2010Nov16

From Filtered Push Wiki
Jump to: navigation, search

Reports

  1. UC Davis (Lei and Bertram): Simulate the curation process by using Ptolemy DE (discrete event model of computation). Such simulation could help us understand use case and design the system architecture. Especially, it might be very useful for performance evaluation in the future.
  2. Zhimin: working on use cases; took a close look at google refine, which looks a very good front end for us with some customization; work with Chinua on expanding field list in the demo since we have more fields in GBIF data
  3. Bob: Partially succeeded in installing Protege web version on umbfp.

Agenda

Discussion

Bertram: How to start on bullet (1) from AHM, i.e., Finding Duplicates in Botanical Collections? What is the architecture? (old or new)

Bob: This case fairly well under control/understood; workflows probably not needed there. Instead on quality control use cases (cf. bullet (3) from AHM final day) and below.

Other questions from Bertram: - ...

1.duplicate finding in hebaria collections

2.Scenario for ebird third party "invalidity annotation": Bertram accesses all e-bird records from November 16 2010 in Davis, for birds in Family XXX. a. Bertram first asks for those already marked invalid by e-bird but not reviewed. He then runs workflow W (his own idea of an invalidity test) and launches its outcome (valid/invalid plus reason) into the e-bird FP network. Subscribers interest in XXX or interested in Davis, or... . One interested party is always e-bird central if Bertram is sufficiently authorized and authenticated.

b.Generalization of a. : Bertram has a predicate P other than "valid" and a workflow that in his view computes P on data retrieved from e-bird by criteria set in his workflow. Rest of scenario substantially similar to a.

Bertram (Bob's interpetation): Make it easy for users like Bertram to be able to make their own workflows. One special case is existing e-bird valid/reviewed computation.

- social issue: do they want to evolve / extend their existing curation wf? - technical issue: share special (analysis?) workflow (e.g. Green birds on the West Coast) Bertram: Steve et al interested in niche modeling and visualization workflows

From DataConservancy: Data curation: An ecological perspective https://jscholarship.library.jhu.edu/handle/1774.2/34025

http://dataconservancy.org/home

==> SVN access for Lei and Bertram

http://firuta.huh.harvard.edu:9000/pUsRg0BmoP