Hi,
For your information, I've appended a posting from the netCDF mailing
list pertaining to an XML representation for netCDF files.
Although this is not strictly related to the XML question, I think it
might be worthwhile to think about a "standard" means for embedding a
"default VisAD mathtype" directly in netCDF files (and consequently,
of course, in any XML representation that emerges). My reasoning is
that this could expand the expressive power of the netCDF for storing
VisAD data objects.
More specifically, I think one or more (reserved?) netCDF attributes
could be used to convey VisAD "mathtype" metadata, probably as text
strings of the form produced by VisAD. Utilized in VisAD's export
utilities, this could permit VisAD applications to re-import the data
and recover the original mathtype. Furthermore, the presence of these
attributes--whether generated by ViSAD, by some other software, or by
a human being--could increase the power of a standard netCDF adapter
to import netCDF files into the VisAD world for display and analysis.
Unfortunately, I don't understand this subject well enough to be
certain of these potentials. Hence, I welcome feedback and (if my
idea makes sense) offers to participate in the development process.
Dave
P.S. I am not asserting that this concept addresses all discrepencies
between the netCDF and the VisAD data models. For example, the latter
can express recursive data types, but the netCDF model cannot; my idea
makes no attempt to resolve this.
=======================================
Dave Fulker
Unidata Program Director
University Corp for Atmospheric Research
----------------------------------------
Mail: PO Box 3000
Boulder, CO 80307-3000 USA
Email: fulker@xxxxxxxxxxxxxxxx
Phone: 303-497-8650 Fax: 303-497-8690
=======================================
---------- Forwarded Message ----------
Date: Monday, 3 April2000 18:34 +0200
From: Lorenzo Bigagli <lorenzo_bigagli@xxxxxxxxxxx>
To: netcdfgroup <netcdfgroup@xxxxxxxxxxxxxxxx>
Subject: XML?
Hi,
we are considering implementing an XML interface to NetCDF data.
Is anybody aware of others' efforts in this directions?
It could be interesting to reach a common definition of a NetCDF-DTD...
Thank you,
Lorenzo Bigagli
--
Lorenzo Alessandro Bigagli
***********************************************************************
e-mail: lorenzo_bigagli@xxxxxxxxxxx cell.: +39 (0)335 8078555
home: Via Brugnano 5, 59014 Prato, Italia phone/fax: +39 0574 813210
work: PIN, P.zza Ciardi 25, 59100 Prato, Italia phone: +39 0574 602535
ICQ#: 10863236
---------- End Forwarded Message ----------
