This email may just show my naivety, but:
All this discussion about how to accommodate web spiders and their inability to trigger cascading builds of treatments brings to mind two recent discussions outside TDWG - maybe there will be a solution here. It seems to me that there is a problem if we constrain our development to using off-the-shelf web systems rather than create our own.
1. If we ever get this standard thang resolved, can we create our own web spiders with the characteristics we need (e.g. some knowledge of the standard wrt version numbering etc, and the ability to interact with a treatment to trigger events). If this were possible, we wouldn't need to rely on standard search sites (Altavista etc) but could have our own search engine with reports from our own spiders. The standard would include a tag to make it identifiable to the spider. This may be a big job, but surely big is what GBIF is all about. David Yeates (University of Queensland entomologist) and I some time ago were toying with this idea. We called it the BioScape, being a view of Web content tailored for biologists.
2. At a recent NSF/ARC-funded meeting at Manassas near Washington, a group of Australians working in bioinformatics and their American counterparts (inc. Bryan Heidorn, Jim Croft, Stuart Poss) discussed some of the big questions and impediments in the field, and opportunities for collaboration. In the descriptive data part of that meeting we developed the idea of a Taxonomic Workbench, created by a New Taxonomy Collective (this lacks only a manifesto, but that could be arranged). The TW would be a site that taxonomists log onto to do their magic, including capturing, storing, sharing and manipulating descriptions. It would be furnished with server-app tools for these jobs. The gene jockeys have analogues of such a workbench for their work, and these are very successful.
Our still-pre-incipient standard would be an important part of this. Again, developing the TW would be a big a big job, but if such a thing is necessary to make it work, then so be it. We would need to report this to TDWG, and boldy go forth.
Cheers - k
participants (1)
-
Kevin Thiele