Worldwide Botanical Knowledge Base

http://wwbota.free.fr/                        presentation sponsored by 
Jean-Marc Vanel

GNU-Linux <--> operating systems
WWBKB    <--> biology

XML: no other way!

How did this project come about?

• thesis in  numerical simulations in fluid dynamics
• my jobs:
• fluid simulations : lots of complex structures: meshes, fluid networks, boundary conditions
• financial instruments
• for 15 years I had living collections of plants
• one day a colleague told me that one can find anything on internet: I challenged him to find Cycas: he had 4 hits with Altavista!

Biodiversity and nature conservancy

• thousands of species are disappearing forever,
• need quick identification of specimens from threatened areas, without qualified scientists;
• the knowledge is on paper like in Linnaeus' time,

There is currently no other botanical project with this scope.
Why ?

• funds go rather to biotechnologies than to descriptive biology,
• pluri-disciplinarity is not easy,
• laboratories are in a position of concurrence
• past efforts from biologists (e.g. Delta, Lucid) have been software first, data afterwards

Specification

• description of species, including 2D and 3D pictures,
• computer-aided identification of specimens,
• geographical distribution,
• distributed knowledge, links with other disciplines
• zoological: pollinating, disseminators, parasites, herbivores
• genetics biochemical: e.g. correlation between molecules and taxonomy
• paleontological
• ecological - phytosociological, pedological, climatic: search of correlation, analyze of vicarious species
• agronomy, plant uses, ethnobotany

A free sofware / free information project

• nobody can own nature
• a great project for humanity, and
• a great, far-reaching, and enjoyable software project
• federate existing resources and people

Collaborators

• Flora of China ( http://flora.harvard.edu/china/ ),
• Flora of Australia
• Taxonomic Databases Working Group
• Kew Botanical Gardens
• Museum National d'Histoire Naturelle, Paris

The data exists

• more than 90% of the 250 000 species are on Floras on paper, and lots are on the Web in HTML (not XHTML!);
• herbarium images; also many are on the Web (e.g. 30 000 species in Florida)

Initial intuitions:

1. relational DB
2. uncouple data (XML) and processing: XML engines, relational DB, OO DB, AI engines
• Web search engine, e.g.:

  +species: +flower:yellow +bark:white +sharp
3. offer easy access to detailed knowledge both for the layman and the professional botanist

The metadata

• Why a relational DB is not enough:
• part-of relations: e.g. bark is a part of stem (and trunk)
• inheritance of properties from upper taxonomic classes: e.g. "oppposite leaves"
• different levels of metadata
• several sources of metadata: Floras, Wordnet, botanical lexicons
• Abstract Data Model: UML diagrams
• uses of metadata: GUI, reasoning (description logic), specify messages

New kind of software needed

• multi-domain documents wrapped in XHTML
• calls the appropriate processors when it sees certain XML namespaces and/or Processing Instructions.
• book-keeping: management of:
• drag'n drop and clipboard with an XML data model (DOM object passed)
• display space between XML processors (tiling, resize, ...)
• mapping between raw XML and displayed XML transformed by XSLT
• Generic display skills are also desirable:
• collapsable tree/graph views for:
• document tree
• inheritance graph
• the ID/IDREF graph
• file/directories, URL/hyperlink graphs
• extended search/query
• editor with the same multi-domain capabilities.
• using a standard dictionary (e.g. wordnet) and some AI techniques will enable to treat well-formed XML with natural language tags: disambiguation, tagging while typing

• document-oriented (HTML & word processor)
• structure-oriented (database type)
• knowledge-oriented (semantic network, AI, RDF, etc)

WW B KB  building works

2D and 3D images

• vectorized images (SVG) from bitmap images
• 3D images (X3D, etc) generated from several pictures through stereoscopic software
• very compact representations adapted to growing beings having recursive structure: L-Systems, MathML, etc
• Artificial Vision techniques

Botany, Zoology, Ecology and the Semantic Web

• express relations between plants and animals in a formal and flexible way (RDF/XLink)
• plants (250 000) are just a beginning: insects alone have 1 000 000 species
• all kinds of properties, and properties about properties
• XML is not just to catalog consumers's preferences and locate the cheapest merchandise

Distributed knowledge on the Web, XML protocols

Vision: field observations

Conclusions

• $£$£$£$£$£$£$£$£$£$£$£$£$£$£$£$£$£$£$£$
• need collaborators, need contacts
• can do great things with little resources, like the physicists in the 19th century
• make standards for 20 years or more ==> clean & flexible design, track standards and implementations
• a new way to make science: GNU spirit

Issues

• AI techniques, XML exchange formats & protocols for AI
• new ways of working for taxonomic scientists: new species created by Web transaction
• Problems with existing standards
• overlapping concepts: RDF & Xlink, RDF Schema & XML Schema, XSLT & X-QL
• non-XML syntaxes: XPath, CSS
• many existing vocabularies are monolythic (e.g. X3D, XML Spec)
• little reuse of Schemas and concepts
• reinventing the wheel; schemas are longer to read and understand
• little modularity

___________
 
 
 
 
 
 
 

structure the presentation as a semantic network