Unit Dictionary Virtual Update VTC - 9/2/2008

A. IM Conference Call - Unit Dictionary - 9/2/2008

B. In attendance
1. cast of thousands - 14 signed up!
C. Update
1. see Powerpoint sent by email
2. revamping from last effort - lessons learned
3. goal - central repository
4. discusison here, not repeating PPT
a) repository on Servilla test server
b) Baker group has been doing some web services etc. demos will
be at EIMC meeting in ABQ
D. Discussion
1. who should make the call on good units?
2. what did you mean by "central data base would be scoped for
a) each unit is defined by a site
1) NWT scope for Niwot units
2) then I'd like to be able to bump them up as more sites use
3) not sure how to define units that more than one site uses
3. idea is to clean up what we have now....
a) best practices will help
b) idea is to link up duplicate units
4. I use custom unit, plus alias to what PIs are using
5. is there a way to have an attribute in STML that could link
back to what is in database? Would like to have some ID that
more directly links back to project.....
a) right now everything is linked to document ID
6. will Karen be presenting later?
a) contacted her, and she's participated in internal VTCs
7. Vetting process?
a) have some general ideas from discussions within group
1) digg method - popularity vote
a> who would vote?
1> whoever is using unit dictionary
2) unit group
3) unit NAZI - one person makes call
b) varies between sites so we need to help find what people are
doing at each site
8. role of scientists?
a) ideally yes....
b) this will be major topic at EIMC meeting
c) bring up in Governance Working Group to set broader
9. How will unit dictionary be USED? Tool for picking units
a) Karen Baker's group has been doing applications
1) plug and play lists via web services
2) PHP and PERL to access unit dictionary
b) larger focus on educational tool
10. using for locating data?
a) needs to be linked to attributes
1) kgN/ha/yr is ammonium-nitrogen would be attribute and
kg/ha/yr is the unit
2) but how do the SCIENTISTs do it?
a> how will we make attribute descriptions match up...?
b> often its left out of documentation
11. Units may be separated completely out of EML in the next
a) conversations with EML developers?
b) Margaret is EML liason.... and she is not here!
c) some discussion with STTML group
d) likely to keep STTML in next version but maybe not with
standard units
e) or extend EML standard list
12. hard to understand how BROAD the problem is with sites using
similar or different units.... What will form of report or
summary look like?
a) right now its all in an EXCEL file
b) 19 sites have custom units defined....
c) can be delivered in many ways....
13. should be easy for sites to spot similarities and identify
places where custom units could be merged
a) need to bring experts together
b) expect iterative process....
c) will do report at IM meeting
14. lots of people have custom units with attributes included in
15. goal for unit part of EML - makes more similar, but can't
search on units to pull up similar data.....
a) e.g., we measure shoots of plants - is that a count? or a
number of shoots
b) will allow unit CONVERSIONS not necessarily searching
1) but requires fully defined custom units
c) but without attribute information can't really do
conversions - ammonium vs ammonium_carbon
d) using STML allows unit conversion but we are dumping more
unstructured information in attribute
e) lays a big load onto attributes
16. how often will the dictionary be updated from EML?
a) not clear
17. one issue on disaggregation of units from attributes is that
many scientist consider what we are saying is an attribute
as part of a unit..... Need to be able to accommodate that