Transcript Florida, USA, April 2007 - American Museum of Natural History

Web-based Specimen Databasing:
Lessons from the Plant Bug
Planetary Biodiversity Inventory Project
presented by
Randall T. Schuh
Division of Invertebrate Zoology
American Museum of Natural History, New York
Plant Bug PBI Specimen Database
Approach
•
•
•
•
•
Tailored to entomological/invertebrate
collections
Browser based
Data entry over Internet to central server
Open source software – MySQL database
Multiple modes usable simultaneously
–
–
–
•
Museum Mode
Edit Mode
Report Mode
Unique specimen identifiers, batch loading
Machine/Network Requirements
•
High resolution monitor
•
Broadband connection
Unique Specimen Identification
Justification
• Facilitate specimen tracking
Necessary Attributes
• Machine readability
• Human readability
• Small size of code-bearing labels
- Matrix codes
• Ease of integration into
existing collection practices
 Accommodate unique specimen
and lot-based approaches
Museum Mode
•
Main data entry mode
•
Efficient data entry navigation using TAB key
or mouse
All fields visible in a single browser window
Most effective using Mozilla Firefox browser
•
•
-
Allows for multiple keystroke searching in
drop-down menus
Retains and displays prior entries in memory
Georeferencing
Performed on a centralized basis or by the
user at the time of locality data entry
Tools Used
- GEOLocate
http://www.museum.tulane.edu/geolocate/default.aspx
- On-line Gazetteers
Edit Mode
Allows the user to:
•
•
•
•
•
Check and emend data entered through
Museum Mode
Add higher taxon names, institutions, and
larger scale locality information, which
cannot be added through Museum Mode
Correct taxon names, institutions, collectors,
and locality information
Query, select, and batch-modify specimen
and host information
Allow for unique views on data and images
Report Mode
Output publication-ready specimensexamined data for:
• New species
• Previously described species
Generate downloadable reports for:
• Host lists
• Specimen lat/lon data for mapping
Prepare specimen labels
Admin Mode
•
Create usernames and passwords
•
Allocate access privileges
•
Create database interfaces and
specify settings
Database Outputs & Extensions
• Reports
Publication-ready specimen data
- Publication ready host data
- Specimen labels
• Species pages and mapping via
www.discoverlife.org
• Simple mapping via on-line systematic
catalog
-
Interface/Web Strengths
•
Easy to use
•
Intuitive
•
Incorporates look-up functions to facilitate
rapid and accurate data entry
•
Allows multiple simultaneous users at diverse
locations around the world via web connection
•
Allows simultaneous use of multiple modes
•
Allows for centralized georeferencing
Database Strengths
•
Relational and highly normalized
•
MySQL enterprise-level multithread backend
•
SQL compliant, facilitating data retrieval
•
Highly scaleable with no restrictions on
database size or numbers of users
•
Rapid response time
Lessons Learned
Positive
•
•
•
Web-based approach well-adapted to
geographically distributed team
Web-based research tools improve productivity
through user feedback network
Common problems resolved through singular
solutions
Negative (applying to all software usage)
•
•
Data model allows for duplications due to lack
of familiarity and training
Effective use requires some adaptation on part
of users
Acknowledgments
Special thanks to:
Nina Gregorev
Sheridan Hewson-Smith
Additional thanks to:
James Ashe
Mark Breedlove
Gerry Cassis
David Furth
Lorenzo Prendini
Michael Schwartz
Tom Trombone
Christiane Weirauch
Denise Wyniger
National Science Foundation
American Museum of Natural History
Australian Museum
http://research.amnh.org/pbi