? Some portion of the chemical community ought to show them how to do it. Or could this service be provided by a specialist Chemistry Portal. In which case ChemWeb (the other one) looks like the best bet. Any views?

To some extent, I addressed this issue in my Webmasters IV talk recently ( http://www.ch.ic.ac.uk/talks/wm4/, an issue not actually mentioned in the report of the meeting http://www.chemsoc.org/gateway/webarticle.htm which cites me out of context ) The essence of my talk was that HTML, whilst it has been comprehensively indexed by eg Alta Vista, has not served the chemistry community very well, and that with HTML "pollution" offered by automatic markup using eg Office 97/98 etc where style is often seen as more important than content, things were going to get a lot worse if the HTML route were to be continued as it is at present (take a look at HTML generated by Office to see if it enhances the chemistry. I have just processed a whole bunch of Word documents for the ECHET98 conference, and was quite underwhelmed) I offered the thought however that there was hope on the horizon, via the new generation of markup specifications such as XML and CML (for a presentation by Peter, see http://www.xml-cml.org/ ) with perhaps an interim solution of much more widespread adoption and use of good generic meta tags (the Dublin Core) and chemical ones (Dublin Chem,, a project currently being formulated). I dont really believe that solutions such as Chemweb.com are the way forward. Apart from anything else, these must still be "human based" solutions, in much the same way that CAS employ 1500 people in Columbus to abstract chemistry. This is EXPENSIVE (and Elsevier also have tended to be expensive!) I doubt ChemWeb can track things better than CAS. The accepted paradigm is that the only way to get good quality search results is to pay a lot of money! It does not have to always be like that! If the 3.8 million "chemistry related" Web documents (plus probably 5 million GIFS containing the actual chemistry) out there were to be replaced by 3.8 Million XML/CML derived documents, then a search engine could produce VERY much better results, not merely for chemistry, but most importantly, for relevant links between chemistry and other disciplines. So the question is: which comes first? Persuading people to write in XML/CML, or at least use good meta tags, or showing how an engine such as Alta Vista could be persuaded to improve the quality of its chemistry retrieval? As I noted in my previous posting, the Netscape "what's relevant" function seems a (small) step in the right direction, perhaps a step that could have been much larger if they had actually made any use of meta tags. Dr Henry Rzepa, Dept. Chemistry, Imperial College, LONDON SW7 2AY; mailto:rzepa@ic.ac.uk; Tel (44) 171 594 5774; Fax: (44) 171 594 5804. URL: http://www.ch.ic.ac.uk/rzepa/ chemweb: A list for Chemical Applications of the Internet. To post to list: mailto:chemweb@ic.ac.uk Archived as: http://www.lists.ic.ac.uk/hypermail/chemweb/ To (un)subscribe, mailto:majordomo@ic.ac.uk the following message; (un)subscribe chemweb List coordinator, Henry Rzepa (mailto:rzepa@ic.ac.uk)