[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: IRC meeting minutes - 04/04
Here are the rough IRC notes we made during the call. Svante
Session Start: Wed Apr 04 00:00:01 2007
Session Ident: #odf-metadata
[Wed:17:11:13] <EliasT_> mm:datatype
[Wed:17:11:28] <Svante> office:datatype
[Wed:17:11:40] <Svante> ^^ current name of the attribute
[Wed:17:12:13] <flr> For the notes: The topic is "Do we require 'a hint' - similar to rdf:type in the manifest - at
[Wed:17:12:13] <flr> the text:meta-field to help to establish a relation between
[Wed:17:12:13] <flr> text:meta-field (the office) and a plug-in.
[Wed:17:12:17] <flr> Right?
[Wed:17:13:15] <Svante> Usage of office:value / office:value-type. This additional type
[Wed:17:13:15] <Svante> information is added by using existing office attributes. We
[Wed:17:13:15] <Svante> should consider to map the possible attribute values to RDF IRIs.
[Wed:17:13:17] <EliasT_> having URIs to define types as opposed to enum('string', 'float', 'int')?
[Wed:17:13:44] <Svante> This attribute value would be mapped to URI of the existing RDF Types
[Wed:17:14:10] <flr> <text:span office:value="dffff">1234</text:span>
[Wed:17:14:22] <flr> <text:span office:value="dffff" ....the other missing attr...>1234</text:span>
[Wed:17:14:38] <Svante> Defined set for the attributes values like "dffff" is already in RelaxNG
[Wed:17:15:38] <EliasT_> m:value, m:datatype
[Wed:17:16:06] <EliasT_> m:content, m:datatype
[Wed:17:16:09] <Svante> I thought we use office:value as we have it already
[Wed:17:16:21] <Svante> otherwise we might use rdf:type
[Wed:17:17:30] <EliasT_> http://www.w3.org/TR/rdf-concepts/#section-Datatypes-intro
[Wed:17:17:31] <Svante> Question: Where ist this allowed?
[Wed:17:17:57] <EliasT_> http://www.w3.org/TR/rdf-syntax-grammar/#section-Syntax-datatyped-literals
[Wed:17:18:34] * ge has joined #odf-metadata
[Wed:17:19:12] <EliasT_> m:about
[Wed:17:19:13] <EliasT_> rdf:about
[Wed:17:19:25] <EliasT_> m:datatype
[Wed:17:19:30] <EliasT_> m:content
[Wed:17:20:36] <Svante> m:datatype="http://www.w3.org/2001/XMLSchema#int"; m:content="7"
[Wed:17:20:47] * EliasT_ has quit IRC (Read error: 104 (Connection reset by peer))
[Wed:17:23:46] <darcusb_> <http:ex.net/1> dc:date 2002-10-12 .
[Wed:17:24:11] <darcusb_> <http:ex.net/1> rdf:value "October 12, 2002"@en .
[Wed:17:26:08] * EliasT_ has joined #odf-metadata
[Wed:17:27:13] <darcusb_> <text:p m:about="http:ex.net/1" m:property="dc:date" m:datatype="xsd:date" m:content="2002-10-12">October 12, 2002</text:p>
[Wed:17:27:26] <Svante> Current list:
[Wed:17:27:26] <Svante> images (draw:image)
[Wed:17:27:26] <Svante> paragraphs (text:p)
[Wed:17:27:26] <Svante> headings (text:h)
[Wed:17:27:26] <Svante> tables (table:table)
[Wed:17:27:26] <Svante> table-cells (table:table-cell)
[Wed:17:27:26] <Svante> bookmarks (text:bookmark, text:bookmark-start, text:bookmark-end)
[Wed:17:27:26] <Svante> metadata fields (text:meta)
[Wed:17:27:48] <Svante> These are the elements were m:about, m:property and xml:id is possible as attributes
[Wed:17:29:29] <Svante> m:about and m:property are describing a RDF literal, but draw:image has no literal!
[Wed:17:29:46] <EliasT_> draw:href="" is the resource object.
[Wed:17:30:06] <Svante> Earlier we defined the literal as the value of the element
[Wed:17:31:25] <Svante> Only xml:id should be allowed to add on draw:image to bind RDF statements from RDF/XML files to the image
[Wed:17:31:29] <EliasT_> <draw:image m:about="http://..."; m:property="foaf:img" draw:href="http://..."; />
[Wed:17:31:36] <Svante> Only xml:id should be allowed to add on draw:image to bind RDF statements from RDF/XML files to the image?
[Wed:17:32:34] <EliasT_> RDFa: link, a, img
[Wed:17:32:34] <flr> Replace draw:image with draw:frame ;-)
[Wed:17:32:58] <flr> 'Cause a draw:image is inside a draw:frame...
[Wed:17:33:26] <EliasT_> Svante: an odf:id
[Wed:17:36:35] <Svante> Flr dislikes content.xml#id
[Wed:17:37:00] <Svante> odf:path="content.xml"
[Wed:17:37:09] <Svante> odf:id="id"
[Wed:17:37:21] <Svante> this would be written metamanifest or some binding.rdf
[Wed:17:37:57] <flr> The uniqueness constraint comes from Svante (more precice from the Sun Writer team) and from me ;-)
[Wed:17:38:00] <Svante> odf:id is in RDF
[Wed:17:38:06] <Svante> xml:id is content.xml
[Wed:17:39:04] <Svante> <mm:manifest odf:baseId="urn:uuid:224ab023-77b8-4396-a75a-8cecd85b81e3">
[Wed:17:39:04] <Svante> <odf:Element rdf:about="IRI1" odf:path="style.xml" odf:id="id1"/>
[Wed:17:39:04] <Svante> <odf:Element rdf:about="IRI2" odf:path="content.xml" odf:id="id1"/>
[Wed:17:39:04] <Svante> </mm:manifest>
[Wed:17:41:12] <flr> ids are not invariant
[Wed:17:42:11] <flr> odf:path="."
[Wed:17:45:48] <Svante> <mm:manifest odf:baseId="urn:uuid:224ab023-77b8-4396-a75a-8cecd85b81e3">
[Wed:17:45:48] <Svante> <mm:Entry rdf:about="http://SOME-NAMED-GRAPH"; odf:path="meta/vcard.rdf">
[Wed:17:45:48] <Svante> <rdf:type rdf:resource="http://www.w3.org/2006/vcard/ns#"/>
[Wed:17:45:48] <Svante> <rdf:type rdf:resource="http://myAdressbook#"/>
[Wed:17:45:48] <Svante> </mm:Entry>
[Wed:17:45:48] <Svante> <odf:Element rdf:about="IRI1" odf:path="style.xml" odf:id="id1"/>
[Wed:17:45:48] <Svante> <odf:Element rdf:about="IRI2" odf:path="content.xml" odf:id="id1"/>
[Wed:17:45:48] <Svante> <odf:Element rdf:about="IRI3" odf:path="content.xml" odf:id="id1"/>
[Wed:17:45:48] <Svante> </mm:manifest>
[Wed:17:45:57] <Svante> Shall we create two files from it?
[Wed:17:46:06] <Svante> binding.rdf and manifest.rdf
[Wed:17:47:27] <EliasT_> I think we only need one file.
[Wed:17:47:53] <Svante> Do we need RDFa on images?
[Wed:17:48:26] <Svante> We have one algorithm to resolve the literal...
[Wed:17:48:40] <Svante> get the literal from the element value..
[Wed:17:49:07] <EliasT_> <meta property="" href="">
[Wed:17:49:14] <EliasT_> <link property="" href=""/>
[Wed:17:49:57] <Svante> <link xml:id href=""/>
[Wed:17:50:16] <Svante> Make a statement in RDF/XML about the link element
[Wed:17:51:59] <EliasT_> Right now we can only build in-content triples that the object is a literal.
[Wed:17:52:05] <EliasT_> plain or datatyped.
[Wed:17:52:15] <EliasT_> do we need an element that can generate a triple with an resource object.
[Wed:17:52:39] <EliasT_> <_:a> foaf:knows <http://elias/>
[Wed:17:55:55] <Svante> * Do we require 'a hint' - similar to rdf:type in the manifest - at
[Wed:17:55:55] <Svante> the text:meta-field to help to establish a relation between
[Wed:17:55:55] <Svante> text:meta-field (the office) and a plug-in. Our last stand was
[Wed:17:55:55] <Svante> adding meta:category (now m:category). More consequent would be
[Wed:17:55:55] <Svante> the usage of rdf:type similar to the metamanifest. The difference:
[Wed:17:55:55] <Svante> in the metamanifest we are able to have multiple rdf:types
[Wed:17:55:55] <Svante> elements on one entry, but we can have only one rdf:type attribute
[Wed:17:55:55] <Svante> on the element.
[Wed:17:55:55] <Svante> But we might even neglect the type on the field, by a
[Wed:17:55:55] <Svante> office/plug-in scenario as the following:
[Wed:17:55:55] <Svante> 1. A plug-in would register for certain rdf:types.
[Wed:17:55:55] <Svante> 2. The office would read the metadata manifest and find a
[Wed:17:55:55] <Svante> responsible plug-in based upon the registered rdf:types for
[Wed:17:55:55] <Svante> certain RDF/XML entries.
[Wed:17:55:55] <Svante> 3. The plug-in would read it's entries (RDF/XML files) and
[Wed:17:55:55] <Svante> would find IRIs in it it related to odf:elements.
[Wed:17:55:55] <Svante> 4. Plug-in would announce responsibility for these IRIs
[Wed:17:55:55] <Svante> 5. Based on the binding the office would be able to map the
[Wed:17:55:55] <Svante> IRIs to xml:ids.
[Wed:17:55:55] * Disconnected
[Wed:17:56:09] * Attempting to rejoin channel #odf-metadata
[Wed:17:56:15] * Rejoined channel #odf-metadata
[Wed:17:56:17] * Svante is now known as Svante_afk
[Wed:17:56:31] * Svante_afk is now known as Svante
[Wed:17:59:25] <EliasT_> <text:field>
[Wed:18:01:10] <Svante> Do you agree that the editor remove m:about/m:property/m:content/m:datatype from the draw:image and table:table?
[Wed:18:01:17] <flr> forget about start and end tags. Its no soo important and the end ;-)
[Wed:18:01:24] <EliasT_> right
[Wed:18:01:26] <flr> But a generic field is important.
[Wed:18:01:32] <EliasT_> we could add two meta field elements
[Wed:18:01:38] <EliasT_> one that has a property ="start"
[Wed:18:01:40] <EliasT_> and the other "end"
[Wed:18:01:43] <EliasT_> so no biggie.
[Wed:18:01:57] <Svante> not for XML, but for the application
[Wed:18:02:02] <darcusb_> meta attributes are only for literals, and so not relevant to images
[Wed:18:02:30] <Svante> and tables?
[Wed:18:02:50] <EliasT_> Svante, any element that has user-generated content...
[Wed:18:03:18] <EliasT_> m:about/m:property/m:content/m:datatype from the draw:image and table:table?
[Wed:18:03:42] <flr> this means everything which has a style should habe m:about/m:propoerty, etc
[Wed:18:03:46] <EliasT_> <element>CONTENT</element>
[Wed:18:04:09] <EliasT_> <table>
[Wed:18:04:21] <EliasT_> datatype="rdf:XMLLiteral"
[Wed:18:05:03] <EliasT_> <sup>2</sup>
[Wed:18:05:21] <EliasT_> <text:p>E = mc<sup>2</sup></text:p>
[Wed:18:05:35] <EliasT_> <text:p m:about=="" m:property="dc:title">E = mc<sup>2</sup></text:p>
[Wed:18:05:53] <EliasT_> e = mc2
[Wed:18:06:01] <Svante> What does m:about="" mean?
[Wed:18:06:12] <EliasT_> <> dc:title "E = mc<sup>2</sup>"^^rdf:XMLLiteral
[Wed:18:06:55] <flr> <sup> === <div + style>
[Wed:18:07:22] <flr> <div + style + RDF semantic>
[Wed:18:08:13] <Svante> We would raise the bar quite a lot as the Office needs to create the XML for certain elements on run time, not just during serialization for the whole document!
[Wed:18:08:39] * flr need to go. cu
[Wed:18:15:18] * ge has quit IRC
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]