----Security_Multipart(Mon_Sep_18_03_26_39_2006_750)--
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Klaus Heinz <k.heinz.sep.sechs@kh-22.de> wrote
  in <20060917162613.GA1442@silence.homedns.org>:

k.> obata@lins.jp wrote:
k.>
k.> > 	In developer.xml <URL:http://www.netbsd.org/People/developers.xml>
k.> > 		http://www.NetBSD.org/XML/htdocs/lang/share/xml/website-netbsd.dtd
k.> > 	Real URL:
k.> > 		http://www.NetBSD.org/share/xml/website-netbsd.dtd
k.> >
k.> > 	So, I can't process those XML files.
k.>
k.> This was done by design. The public identifier in our documents is _not_
k.> meant to be used for downloading a DTD through a network.
k.> In pkgsrc, various catalog files, managed by xmlcatmgr, are responsible
k.> for maintaining a mapping from this identifier to the actual local DTD
k.> file.
k.>
k.> I do not remember the exact reason hrs@ introduced this separate XML
k.> URL namespace.

 Yes.  This is because the URLs are used as an abstraction layer
 to access a file from XML documents.  For example,
 "http://www.NetBSD.org/XML/htdocs/lang/share/xml/website-netbsd.dtd"
 is for a localized DTD since it includes /lang/ part.  The /lang/ part
 is resolved by XML catalog file.  This is needed for preventing the
 documents from being bound by the actual directory structure.
 Without it, relocation of the document files becomes very difficult.

 In short, tools which does not support XML catalog cannot process
 XML files in the htdocs tree.  XML support of WWW browser is very
 limited, so you cannot get meaningful results by using it even if
 the URLs can actually be accessed by HTTP.

--
| Hiroki SATO

----Security_Multipart(Mon_Sep_18_03_26_39_2006_750)--
Content-Type: application/pgp-signature
Content-Transfer-Encoding: 7bit

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (FreeBSD)

iD8DBQBFDZNfTyzT2CeTzy0RAhJJAJ0e1/d3HzNDdz+pTzoIqMR9AvGffACgj2f5
E62VTbddYCFIWb3f3PjkMxU=
=cmDN
-----END PGP SIGNATURE-----

----Security_Multipart(Mon_Sep_18_03_26_39_2006_750)----