2012-04-09 conference call¶

• Table of contents
• 2012-04-09 conference call
  • To do
  • New data sources
  • Geo-validation
  • Martha's notes

To do¶

1. reload all datasources except those excluded by Brad
   • investigate host vs. VM performance first?: VM is actually 9% faster
2. import CTFS VegX raw data instead
3. parallelize Python import scripts using column-based import instead
   • split time between this and importing CTFS
   • will allow SpeciesLink to be imported much faster by using all cores at once
4. automate validation of new data sources
   • replace column names for each datasource in a SQL validation script that uses DwC2 names
5. serialization of VegX
   • CSV? JSON?
6. find John Donoghue's geo-validation scripts

New data sources¶

• quality over quantity: do existing sources well first
• no new plots data until we have all BIEN2 data in VegBIEN

1. TurboVeg
2. U.S. National Parks data from Brad
3. Argentina/-Chile-
4. GBIF reimport
   • talk to Dave Remsen
5. BCI DiGIR servers

Geo-validation¶

• sources of geographical names
• spelling errors in names
• names in different languages for the same place
• shapefiles (also correct their spelling errors)
• Biogeomancer

• TNRS: PHP/MySQL from Brad on nimoy
• Geoscrub: Python/PostgreSQL/R from John on eos, and PHP/MySQL from Brad on nimoy

Martha's notes¶