We are trying to fill the gap between
-
Delta, very formal, but few data available,
-
classic floras, less formal, but almost exhaustive
Data flow
When the botanical descriptions have become a marked-up XML document (see
parsing),
we can :
-
export to Delta
-
export to a relational database
-
export to a Object database
-
process it in a server with the W3C's Document Object Model, the Xpath
query language, and the XSL style language
-
apply the same processing on a browser (today, mostly with Internet Explorer
5)
-
use some AI techniques to help in identification of plant specimens
Issues
-
several servers, specialised in taxonomic groups (families and genus),
or in geographical zones, details of the HTTP protocol to dispatch an all-world
request and gather the results for the client;
-
descriptions in several languages,
-
treatment of taxonomic synonyms (marked by an <accepted> tag
?),
-
meta-data: semantic net: synonyms, near-synonyms, generalization, containment(agregation);
example: bark, branchlets, trunk, stems would all be concerned by a general
query for "bark contains(yellow)"; in fact, to achieve understanding beetwen
different vocabularies in the same domain, there is a need of a semantic
net, especially synonyms, near-synonyms
-
detailed object model outside species description:
-
plant, sample, living collection sample, herbarium sample, description;
-
institution, herbarium, arboretum, botanical garden, etc;
-
person, botanist, collector, etc;
-
character, organ;
-
taxon, family, genus, species, etc;
-
association, vegetation types, phytosociology, etc
-
RDF, RDF-Schema, knowledge representation (KR) techniques