On 1/3/11 12:48 PM, Rich Signell wrote:
> the problem that we were trying to solve
> with "ocean_binary_mask", which was to make it easy for providers to
> make their data CF-compliant.
I'm still confused -- I write code that both generates and reads netcdf
files -- as a rule the code to read generic files is much harder to
write -- because it's generic. You have to write a bunch of code to
figure out what is in the file, and what it means to your application.
This is greatly simplified by the CF standard, but the more duplication
in the standard, the harder it is to use.
On the other hand, when writing a file, once I've taken the time to
figure out what the standard name is for a quantity, and written the
code to write out that variable, adding a "not" to swap the ones and
zeros would be negligible effort.
To quote the "Zen of Python" (
http://www.python.org/dev/peps/pep-0020/):
"There should be one-- and preferably only one --obvious way to do it."
on the other hand, it also says:
"Although practicality beats purity"
So if there are a lot of "sea_binary_mask" files already in the wild,
then so be it.
All that being said, I like Steve's suggestion. In fact, in thinking
about this, I took a look at the THREDDS interface to Rob Hetland's ROMS
model for the Texas coast. There is a mask variable for rho, U, V, and
psi. I think Steve's suggestion would be a great way to handle that, as
I suspect the mask is the the same for all of those quantities.
Interesting that there is no explicit land-sea mask that I can find, by
any name.
-Chris
--
Christopher Barker, Ph.D.
Oceanographer
Emergency Response Division
NOAA/NOS/OR&R (206) 526-6959 voice
7600 Sand Point Way NE (206) 526-6329 fax
Seattle, WA 98115 (206) 526-6317 main reception
Chris.Barker at noaa.gov
Received on Tue Jan 04 2011 - 10:11:39 GMT
