Activity for Christopher Barker
-
Christopher Barker posted a comment on discussion Open Discussion
hmm -- I'm inclined to think that the the record dimensions are an impiimentation...
-
I think consisentcy with ncML is good, too.
-
every "option" makes more work for the reader -- probably not that much work in this...
-
As I said (or implied), I prefer un-typed attributes (let JSON map them to what it...
-
near real time conversation! maybe not needed, but some more thoughts I already write...
-
BTW: how does CDL deal with typed attributes? If CDL doesn't fully handl eit, then...
-
BTW: how does CDL deal with typed attributes? If CDL doesn't fully handl eit, then...
-
Interesting -- I honestly had no idea attributes were typed -- I don't think I've...
-
Funny -- if I needed to get netCDF meta data into a DB with Python, I'd jsut use...
-
I think the separate object for attributes is orthoganal to the type of attributes....
-
except that the attribues of the variable should be its own JSON object: I agree...
-
netCDF dimensions is a JSON object with key "dimensions" and value a list JSON objects...
-
netCDF dimensions is a JSON object with key "dimensions" and value a list JSON objects...
-
Is NCO going to read JSON, too? That's where you'd get a real benifit from a library....
-
Personally, I do not want to see a group explicitly labeled "root". If those who...
-
The major remaining design question on which I would like feedback is whether variables...
-
@Charlie Zender wrote: """ 1. We agree that bracketing rather than flattening arrays...
-
Looking at the group example now: Same thing, of course for the variables object...
-
Taking a quick look, and a few thoughts I'd like to see it nested a bit more -- i.e....
1