DC comments & questions

[Robert Hanisch]

The counterargument that Don Wells has made over the years concerning the
FITS format, is that if you do not use the archival/standard format
directly, then you are less likely to produce truly conforming data.  I
think this situation has been seen already, i.e., when people convert from
various internal formats to FITS, they get things somewhat wrong more often
than when they use FITS directly, as it is then just an "export" format.

Of course, the whole VO initiative is about interfaces and standards.  So,
though it seems to me that it makes a bit more work for resource providers
to maintain dynamic conversion utilities than to encode metadata to DC
directly, at this point I am more concerned that we do Something rather than
debate the merits of one way or the other.

Bob

----- Original Message ----- 
From: "Tony Linde" <ael at star.le.ac.uk>
To: <registry at ivoa.net>
Sent: Sunday, October 05, 2003 12:07 PM
Subject: RE: DC comments & questions


> One more comment before I go do something interesting for the last couple
of
> hours of the weekend...
>
> I was just checking out OAI and in the 'Guidelines for Repository
> Implementers' at
> http://www.openarchives.org/OAI/2.0/guidelines-repository.htm it states:
>
> 'Repositories are not required to store their metadata in DC; DC is
> something that is frequently "converted to", rather than "stored in". Many
> repositories store their metadata in some other format, and dynamically
> convert to DC in response to harvester requests.'
>
> Which is pretty much what I have been arguing for.
>
> Cheers,
> Tony.