It seems there is still no consensus on metadata. I know this has been
covered, but I am really at a loss for what I would do with the proposed
metadata implementation.

In thinking about it, I did come up with one idea. Potentially, a
metadata implementation might simply be a schema for a collection that
is able to be queried. Further, if a schema was attached to a collection
one might force validation of documents in that collection.

The above seems simple enough to implement and potentially useful. In
looking at how we are using Xindice, we would have immediate benefit.
Could some explain how the proposed metadata implementation is different
and/or better?

Matt Liotta
President & CEO
Montara Software, Inc.
http://www.montarasoftware.com/
888-408-0900 x901

> -----Original Message-----
> From: Vladimir R. Bossicard [mailto:[EMAIL PROTECTED]
> Sent: Tuesday, December 24, 2002 2:21 AM
> To: [EMAIL PROTECTED]
> Subject: Re: proposal for metadata
> 
> Sorry to bother,
> 
> XNode in scratchpad: +1
> Metadata: -0.  Isn't is possible to create this a service for a
> collection?
> 
> - where are the tests?  Can't this code fail?
> - where is the documentation?
> - can we see the files from the meta package?
> - the '{' are placed at the end of the line
> 
> > This will allow developers to see if the concept of metadata is
useful
> in
> > Xindice.
> 
> and if 99% think that this is not useful, we will have to keep (and
> maintain) the code for 1% of the users...
> 
> But if you want to commit right away, go ahead.  I don't want to be
the
> only one that stops the code from floading into the codebase.
> 
> -Vladimir
> 
> --
> Vladimir R. Bossicard
> Apache Xindice - http://xml.apache.org/xindice

Reply via email to