[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Satellite Data Image Conventions

> Organization: NCAR / MMM
> Keywords: 199502282325.AA15817

Hi David,

> One of the ideas I had at the back of my head when I was talking to you 
> and Joe last fall was to try to find out whether there was (still) enough 
> commonality between netCDF and TDF that at least some subsets of either 
> data system might be functionality equivalent.  For example, Joe already 
> has what he calls an "instant" routine (for "instantiate") which resolves 
> many of the advanced TDF features such as variable links and external 
> references that may not have an equivalent in netCDF.  Even if a full 
> implimentation of TDF supports things that are foreign to netCDF it 
> seemed to me that the most basic version of a data file wouldn't necessarily 
> have to make use of all the features of the complete data model.   Thus 
> through a routine very similar to "instant", you might be able to produce 
> a fully acceptable TDF file that is still compliant with netCDF (with the 
> possible exception of a data header or the like).  Similarly, having some 
> form of a fully acceptable netCDF data set, perhaps with some restrictions 
> on data packing or the like, that would also be acceptable to Sea Space's 
> Terascan software (and TDF implimentation) is real attractive.  That would
> facilitate data exchange, without requiring either data system to fully
> support all the features of the other system and with the need for a formal
> file conversion filter.  In that case, cross data set compatibility could 
> be consider a user "option" that could be invoked, if desired.  

I think the fact that netCDF encodes all the data using XDR whereas TDF
doesn't use XDR makes it impractical to make use of data among TDF- and
netCDF-based applications without an actual file conversion program.  Every
netCDF dataset has all its values as well as the header information encoded
using XDR.

--Russ