Re: New draft

[Norman Walsh <ndw@nwalsh.com>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

/ Paul Grosso <pgrosso@arbortext.com> was heard to say:
|>This draft addresses the issues discussed today.
|
| Regarding (section 5.3. Catalog Circularities):
|
|   "The URIs identified by catalog entries are not subject to catalog resolution."
|
| I don't know what "URIs identified by catalog entries" means.
| I'm not sure what "identified by catalog entries" means.
|
| Do we simply mean to say that none of the attributes in the XML Catalog
| namespace are to be "of type URI reference" in the sense that they shouldn't
| be "resolved" by the catalog process?  That is, they are all strings
| that might look like URI references and that, after getting returned to the
| calling program, are expected to be treated as URI references, but they
| are URI references as far as the catalog is concerned so they shouldn't
| be "resolved" by the catalog process?

No, they are URI references (and they're defined that way explicitly
so that xml:base will do the right thing), but they aren't subjected
to catalog resolution.

When the resolver processes <nextCatalog uri="foo"/>, it does not
attempt to find "foo" in a <uri .../> entry before retrieving it.

But I guess I'm not saying that clearly. I'm too tired right now to
try again, but I will.

                                        Be seeing you,
                                          norm

- -- 
Norman.Walsh@Sun.COM    | Faith makes many of the mountains which it
XML Standards Architect | has to remove.--W. R. Inge
Web Tech. and Standards |
Sun Microsystems, Inc.  | 
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.7 <http://mailcrypt.sourceforge.net/>

iD8DBQE+qE6VOyltUcwYWjsRAucTAJ42hxYCF493cQ0jbWpkYUitAg/n7QCeK7Bx
rgmEuq2VZA8TGD7iy4CXC5g=
=p+jE
-----END PGP SIGNATURE-----