13:40:59 <RRSAgent> RRSAgent has joined #csvw
13:40:59 <RRSAgent> logging to http://www.w3.org/2015/11/04-csvw-irc
15:00:06 <gkellogg> gkellogg has joined #csvw
15:00:15 <JeniT> JeniT has joined #csvw
15:02:00 <gkellogg> Looks like it needs Ivan to start it?
15:02:27 <danbri> really? uhoh. he's in china
15:02:39 <jtandy> jtandy has joined #csvw
15:02:53 <gkellogg> We could try Google Hangouts, or Skype, I suppose.
15:02:57 <JeniT> ok, so perhaps we can just do this by IRC
15:03:09 <danbri> I've a schema.org bridge we can borrow if we want. But text mode is fine by me too.
15:03:15 <JeniT> I think all we need to do is propose & agree the transition to Proposed Recommendation
15:03:19 <danbri> yeah
15:03:22 <jtandy> ok
15:03:24 <JeniT> unless anyone had anything else that you think we need to discuss?
15:03:29 <danbri> there was one outstanding issue on you JeniT
15:03:44 <jtandy> ... and the UCR stuff?
15:03:52 <JeniT> the only issue I see on me is the Primer
15:04:00 <danbri> looks like you did it, whatever it was :)
15:04:00 <JeniT> given I made the change to the model document the other day
15:04:06 <danbri> https://github.com/w3c/csvw/issues?q=is%3Aopen+is%3Aissue+label%3A%22For+Proposed+Rec%22 wasn't empty before. Yeah that was it.
15:04:16 <JeniT> jtandy: and you did the UCR stuff?
15:04:21 <danbri> ok so is it just "This WG resolves that the specifications are complete, and requests transition to Proposed Recommendation (on Nov 24th 2015)."?
15:04:27 <jtandy> no -
15:04:34 <gkellogg> Theres https://github.com/w3c/csvw/issues/775, but I think there’s nothing to do
15:04:41 <jtandy> it was on Ivan's list of things ...
15:05:32 <danbri> see mail from Ivan, "UCR finalization?"
15:05:45 <jtandy> for the UCR stuff we're looking at ISSUES 536, 537, 538 and 539
15:05:58 <jtandy> (oh and 481)
15:06:06 <JeniT> that shouldn’t hold up our transition on the other documents
15:06:27 <JeniT> UCR is a Note, final version to be published when the other docs move to Rec (same time as Primer)
15:06:57 <jtandy> I see that Ivan asks if this can be done before 24-Nov. Unfortunately that's still a "No"
15:07:20 <jtandy> Quoting Ivan: "Do you think this is something you could do between now and the 24th, or should I gloss over this in my PR Transition request? The former would be better…"
15:07:43 <danbri> I think a fine gloss paint is in order here, so long as the larger schedule is considered viable / safe.
15:07:46 <JeniT> jtandy: it looks like the PR transition request that Ivan has currently put together glosses over it
15:08:04 <JeniT> danbri: I think we need to explicitly list the specs that we are agreeing to transition
15:08:09 <JeniT> in our resolution
15:08:11 <jtandy> that's good ... I've not read the details of the PR transition
15:08:18 <JeniT> https://github.com/w3c/csvw/blob/gh-pages/admin/PRTransitionRequest.txt
15:09:02 <JeniT> everyone OK if we go through the resolutions?
15:09:04 <jtandy> ah yes - I see: "The group intends to publish a final version of the UCR document as a WG Note at the time of publishing the REC"
15:09:25 <danbri> JeniT you want one big sausage of a resolution, or several small ones?
15:09:45 <jtandy> ... that would make it the same time as the Primer note gets published ... and danbri 's note about wrapping stuff in HTML
15:09:49 <JeniT> One big one is fine, but I think it needs to list the documents and say that we resolve to request the transition
15:09:52 <danbri> rrsagent, scribenick is danbri
15:09:52 <RRSAgent> I'm logging. I don't understand 'scribenick is danbri', danbri.  Try /msg RRSAgent help
15:10:04 <JeniT> ScribeNick: danbri
15:10:14 <danbri> attending: Jeni, Gregg, Jeremy, Dan; regrets: Ivan (china travel)
15:10:40 <jtandy> present+ JeniT
15:10:57 <jtandy> present+ jtandy, gkellogg, danbri
15:12:05 <gkellogg> rssagent, start
15:12:16 <danbri> rrsagent, start
15:13:32 <danbri> rrsagent, please make logs public
15:13:35 <danbri> rrsagent, please make minutes public
15:13:35 <RRSAgent> I'm logging. I don't understand 'please make minutes public', danbri.  Try /msg RRSAgent help
15:13:42 <danbri> rrsagent, please draft minutes
15:13:42 <RRSAgent> I have made the request to generate http://www.w3.org/2015/11/04-csvw-minutes.html danbri
15:14:03 <danbri> (it's funny, we each remember different bits of the bot command language)
15:14:19 <danbri> Ok, logs are logged.
15:14:42 <danbri> "Dearly beloved, we are gathered here today, in the presence of rrsagent, …"
15:15:00 <danbri> ok drafting give me a couple mins
15:16:23 <danbri> Do we want exact URLs in there?
15:16:30 <danbri> text otherwise I suggest is """The WG resolves that its work on "Model for Tabular Data and Metadata on the Web", "Metadata Vocabulary for Tabular Data", "Generating RDF from Web Tabular Data" and "Generating JSON from Web Tabular Data" is complete and requests that W3C advance them to Proposed Recommendation status."""
15:16:46 <danbri> Comments on the text welcomed, before I type it prefixed PROPOSED, after which we should +1.
15:17:16 <JeniT> just a sec
15:17:40 <JeniT> can we adjust the titles to the correct ones
15:17:52 <danbri> I copied these from Ivan :/
15:17:55 <JeniT> namely, the transformation specs actually are called “Generating JSON from Tabular Data on the Web” and “Generating RDF from Tabular Data on the Web”
15:18:12 <danbri> ok
15:19:02 <danbri> revision is """The WG resolves that its work on "Model for Tabular Data and Metadata on the Web", "Metadata Vocabulary for Tabular Data", “Generating RDF from Tabular Data on the Web” and “Generating JSON from Tabular Data on the Web” is complete and requests that W3C advance them to Proposed Recommendation status.
15:19:02 <danbri> URLs: http://w3c.github.io/csvw/syntax/index.html?specStatus=PR;publishDate=2014-11-24 http://w3c.github.io/csvw/metadata/index.html?specStatus=PR;publishDate=2014-11-24 http://w3c.github.io/csvw/csv2rdf/?specStatus=PR;publishDate=2014-11-24 http://w3c.github.io/csvw/csv2rdf/?specStatus=PR;publishDate=2014-11-24#h-sotd """
15:19:13 <danbri> shall I propose it?
15:19:23 <jtandy> yes
15:19:25 <JeniT> yep
15:19:38 <danbri> PROPOSED: The WG resolves that its work on "Model for Tabular Data and Metadata on the Web", "Metadata Vocabulary for Tabular Data", “Generating RDF from Tabular Data on the Web” and “Generating JSON from Tabular Data on the Web” is complete and requests that W3C advance them to Proposed Recommendation status.
15:19:38 <danbri> URLs: http://w3c.github.io/csvw/syntax/index.html?specStatus=PR;publishDate=2014-11-24 http://w3c.github.io/csvw/metadata/index.html?specStatus=PR;publishDate=2014-11-24 http://w3c.github.io/csvw/csv2rdf/?specStatus=PR;publishDate=2014-11-24 http://w3c.github.io/csvw/csv2rdf/?specStatus=PR;publishDate=2014-11-24#h-sotd 
15:19:42 <jtandy> +1
15:19:45 <danbri> +1
15:19:49 <gkellogg> +1
15:19:56 <JeniT> +1
15:20:00 <JeniT> +1 (from Ivan)
15:20:05 <danbri> RESOLVED: The WG resolves that its work on "Model for Tabular Data and Metadata on the Web", "Metadata Vocabulary for Tabular Data", “Generating RDF from Tabular Data on the Web” and “Generating JSON from Tabular Data on the Web” is complete and requests that W3C advance them to Proposed Recommendation status.
15:20:06 <danbri> URLs: http://w3c.github.io/csvw/syntax/index.html?specStatus=PR;publishDate=2014-11-24 http://w3c.github.io/csvw/metadata/index.html?specStatus=PR;publishDate=2014-11-24 http://w3c.github.io/csvw/csv2rdf/?specStatus=PR;publishDate=2014-11-24 http://w3c.github.io/csvw/csv2rdf/?specStatus=PR;publishDate=2014-11-24#h-sotd 
15:20:09 <danbri> rrsagent, pointer?
15:20:09 <RRSAgent> See http://www.w3.org/2015/11/04-csvw-irc#T15-20-09
15:20:17 <JeniT> excellent, thanks all
15:20:21 <danbri> Congratulations everyone :)
15:20:27 <JeniT> danbri: would you be able to put that in an email to the list please?
15:20:33 <danbri> Do we have any more business for today?
15:20:39 <jtandy> hooray ... ticker tape and trumpets blare
15:20:42 <danbri> Sure. Let's handle it as minutes.
15:20:49 <danbri> rrsagent, please draft minutes
15:20:49 <RRSAgent> I have made the request to generate http://www.w3.org/2015/11/04-csvw-minutes.html danbri
15:21:02 <JeniT> thanks everyone, for all your amazingly hard work
15:21:14 <danbri> actually Ivan tunes that garbled HTML into lovely minutes usually, I'll send a mail to be clear.
15:21:15 <gkellogg> nice to get over that hump!
15:21:17 <jtandy> thanks for thanks ... still a bit more to do tho
15:21:19 <danbri> Any other business for today?
15:21:36 <jtandy> none from em
15:21:38 <JeniT> just to say that I intend next to work on the Primer
15:21:40 <jtandy> s/em/me
15:22:03 <gkellogg> Danbri, you and I can work on the HTML stuff, if you like
15:22:10 <jtandy> (and to danbri - I've been doing a bit of follow up on weather schema)
15:22:14 <danbri> oh, we didn't include the proposed date. But we can leave that in Ivan + team's hands?
15:22:33 <JeniT> I think so
15:22:37 <danbri> I believe date is proposed for Nov 24. But we request the W3C to transition whenever it makes sense.
15:23:02 <jtandy> btw - I am travelling next week and won't be able to make any call should we choose to have one.
15:23:15 <danbri> gkellogg, on the HTML stuff … maybe a quick exchange. I wanted to ask if you know any clean way of saying "scoop out the <script> tags wherever you find them?
15:23:35 <jtandy> ciao
15:23:36 <danbri> I think we have to defer to the HTML(5+) parsing rules
15:23:46 <gkellogg> danbri: other than XPath?
15:23:47 <danbri> bye Jeni, and thanks!
15:23:49 <JeniT> I shall be on the call next week
15:23:52 <danbri> other than XPath.
15:23:52 <JeniT> bye
15:23:54 <gkellogg> bye Jeni!
15:23:57 <jtandy> bye
15:24:10 <gkellogg> Should be able to do it okay in JQuery
15:24:22 <danbri> More for spec language than implementation
15:24:30 <danbri> Implementors will do all kinds of horrid things, I'm sure.
15:25:01 <gkellogg> Ahh. I’d need to look at other specs. Turtle describes it using <script>
15:25:16 <gkellogg> http://www.w3.org/TR/turtle/#in-html-parsing
15:25:18 <danbri> But if we're writing a Note that says here is how to find JSON describing CSVs in a Web page, … we'll need to figure out what the raw materials are that we use to express that
15:25:26 <danbri> it will affect base URIs at least, right?
15:26:00 <danbri> there is something in the JSON-LD spec but I remember it left some point(s) vague - maybe just base URI?
15:26:14 <gkellogg> I believe the base URI for each script is that of the document containing them. We had discussed using @id, but I don’t think any spec does that
15:26:28 <gkellogg> JSON-LD does it similar to Turtle
15:26:39 <danbri> what if the document has base href="foo" declarations? ignored?
15:26:59 <danbri> so you might find it on a straw :8080 server and it'd mess up all the otherwise potentially wellknown URIs
15:27:00 <gkellogg> I believe that would set the base for <script> elements too.
15:27:14 <gkellogg> http://www.w3.org/TR/json-ld/#embedding-json-ld-in-html-documents
15:27:16 <danbri> hmm we should have said "Adjourned" to the bots. Ah well.
15:27:44 <danbri> "Defining how such data may be used is beyond the scope of this specification. " is vague w.r.t. base.
15:27:56 <danbri> but it provides justification for building on that definition
15:28:13 <danbri> Note: "Embedding CSVW JSON-LD in HTML Documents"
15:28:19 <danbri> 1. we build on http://www.w3.org/TR/json-ld/#embedding-json-ld-in-html-documents
15:28:30 <gkellogg> Pretty sure my implementation uses the document base as the base for the contents of <script>
15:28:34 <danbri> 2. we say how base URI is specified
15:28:54 <danbri> 3. we say that multiple blocks of script are extracted and concatenated (textually? as triples?)
15:29:22 <danbri> 4. we note that the surrounding markup (e.g. svg or whatever) is irrelevant - we extract it all (do we?)
15:29:42 <danbri> 5. we give an example of a single CSV being described, and a bundle of linked CSVs
15:29:58 <danbri> is that a rough todo? were you expecting something substantively different?
15:30:00 <gkellogg> For 3) both Turtle and JSON-LD treat these as separate documents, not concatenated
15:30:30 <danbri> I don't mind re (3) so long as we say something :)
15:30:41 <gkellogg> That’s pretty much it. I had considered if we could use a <script> to contain text/csv as well, but I don’t think we need to go there.
15:31:25 <gkellogg> For 3) perhaps we just treat each <script typle=“application/csvw+json”> element as a separate metadata document.
15:31:38 <gkellogg> Not of much practical use, but no reason it can’t work that I can think of.
15:32:49 <gkellogg> I can work on examples and code that makes it work. We can also do a separate non-normative test suite easily enough.
15:34:22 <danbri> the other 'obvious' thing is HTML tables
15:34:35 <danbri> but let's not get dragged down into handling those properly just yet
15:34:56 <danbri> it would be nice to see how our vocab looks interweaved into <table> via RDFa, but that is an added extra
15:34:58 <gkellogg> Yes, there may be some issues going there.
15:35:16 <gkellogg> Personally, I think it would be useful, but it could be a can of worms.
15:35:38 <gkellogg> Better do something like that as a social standard, and look to formalizing sometime later.
15:36:09 <gkellogg> However, I can also provide some examples for discussion, and that may reveil hidden issues.
15:37:14 <danbri> the last outstanding headache for me w.r.t. Embedding JSON descriptions of CSV in HTML" (apart from how odd it sounds :) is ---
15:37:27 <danbri> how does the Web page actually LINK to the CSVs?
15:37:41 <danbri> should it be "Embedding JSON descriptions of CSV in HTML (with a tiny bit of RDFa)"?
15:37:53 <danbri> a title designed to be mocked, if ever there was one
15:38:32 <gkellogg> I don’t think RDFa is necessary, but certainly compatible. My RDFa processor would extract RDFa, CSVW, Turtle, RDF/XML, JSON-LD and whatever into a single graph.
15:38:57 <danbri> but we agree that the human-facing html should link to the CSVs?
15:39:01 <gkellogg> How about “Describing Tabular Metadata in HTML”
15:39:05 <danbri> sure
15:39:46 <gkellogg> I think it would be a good idea for a <link rel=“describes”> element to point to the CSV, but I don’t see that actually being used. (for some value of @rel)
15:40:38 <gkellogg> There may also be a <link rel=“describedby”> pointing to an actual JSON file containing the same contents as the script tag.
15:41:20 <gkellogg> What about <script type=“application/csvw+json” src=“./metadata.json”></script> ?
15:42:11 <gkellogg> Not done for any other format, and could confuse HTML, so probably not. But, a describedby link pointing to a regular metadata file might be a suggested practice.
15:44:11 <danbri> fair enough
15:44:30 <gkellogg> I’ll work up some examples in a branch and ping you when I have something.
15:44:32 <danbri> you could imagine a utility Javascript thingy fleshing out the HTML with a view of the JSON (or more)
15:44:35 <danbri> great
15:45:08 <gkellogg> Got to go, have a great rest of your day!
15:49:30 <danbri> got to go too, take care!
17:20:11 <JeniT> JeniT has joined #csvw