2011 working group Th VegBank conference call¶

• Mike Lee; SB: Mark Schildhauer, Nick Spencer, Aaron, Jim Regetz, Bob Peet, Steve Dolins (CS prof), Brad Boyle (later?), Shash
• Mike: program-director for VegBank (11 yrs)
• db infrastructure
• broad-based, community-oriented system rather than internal to group
• discipline-wide resource
• need central core db to house data
• architecture that holds plant, plot records
• VegX mechanism to port things in and out
• core will be designed around VegBank architecture
• technology of VegBank out of date
• tweaks to VegBank:
  • doesn't hold all DwC stuff
• clarify: recent db updates, framework version to VegBank?
  • ask Nick Brand
  • updated Postgres to newest version
  • problems with ant being a new version and working slightly differently
  • moved over to svn
• Mike mostly debugging of Java/Spring code
• modernizing the model: db platform and design itself?
• schema modification: Postgres
• PostGIS w/ spatial operators
• focus on replacing Java web app framework
• store aggregated data, not just raw obs
• store DwC data
• modeling of individual trees, stems
  • attrs: taxon names, stems
  • specimens, aggregate obs plots, indiv obs plots
• individual stem measurements
• large # of tables to see data
• need recursive loop for stems to point to individual they're part of
• herbarium labels
• complexity of dealing with stems through multi-table version
• combine stems, aggregates into flattened table
• taxonInterpretation and stemLocation
• link stemLocation directly to observation
• vouchers: plots where specimen vouchers connected to them
• specimens in taxonInterpretation table
• treat each voucher in plot as subplot with single organism
• CVS based on VegBank
• split up plots into
• recursive relationship on plot and observation
  • observation link: timeseries link
  • plot link: spatial subplot link
• plot, voucher have many same attrs: country, state, lat/long
  • make attrs live in same place
• management issue
• data largely static once entered
• relationship betw plot, plot observation
• not bringing along UI, so can restructure schema
• add table/change links
• access to Mike
  • ESA pays some time to maintain VegBank
  • not VegX-VegBank linkage
• NZ group revive VegX-VegBank?
  • db is similar to VegBank
  • getting data of VegBank into NVS
• if keeping VegBank as is, large undertaking
• share between VegBank and new tool
• capabilities of VegBranch
  • client db for VegBank
  • takes info and imports into simpler tables, do error checking, standardize names, missing metadata
  • moves data into VegBank impl in Access -> export to VegBank XML format
  • dematrifying data
• limitations of VegBranch
  • fully compat w/ VegBank w/ all its complication
  • bells and whistles in data model are distraction
  • simpler tool
  • link to NPS plots dbs and pull data in
  • NPS now using new system
• user base for VegBranch: small
• have lots of plots: 72,000
• CA heritage program has 30,000 plots

Whitepaper¶

• when finished in a year, what will have been done
• hinge on VegX
• vision of pieces
• separate from overall BIEN whitepaper
• build on tech whitepaper in Dec
• outline form of Brad's notes
• skeleton requirements paper
• data end products key requirements
• constantly refer to doc
• detailed specimen descriptions
• roadmap
• will VegBank do the job?

• taxonomy tables
• db walkthrough with Mike
• important to include explanation in name of table
• stemCount records individual organism
• snippet of populated data
• ask Mark for access to VegBank database
• data dictionary: derived from database or a story item
• Visio to reverse-engineer data dictionary
• ER studio auto-generates data dictionary
• e-mail Mike