IRC meeting minutes - 03/21

[Svante Schubert <Svante.Schubert@Sun.COM>]

Session Start: Wed Mar 21 2007
Session Ident: #odf-metadata
[Wed:15:52:14] * Attempting to rejoin channel #odf-metadata
..
[Wed:16:01:09] * danbri can only lurk, in another meeting, but good to 
see irc being used!
[Wed:16:01:27] * Svante sets mode: +o danbri
[Wed:16:01:30] <danbri> also sorry for general non-participation on the 
oasis group... i do try follow some by email tho
[Wed:16:01:56] <darcusb> hey Dan!
[Wed:16:02:07] <darcusb> big topic for today is URIs
[Wed:16:02:07] <Svante> Very good, Dan! Nice to have you around
[Wed:16:02:27] * EliasT has joined #odf-metadata
[Wed:16:02:39] * EliasT waves at everybody!
[Wed:16:02:51] <darcusb> hey elias!
[Wed:16:03:14] * Svante sets mode: +o EliasT
[Wed:16:03:28] <EliasT> is danbri actually herE?
[Wed:16:04:14] * danbri is in another meeting, but lurking; hi!
[Wed:16:04:19] <darcusb> yes, he said "can only lurk, in another meeting"
[Wed:16:04:31] <EliasT> danbri, want to start an ODF Fight Club?
[Wed:16:04:40] <Svante> ;-)
[Wed:16:05:10] * flr has joined #odf-metadata
[Wed:16:05:11] <EliasT> Svante, don't read too much into it... danbri 
was posting some funny links to #swig yesterday.
[Wed:16:05:12] <flr> Hi
[Wed:16:05:20] * EliasT waves at flr
[Wed:16:05:23] * Svante sets mode: +o flr
[Wed:16:05:40] * Svante gives power to the people
[Wed:16:08:59] <flr> Atendees: Bruce, Florian, Svante, Dan, Bernd, 
Michael, Gary
[Wed:16:09:13] <Svante> and Elias
[Wed:16:09:37] <flr> First topic: binding identification
[Wed:16:10:27] <Svante> Agenda:
[Wed:16:10:27] <Svante> 1. Rollcall - 5 Minutes
[Wed:16:10:27] <Svante> 2. Binding/Identification proposals
[Wed:16:10:27] <Svante> 3. Open issues on manifest ontology
[Wed:16:10:27] <Svante> 4. Examples for the proposal
[Wed:16:10:27] <Svante> 5. Deliverable deadline
[Wed:16:10:37] <flr> Discussion about 3.
[Wed:16:11:41] <flr> open issues: URLs
[Wed:16:12:46] <darcusb> Integrating Elias' W3C best practice approach 
with the dates included would instead yield something like this I guess:
[Wed:16:12:46] <darcusb>     manifest --> 
<http://docs.oasis-open.org/odf/2007/03/meta/manifest>
[Wed:16:12:46] <darcusb>     meta --> 
<http://docs.oasis-open.org/odf/2007/03/meta>
[Wed:16:13:35] <flr> #
[Wed:16:13:55] <EliasT> 
http://docs.oasis-open.org/odf/2007/03/meta/manifest#
[Wed:16:14:17] <Svante> Currently in OWL we use 
http://opendocument.xml.org/2007/03/meta-manifest/ns/
[Wed:16:14:56] <EliasT> 
<http://docs.oasis-open.org/office/2007/03/meta/manifest#>
[Wed:16:14:58] <flr> 
http://docs.oasis-open.org/office/2007/03/meta/manifest#
[Wed:16:15:28] <flr> open issue Category
[Wed:16:15:40] <EliasT> <manifest:category>
[Wed:16:15:47] * ge has joined #ODF-METADATA
[Wed:16:16:33] <flr> change manifest:category in the manifest to ...
[Wed:16:16:44] <EliasT> a plugin will add extra triples to the manifest 
entry that they own
[Wed:16:16:51] <EliasT> so they can query for those triples when they 
want to find it.
[Wed:16:17:02] <EliasT> no need to force a manifest:category
[Wed:16:17:34] <flr> leave it up to the plug-in to define their category
[Wed:16:17:46] <EliasT> <_:entry1> a :Entry ; rdf:type 
<http://torrez.us/myplugin>
[Wed:16:17:49] <EliasT> <_:entry1> a :Entry ; rdf:type 
<http://torrez.us/myplugin> .
[Wed:16:18:55] <flr> kepp the issues of category identification seperate
[Wed:16:19:15] <flr> open issue Named Graph
[Wed:16:19:59] <EliasT> ACTION: to add text to the spec documenting that 
the named graph URI will be the about of an :Entry in the manifest.
[Wed:16:20:25] <EliasT> ACTION: Svante to add text to the spec 
documenting that the named graph URI will be the about of an :Entry in 
the manifest.
[Wed:16:22:09] <flr> Discussion about 4.
[Wed:16:22:33] <flr> Discussion about  4. Examples for the proposal
[Wed:16:22:46] <EliasT> +1
[Wed:16:23:05] * Svante sets mode: +o ge
[Wed:16:23:30] <flr> so we need to assemble samples ...
[Wed:16:23:34] <flr> Discussion about 5. Deliverable deadline
[Wed:16:24:54] <flr> Deadline for ODF1.2 comitte draft:
[Wed:16:25:24] <flr> May 2007
[Wed:16:27:16] <flr> Target date: early april?
[Wed:16:28:03] <flr> Best target date: end of march... / before 4th of april
[Wed:16:31:16] <flr> Deadline given by the metadata SC itself: 4th of 
april 2007
[Wed:16:32:22] <flr> Discussion about  2. Binding/Identification proposals
[Wed:16:33:33] <EliasT> document-id
[Wed:16:33:33] <EliasT> =========
[Wed:16:33:33] <EliasT> We therefore agreed on the existence of an ID, 
but have different 
[Wed:16:33:33] <EliasT> proposals about the ID:
[Wed:16:33:33] <EliasT> 
[Wed:16:33:34] <EliasT> If someone external would like to refer to the 
document, he might use a 
[Wed:16:33:36] <EliasT> string similar to this
[Wed:16:33:38] <EliasT>     "tag:odf:550e8400-e29b-41d4-a716-446655440000"
[Wed:16:33:40] <EliasT> It is unclear, if this ID is the value of an 
attribute of the meta 
[Wed:16:33:40] * EliasT has quit IRC (Excess Flood)
[Wed:16:33:43] <flr> No more work on revision ids
[Wed:16:34:04] * EliasT has joined #odf-metadata
[Wed:16:34:51] <Svante> 
tag:odf:550e8400-e29b-41d4-a716-446655440000:/content.xml#myID
[Wed:16:34:51] <Svante> or
[Wed:16:34:51] <Svante> odf:/content.xml#myID
[Wed:16:35:36] <flr> difference between svantes and bruce proposal is 
the "document-id"
[Wed:16:36:23] <flr> EliasT: standard RDF will not parse 
"odf:/content.xml#myID"
[Wed:16:37:13] <Svante> Even our RDFa have to be parsed
[Wed:16:37:18] <EliasT> Svante: either way you still have to parse URI 
element id to find the element in content.xml
[Wed:16:37:47] <EliasT> but RDFa attributes contain a full URI
[Wed:16:38:03] <EliasT> parsing triples from a graph is different than 
parsing a specific URI
[Wed:16:38:17] <EliasT> meta:about="absolute-URI"
[Wed:16:38:54] <EliasT> meta:about="foo"
[Wed:16:39:00] <flr> issue is: how to make relativa uris absolute
[Wed:16:39:00] <EliasT> content.xml#foo
[Wed:16:39:58] <EliasT> meta:about="foo" -> content.xml#foo
[Wed:16:40:33] <flr> we have to specifcy how the realtive URI 
meta:about="foo" will be transformed into the absolute URI content.xml#foo
[Wed:16:40:57] <EliasT> http://
[Wed:16:40:59] <EliasT> file:///
[Wed:16:42:00] <EliasT> urn:tag:torez.us:foo
[Wed:16:42:08] <EliasT> tag:torrez
[Wed:16:42:12] <EliasT> odf:/content.xml
[Wed:16:43:01] <flr> argument against odf:/content.xml is that odf is 
not registered
[Wed:16:43:18] <EliasT> meta:about="urn:uuid:123"
[Wed:16:43:32] <EliasT> urn:uuid:123 a odf:Element ;
[Wed:16:43:41] <EliasT> urn:uuid:123 odf:name "foo" .
[Wed:16:45:10] <EliasT> urn:tag:doc-UUID:foo
[Wed:16:50:30] <flr> bruce: problem is the xml:id resolution
[Wed:16:51:32] <EliasT> urn:uuid:123 odf:name "content.xml#foo" .
[Wed:16:53:47] <EliasT> 1. xml:id
[Wed:16:53:54] <flr> svante: shall we define properies liek odf:name
[Wed:16:54:02] <EliasT> 2. meta:about to always an absolute URI
[Wed:16:55:44] <flr> eliast: "encourage" programs to generate unique 
URIs across documents
[Wed:16:58:39] <Svante> I assume we all agree that a doc ID can be 
helpful and should always be inserted to the document, although 
currently they are hardly unique due to bad user habits
[Wed:17:00:53] <EliasT> odf:name "content.xml#foo"
[Wed:17:08:45] <EliasT> ADJOURNED
[Wed:17:09:06] * darcusb has left #odf-metadata
[Wed:17:09:18] * ge has quit IRC