From Filtered Push Wiki
Jump to: navigation, search

1. recap demo at GBIF

2. brief discussion from Zhimin? of http://cloudbase.sourceforge.net/#homePage Tim suggested looking at it, but I am not sure. Maybe for a fp cache? If possible, everyone should browse a bit on the cloudbase sourceforge site.

3. Set goals for next few weeks

4. Sketch by Maureen of any Specify issues (especially if Jim or Rod join us).

5. Medical records applications. Sylvia Spengler has been sending mail with pointers to NIH (in general).

6. IPT

1. Discussion of Morphbank as a client and maybe server of image access metadata regarded as an annotation of a specimen.

2. another from facebook, and yahoo. vs HBase they are database warehouses. Maybe hive? Paul, Zhimin, need global network knowledge and analysis different from transactional dbs with the local data. Performance requirement of the Find-duplicates use case suggests either a separate local db for local cache, or a distributed network store that can intelligently cache likely answers to the find duplicates for this collector/number query near relevant nodes for very rapid retrieval. Other use cases require a distributed global store of annotations that doesn't have the high performance requirements of the find-duplicates use case.


4. (No FP work since last meeting)

5. Medical records

6. Paul: maybe FP as a qc message service to provide for users to generate fp messages for providers. Possibility of adding a FP annotation injection client into IPT, with IPT providing the UI for the creation of a semantic annotation of a specimen record served through IPT and IPT injecting this annotation into the FP network. IPT might also make a platform for viewing annotations in the network store.

Next meeting Thursday 12 March at UMB