Chatlog 2010-08-05

See CommonScribe Control Panel, original RRSAgent log and preview nicely formatted version.

13:50:13 <RRSAgent> RRSAgent has joined #rdfa
13:50:13 <RRSAgent> logging to http://www.w3.org/2010/08/05-rdfa-irc
13:50:15 <trackbot> RRSAgent, make logs world
13:50:15 <Zakim> Zakim has joined #rdfa
13:50:17 <trackbot> Zakim, this will be 7332
13:50:17 <Zakim> ok, trackbot; I see SW_RDFa()10:00AM scheduled to start in 10 minutes
13:50:18 <trackbot> Meeting: RDFa Working Group Teleconference
13:50:18 <trackbot> Date: 05 August 2010
13:50:25 <manu> Agenda: http://lists.w3.org/Archives/Public/public-rdfa-wg/2010Aug/0006.html
13:50:29 <manu> Chair: Manu
13:50:55 <manu> Present: Knud, Manu, Shane, MarkB, Toby
13:53:02 <manu> Regrets: Ivan
13:58:04 <ShaneM> ShaneM has joined #rdfa
13:58:18 <Zakim> SW_RDFa()10:00AM has now started
13:58:25 <Zakim> +Knud
13:58:33 <markbirbeck> markbirbeck has joined #rdfa
13:58:43 <toby> toby has joined #rdfa
13:58:55 <Zakim> +manu
13:59:01 <Zakim> + +44.785.583.aaaa
13:59:09 <Knud> Scribe: Knud
13:59:12 <toby> zakim, aaaa is me
13:59:12 <Zakim> +toby; got it
13:59:24 <toby> zakim,mute me
13:59:24 <Zakim> toby should now be muted
14:01:43 <manu> zakim, who is on the call?
14:01:43 <Zakim> On the phone I see Knud, manu, toby (muted)
14:01:53 <markbirbeck> zakim, code?
14:01:53 <Zakim> the conference code is 7332 (tel:+1.617.761.6200 tel:+33.4.26.46.79.03 tel:+44.203.318.0479), markbirbeck
14:02:46 <toby> zakim, unmute me
14:02:46 <Zakim> toby should no longer be muted
14:03:43 <Zakim> +??P10
14:03:46 <markbirbeck> zakim, i am ?
14:03:46 <Zakim> +markbirbeck; got it
14:04:15 <Zakim> +ShaneM
14:05:03 <toby> zakim, who's here?
14:05:03 <Zakim> On the phone I see Knud, manu, toby, markbirbeck, ShaneM
14:05:05 <Zakim> On IRC I see toby, markbirbeck, ShaneM, Zakim, RRSAgent, trackbot, manu, Knud
14:05:15 <Zakim> + +8014aabb
14:05:21 <manu> zakim, I am ??bb
14:05:21 <Zakim> sorry, manu, I do not see a party named '??bb'
14:05:24 <manu> zakim, I am bb
14:05:24 <Zakim> sorry, manu, I do not see a party named 'bb'
14:05:32 <manu> zakim, aabb is me
14:05:32 <Zakim> +manu; got it
14:06:01 <toby> zakim, mute me
14:06:01 <Zakim> toby should now be muted
14:06:09 <manu> zakim, who is making noise?
14:06:14 <toby> zakim, unmute me
14:06:14 <Zakim> toby should no longer be muted
14:06:19 <markbirbeck> zakim, mute me
14:06:19 <Zakim> markbirbeck should now be muted
14:06:20 <Zakim> manu, listening for 10 seconds I heard sound from the following: manu.a (100%), Knud (57%)
14:06:28 <Knud> Manu: Any additions or changes to the agenda?
14:06:28 <Zakim> Knud should now be muted
14:07:11 <Knud> no other additions to the agenda
14:07:13 <markbirbeck> zakim, unmute me
14:07:13 <Zakim> markbirbeck should no longer be muted
14:07:24 <Knud> TOPIC: ISSUE-31: setMapping API location
14:07:47 <Knud> Manu: setMapping call should either be duplicated or be moved from context to document interface
14:08:00 <manu> document.setMapping("foaf", "http://xmlns.com/foaf/0.1");
14:08:07 <manu> document.data.context.setMapping("foaf", "http://xmlns.com/foaf/0.1");
14:08:51 <Knud> manu: document.setMapping could be an alias to context.setMapping
14:09:28 <markbirbeck> http://lists.w3.org/Archives/Public/public-rdfa-wg/2010Jun/0054.html
14:10:09 <Knud> mark: still open as to how to implement this
14:10:24 <markbirbeck> document.setMapping("foaf", "http://xmlns.com/foaf/0.1/");
14:10:24 <markbirbeck>   var people = document.getItemsByType("foaf:Person");
14:10:24 <Knud> ... as long as programmers can get to setMapping via document
14:11:43 <Knud> manu: would like to keep the idea of context
14:12:07 <Knud> ... also, some people might be concerned that we pollute the document interface too much
14:12:20 <markbirbeck> q+
14:12:35 <manu> ack markbirbeck
14:12:47 <toby> Is {document.setMapping("foaf","http://xmlns.com/foaf/0.1/");var people=document.getElementsByType("foaf:Person");} any easier than {var FOAF="http://xmlns.com/foaf/0.1/";var people=document.getElementsByType(FOAF+"Person");} ??
14:12:51 <Zakim> -manu.a
14:13:12 <Zakim> +manu.a
14:13:16 <manu> ack markbirbeck
14:13:49 <markbirbeck> q+
14:14:03 <manu> ack markbirbeck
14:14:58 <Knud> mark: now remembers that we steered clear of putting too much into document interface
14:15:19 <manu> var context = document.data.context;
14:15:34 <manu> context.setMapping("foaf", "http://xmlns.com/foaf/0.1")?
14:15:37 <Knud> ... so maybe we should rethink how we give people access to something like setMapping
14:15:40 <manu> so, people would just do that.
14:16:44 <Knud> ... there could also be a context parameter to getElementsByType
14:17:28 <Knud> ... would therefore like to withdraw the suggestion to move setMapping to document
14:18:03 <ShaneM> no objection from me
14:18:08 <Knud> fine
14:18:22 <manu> PROPOSAL: Close ISSUE-31 and have Mark re-submit an alternative mechanism to make mappings easier to use.
14:18:33 <manu> +1
14:18:34 <Knud> Knud: +1
14:18:35 <markbirbeck> +1
14:18:37 <ShaneM> +1
14:18:43 <toby> +1
14:18:47 <manu> RESOLVED: Close ISSUE-31 and have Mark re-submit an alternative mechanism to make mappings easier to use.
14:19:02 <Knud> TOPIC: ISSUE-30: hasFeature clarifications
14:19:19 <manu> http://lists.w3.org/Archives/Public/public-rdfa-wg/2010Jun/0053.html
14:19:40 <Knud> Mark: hasFeature already exists in the DOM in other specs
14:20:00 <manu> q+ to discuss RDFa version number
14:20:00 <Knud> ... our use of it should be aligned, e.g., with respect to the version number
14:20:03 <manu> ack manu
14:20:03 <Zakim> manu, you wanted to discuss RDFa version number
14:20:37 <Knud> manu: mostly agrees with proposal, except: we now want to make RDFa versionless
14:20:56 <Knud> ... so, just check if RDFa support is available, not a specific version
14:21:31 <Knud> mark: maybe we should change feature string to "rdfa_api"
14:22:13 <manu> document.hasFeature("rdfa-api 1.1")
14:22:37 <markbirbeck> document.hasFeature("rdfaapi", "1.1")
14:23:03 <manu> document.hasFeature("rdfa-api 1.0")
14:23:20 <manu> oh woops, Mark is correct
14:23:47 <Knud> mark: api version is in sync with RDFa core?
14:23:54 <manu> document.hasFeature("rdfa-api, "1.1")
14:24:56 <manu> So, something like this: document.implementation.hasFeature("rdfa-api, "1.1")
14:25:31 <Knud> mark: we should not say "we provide hasFeature", but instead "when you call hasFeature, this should happen:"
14:25:43 <Knud> ... this is in line with how other specs do it
14:26:30 <toby> <script implements="http://www.w3.org/TR/rdfa-api/">...</script>
14:26:36 <markbirbeck> This is boilerplate text:
14:26:38 <markbirbeck> http://www.w3.org/TR/DOM-Level-2-Core/core.html#ID-BBACDC08
14:27:42 <ShaneM> toby: yeah, that.  I really wanted that feature.
14:29:17 <manu> PROPOSAL: Bring hasFeature() in line with the DOMImplementation interface, use "rdfa-api" as the feature string, and "1.1" as the version.
14:29:25 <toby> +1
14:29:28 <ShaneM> +1
14:29:35 <Knud> Knud: +1
14:29:40 <markbirbeck> http://www.w3.org/TR/DOM-Level-2-Core/introduction.html#ID-Conformance
14:31:10 <markbirbeck> http://www.w3.org/TR/SVG/feature.html
14:31:12 <manu> RDFaApi
14:33:32 <toby> "rdfa-api" could be considered a relative URL reference with base http://www.w3.org/TR/.
14:34:09 <markbirbeck> +1
14:34:24 <Knud> general discussion on what to use as the feature string: URI, camel case, hyphen, ...
14:34:45 <ShaneM> consistency is the hobgoblin of little minds
14:35:49 <manu> PROPOSAL: Bring hasFeature() in line with the DOMImplementation interface, use "RDFaAPI" as the feature string, and "1.1" as the version.
14:36:13 <ShaneM> +1
14:36:13 <manu> +1
14:36:14 <Knud> Knud: +1
14:36:17 <toby> +1 (though I'd still prefer 'rdfa-api')
14:37:05 <manu> http://www.w3.org/TR/DOM-Level-2-Core/introduction.html#ID-Conformance
14:37:36 <markbirbeck> +1
14:37:44 <manu> RESOLVED: Bring hasFeature() in line with the DOMImplementation interface, use "RDFaAPI" as the feature string, and "1.1" as the version.
14:38:09 <Knud> TOPIC: ISSUE-32: TypedLiteralConverter improvements
14:38:15 <manu> http://lists.w3.org/Archives/Public/public-rdfa-wg/2010Jun/0055.html
14:38:18 <markbirbeck> http://dev.w3.org/2006/webapi/DOM-Level-3-Events/html/DOM3-Events.html#feature-detection
14:38:45 <markbirbeck> RDFa.API
14:40:32 <manu> q+ to point out that we allow functions already.
14:41:12 <Knud> Mark: we should make very clear how callback functions/methods should be implemented
14:41:39 <manu> We already allow callback functions for TypedLiteralConverters: http://www.w3.org/2010/02/rdfa/sources/rdfa-dom-api/#idl-def-TypedLiteralConverter
14:42:44 <manu> ack manu
14:42:44 <Zakim> manu, you wanted to point out that we allow functions already.
14:42:58 <Knud> ... also, people might only want to define one callback function for different datatypes, so the callback function should have a parameter to say what it should convert
14:43:25 <Knud> ... finally, we should define how the callback reacts when conversion is not possible (exception, return null, etc.)
14:43:36 <toby> first class functions
14:45:25 <Knud> ... for the third point, the parameter should say what the function is converting _to_
14:45:32 <manu> q+ to discuss point 3
14:45:34 <manu> ack manu
14:45:34 <Zakim> manu, you wanted to discuss point 3
14:47:10 <markbirbeck> var converter = function (value) { return new String(value) };
14:47:10 <markbirbeck> document.data.context.registerTypeConverter("xsd:string", converter);
14:47:51 <markbirbeck> document.data.context.registerTypeConverter("xsd:string",
14:47:55 <markbirbeck>   function (value) { return new String(value) }
14:47:55 <Knud> mark: if there is one function per conversion type, then we wouldn't need the extra parameter.
14:47:59 <markbirbeck> );
14:48:23 <Knud> ... when we want to allow one function for all conversion types, then we need an extra parameter
14:48:46 <toby> http://esw.pastebin.com/m9CXB8qh
14:50:37 <manu> PROPOSAL: Add a parameter to convertType() that indicates the type that is being converted.
14:50:44 <Knud> Knud: shouldn't it be 'converted to'?
14:51:00 <toby> +0
14:51:03 <markbirbeck> +1
14:51:14 <Knud> Knud: +1
14:51:27 <ShaneM> abstain - I don't grok the issue
14:51:36 <manu> +1
14:52:19 <manu> RESOLVED: Add a parameter to convertType() that indicates the type that is being converted.
14:52:44 <Knud> we are resolving this issue with a string of different proposals
14:53:33 <markbirbeck> "23 elephants"^^xsd:integer
14:53:41 <Knud> SubTopic: Conversion guidance when authoring TypedLiteralConverters
14:54:49 <Knud> Mark: We need to provide better guidance on what do to when authoring TypedLiteralConverters. For example, "23 elephants"^^xsd:integer fails the range check for xsd:integer, but applying a Javascript integer converter would return the number 23 even though this is an RDF range violation. We should suggest that authors be aware of this sort of issue when writing range checks.
14:54:00 <toby> q+
14:54:49 <manu> ack toby
14:54:52 <manu> Toby: We also want to let authors know that RDF processors may not even allow "23 elephants" to be expressed or they may modify it to "23" before sending it up to the RDFa API. RDF processors are allowed to do this kind of modification of Typed Literals based on range checking. Authors shouldn't depend on ever getting "23 elephants"^^xsd:integer.
14:56:56 <manu> Mark: It's beside the point, though, because we could have something like "23 blah"^^my:foo that only allows three values "red", "green" and "purple". The RDF processor is going to send that up, but the developer is still going to have to recognize that "23 blah" is not in the correct range. We need to let developers know that they should be aware of ranges.
14:58:26 <manu> q+ to end the call.
15:00:41 <manu> ack manu
15:00:41 <Zakim> manu, you wanted to end the call.
15:00:55 <Zakim> -toby
15:00:57 <Zakim> -markbirbeck
15:00:59 <Zakim> -Knud
15:01:04 <Zakim> -manu.a
15:01:06 <Zakim> -ShaneM
15:01:11 <manu> zakim, who is on the call?
15:01:11 <Zakim> On the phone I see manu
15:01:17 <manu> zakim, drop manu
15:01:17 <Zakim> manu is being disconnected
15:01:18 <Zakim> SW_RDFa()10:00AM has ended
15:01:20 <Zakim> Attendees were Knud, manu, +44.785.583.aaaa, toby, markbirbeck, ShaneM, +8014aabb
# SPECIAL MARKER FOR CHATSYNC.  DO NOT EDIT THIS LINE OR BELOW.  SRCLINESUSED=00000211