Christina's LIS Rant
Tuesday, June 17, 2008
SLA2008: Cyberinfrastructure Informatics Across the Biological Sciences
Biodiversity Informatics: Evolving in the Biological Sciences
Biodiversity Heritage Library
Cathy Noorden, Woods Hole/Marine Biological Library
tuatara - outside has stayed the same but inside has changed completely, most quickly evolving animal... so a good metaphor for librarians.
Part of the Encyclopedia of Life project. Group of 10 museums that are scanning the taxonomic literature to form the base of the EOL.
Importance of all materials. Not just one type of discovery tool, not only libraries, equal group of contributors.
Goal: core information open access on the web. Can do this because early information - first time something described - still very important to taxonomists.
Partners: internet archive, publishers as content providers, copyright holders.
Domain: 5.4M books dating back to 1469, 50% pre-1923 so out of copyright. But also non-profit society journals who can't afford to digitize agree and also get copies for their page, and some agreements with commercial publishers.
How do they differ from google? Taxonomic information - including what has changed, different languages, historic information. Other organizations need this too - PubMed doesn't have all of these names. Example: this particular salamander, many different spellings so you would not get all of the articles. Reconciliation, link to alternative names for the same organism.
uBio - Universal Biological Index and Organizer. "Taxonomic intelligence is the inclusion of taxonomic practices skills and knowledge within informatics services to manage information about organisms" Giant index with 10.8 million names and merging maps - including common names. Linkage to other data types (molecular, morphological, phenotype). Find it scientific name recognition algorithm (after scanning and OCR-ing) - fairly easy starts upper case then lower case and in italics. Training and improving algorithm.
uBioRSS taxonomically intelligent RSS feed aggregator - their lab works on 200 organisms, so they go out and search selected sources on those organisms. Also present search results for their authors' articles right on the home page.
See also poster session tonight for more details.
Q: zoo record? A: yes, Q2: images? A: yes, and others, too, but rights are slightly more difficult.
Ecological Informatics: Building Solutions for Multi-Decadal Research.
Dr. Bill Michener, University of New Mexico
Long Term Ecological Research (LTER), http://www.lternet.edu/
LTER is required for slow or transient processes, episodic or infrequent events, trends, multi-factor, delayed effects.
Issues: data dispersion, field stations, museums, local agencies, individual collections - data entropy - basically, scientists lose or forget details about their datasets after publication... Data integration- different syntax (format), schema (model), semantics (meaning).
ecological informatics- status: some data archives, some tools (such as morpho, meta cat, kepler - a workflow system). Vegbank- value added database, societal and NSF effort.
future: Smith, Knapp, Collins in press - the global change hockey stick applies to population, temp, environmental nitrogen...
coupled science and cyberinfrastructure, facilitate access and use (usability of preserved data). Global communities of practice. Diverse scales and scopes. Enabling the scientists, whole life cycle data management, domain agnostic solutions....
citizen science toolkit - citizenscience.org
Empty Suits
Quentin Wheeler
(ppt issue so no slides)
Creating the taxonomic content for taxonomic knowledge - lack of funding for taxonomic work based on mistaken idea that we're done or can rely on history and historical data.
We only know 10% species on earth and we might lose 25-50% of the species within this century. How can we detect invasive species or find bioterrorism if we don't have a biodiversity baseline. Major discoveries every year in biodiversity (biggest stick bug found).
Still major, heated, furious debate on what a species is - still more than 2 dozen competing definitions. Tradition of argumentation... biological challenge and process challenges - fundamental problems. Disservice to turn taxonomy into a service for other areas of science instead of as for a basic science. More money into mobilizing bad data (databases) instead of generating good data and capturing new good data as it's created.
NSF - Planetary biodiversity inventories - required communities of taxonomists work together (new idea, generally taxonomists are very independent). Example: cat fishes - 200 ichthyologists working together - now can work together. Also information access in smaller or remote institutions. Remote microscopes.
Challenges - science is not the issue (phylogenics has had a huge impact, and they have rigorous testable hypotheses)... it's more the culture of not collaborating, and the need for support through cyberinfrastructure.
Next steps: more on bugs. more on history and philosophy of science (to look at how marginalized to prevent future), sociologists to help work with rewards structure, change image of taxonomy, advertising...
saw video: Planet Bob.
Biodiversity Heritage Library
Cathy Noorden, Woods Hole/Marine Biological Library
tuatara - outside has stayed the same but inside has changed completely, most quickly evolving animal... so a good metaphor for librarians.
Part of the Encyclopedia of Life project. Group of 10 museums that are scanning the taxonomic literature to form the base of the EOL.
Importance of all materials. Not just one type of discovery tool, not only libraries, equal group of contributors.
Goal: core information open access on the web. Can do this because early information - first time something described - still very important to taxonomists.
Partners: internet archive, publishers as content providers, copyright holders.
Domain: 5.4M books dating back to 1469, 50% pre-1923 so out of copyright. But also non-profit society journals who can't afford to digitize agree and also get copies for their page, and some agreements with commercial publishers.
How do they differ from google? Taxonomic information - including what has changed, different languages, historic information. Other organizations need this too - PubMed doesn't have all of these names. Example: this particular salamander, many different spellings so you would not get all of the articles. Reconciliation, link to alternative names for the same organism.
uBio - Universal Biological Index and Organizer. "Taxonomic intelligence is the inclusion of taxonomic practices skills and knowledge within informatics services to manage information about organisms" Giant index with 10.8 million names and merging maps - including common names. Linkage to other data types (molecular, morphological, phenotype). Find it scientific name recognition algorithm (after scanning and OCR-ing) - fairly easy starts upper case then lower case and in italics. Training and improving algorithm.
uBioRSS taxonomically intelligent RSS feed aggregator - their lab works on 200 organisms, so they go out and search selected sources on those organisms. Also present search results for their authors' articles right on the home page.
See also poster session tonight for more details.
Q: zoo record? A: yes, Q2: images? A: yes, and others, too, but rights are slightly more difficult.
Ecological Informatics: Building Solutions for Multi-Decadal Research.
Dr. Bill Michener, University of New Mexico
Long Term Ecological Research (LTER), http://www.lternet.edu/
LTER is required for slow or transient processes, episodic or infrequent events, trends, multi-factor, delayed effects.
Issues: data dispersion, field stations, museums, local agencies, individual collections - data entropy - basically, scientists lose or forget details about their datasets after publication... Data integration- different syntax (format), schema (model), semantics (meaning).
ecological informatics- status: some data archives, some tools (such as morpho, meta cat, kepler - a workflow system). Vegbank- value added database, societal and NSF effort.
future: Smith, Knapp, Collins in press - the global change hockey stick applies to population, temp, environmental nitrogen...
coupled science and cyberinfrastructure, facilitate access and use (usability of preserved data). Global communities of practice. Diverse scales and scopes. Enabling the scientists, whole life cycle data management, domain agnostic solutions....
citizen science toolkit - citizenscience.org
Empty Suits
Quentin Wheeler
(ppt issue so no slides)
Creating the taxonomic content for taxonomic knowledge - lack of funding for taxonomic work based on mistaken idea that we're done or can rely on history and historical data.
We only know 10% species on earth and we might lose 25-50% of the species within this century. How can we detect invasive species or find bioterrorism if we don't have a biodiversity baseline. Major discoveries every year in biodiversity (biggest stick bug found).
Still major, heated, furious debate on what a species is - still more than 2 dozen competing definitions. Tradition of argumentation... biological challenge and process challenges - fundamental problems. Disservice to turn taxonomy into a service for other areas of science instead of as for a basic science. More money into mobilizing bad data (databases) instead of generating good data and capturing new good data as it's created.
NSF - Planetary biodiversity inventories - required communities of taxonomists work together (new idea, generally taxonomists are very independent). Example: cat fishes - 200 ichthyologists working together - now can work together. Also information access in smaller or remote institutions. Remote microscopes.
Challenges - science is not the issue (phylogenics has had a huge impact, and they have rigorous testable hypotheses)... it's more the culture of not collaborating, and the need for support through cyberinfrastructure.
Next steps: more on bugs. more on history and philosophy of science (to look at how marginalized to prevent future), sociologists to help work with rewards structure, change image of taxonomy, advertising...
saw video: Planet Bob.
Labels: biodiversity informatics, SLA2008
¶ 11:52 AM| |cites (technorati) |
Comments:
Post a Comment
