Hi Tim (et. al.),
A slew of good comments and ideas in your response (Oct. 12). Here's a
suggestion for how we can incorporate them into the strawman: Those comments
that represent new issues (especially all of the ideas on how to represent raw
cruise data) we need to condense into specific suggestions. The kind of thing
I'm getting at is this style: If Sue Schmue at the Acme Oceanographic
Institute collected her raw data in netCDF files what conventions would she
need to observe in order for her files to be intelligible to your software?
You raised a number of subtle points about the breakdown of simple distinctions
between independent and dependent variables (even "latitude" becomes ambiguous
- e.g. do we want "GPS_lat" or "SatNav_lat"?). How do the dependent variables
map to the independent variables: If your ship was towing a "fish" you would
have an associated variable "depth". Should your file convey that "depth" is
an independent axis of "fish_temp" but not of, say, "ship_temp"? If so, how?
(Some conventions may not effect the content of the file, just the
interpretation of it.)
For your comments that are discussions of the numbered issues in the straw man
(e.g. length of variable names) I would prefer to file them away for now and
bring them back in "phase 2" for detailed discussion (and the daunting task of
reaching consensus).
| NOAA/PMEL | ph. (206) 526-6080
Steve Hankin | 7600 Sand Point Way NE | FAX (206) 526-6744
| Seattle, WA 98115-0070 | hankin@xxxxxxxxxxxx
