W3C

- DRAFT -

RDF Working Group Teleconference

08 Aug 2012

See also: IRC log

Attendees

Present
Regrets
Chair
SV_MEETING_CHAIR
Scribe
pfps

Contents


<trackbot> Date: 08 August 2012

I can scribe.

<davidwood> scribenick: pfps

<gavinc> Not so much, http://www.w3.org/2011/rdf-wg/meeting/2012-07-25 404 :(

<gavinc> also http://www.w3.org/2011/rdf-wg/meeting/2012-07-11 not so much :(

Administration

david: last two minutes are not ready

<gavinc> 2012.07.25

<sandro> https://www.w3.org/2009/CommonScribe/panel/

gregg: 11 July minutes are OK, except that they don't identify markus
... I'll try to fix them up

<sandro> http://www.w3.org/2011/rdf-wg/meeting/2012-07-25

Both minutes appear to be OK to me.

<sandro> http://www.w3.org/2011/rdf-wg/meeting/2012-07-25-v1.html#line0097

david: is there a proposal from the straw poll on inverse in Turtle?

gavin: no, and no action

david: will you take an action?

gavin: OK

<davidwood> ACTION: Gavinc to create a proposal for inverse property syntax in Turtle [recorded in http://www.w3.org/2012/08/08-rdf-wg-minutes.html#action01]

<trackbot> Sorry, couldn't find user - Gavinc

<davidwood> ACTION: Gavin to create a proposal for inverse property syntax in Turtle [recorded in http://www.w3.org/2012/08/08-rdf-wg-minutes.html#action02]

<trackbot> Created ACTION-180 - Create a proposal for inverse property syntax in Turtle [on Gavin Carothers - due 2012-08-15].

<PatH> I'll have a peppermlnt, thanks.

<gavinc> Scribe problem: the name 'mlnt' does not match any of the 52 active names.

<gkellogg> http://www.w3.org/2011/rdf-wg/meeting/2012-07-11

<PatH> sounds like the olympics.

Now 11 July minutes look fine.

david: Any problems with 11 July minutes?

25 July minutes do not call out the straw pool

<gavinc> http://www.w3.org/2011/rdf-wg/meeting/2012-07-25#line0099

sandro: you can use the line links

<sandro> eg http://www.w3.org/2011/rdf-wg/meeting/2012-07-25#line0097

Proposed: accept minutes for 11 and 25 July

Subtopic: Action item review

david: my action is done, close?

<sandro> +1

OK

Close action 177

<sandro> resolved: accept minutes for 11 and 25 July

Actions

<davidwood> http://www.w3.org/2011/rdf-wg/track/actions/open

<sandro> action-177?

<trackbot> ACTION-177 -- David Wood to create blog post on swig blog for Turtle LC -- due 2012-07-18 -- CLOSED

<trackbot> http://www.w3.org/2011/rdf-wg/track/actions/177

pat: I claim success on action 178!

<davidwood> http://www.w3.org/2011/rdf-wg/track/actions/178

<ivan> Pat's mail on the provenance issue: http://lists.w3.org/Archives/Public/public-rdf-wg/2012Aug/0014.html

pat: I'll fix up the action status and include the email link.

<sandro> https://www.w3.org/2005/06/tracker/users/my

<mlnt> thanks david

Subtopic: next meeting

david: 22 Aug, then 5 Sept, then weekly
... 22 Aug will be on named graphs

SubTopic: F2F

<PatH> yeh, that *hurt*.

<gavinc> Is there still no remote possibility at TPAC?

F2F is at TPAC (29-30 Oct)

david: start wiki page on F2F

<ivan> Participants should also register on https://www.w3.org/2002/09/wbs/35125/TPAC2012/

<scribe> ACTION: david to start wiki page for F2F [recorded in http://www.w3.org/2012/08/08-rdf-wg-minutes.html#action03]

<trackbot> Created ACTION-181 - Start wiki page for F2F [on David Wood - due 2012-08-15].

david: If you are attending in person please register for TPAC and also pay (separate page)
... sandro/ivan remote participation?

sandro: phone bridge via speaker phone and usual machinery

david: should we have a remote meeting room (US?)

sandro: no one will be at MIT

<PatH> In the past, skype with video has been useful.

david: I'll ask on in the wiki page and ask who would like to be in a remote meeting

arnaud: note that we can't change the meeting time

david: sure, so it'll be very early in the US

<PatH> share the jetlag. Seems kind of socialist to me.

Provenance LC reviews

david: volunteers?

path: I checked some issues, but I don't know whether that it a review

david: we owe the PWG a formal review, any volunteers?

pat: what is a formal review

david: at least everything relevant to us, but maybe everything
... anyone can comment
... the RDF WG should have a formal response, however

Turtle LB

Turtle LC

david: gavin, how did you set up LC tracking

gavin: poorly, as I had problems with connecting

david: it is important that *everyone* gets a response, but an entire thread can have a single response, as long as everyone is CC'd
... any non-editorial issues need to be highlighted so the WG looks at it

gavin: only one so far is inverses

ivan: when we come to PR we need an overview for all comments, so tracking using W3C resources can help

david: I had proposed to use tracker with a special product for Turtle LC

ivan: OK
... adding inverse properties will likely trigger a second LC

david: it would be nice if we can get a second LC started before the first LC ends

ivan: process probably requires second LC start after the first one ends

sandro: do we need to wait? It would be odd, but why would anyone be bothered?

<PatH> +1 sandro

macted: if there is another change, then it might trigger a third last call

Gavin: inverses may trigger a lot of comments

<sandro> (gavin is the person talking other than Ivan, David, and me)

david: did anyone ask Dave Beckett, he had an opinion on no changes

gavin: Dave felt that changes should be only for major pain points, evidenced by screaming

sandro: It appears that inverse meets the bar

<SteveH> I'm not convinced it's good idea either FWIW

sandro: If people add inverse properties solely to get the effect of inverse Turtle lines then that is good evidence

<AndyS> The divergence from SPARQL (Update) is a cost.

<PatH> And RDFa effectively has inverse and it works fine.

gavin: there is no implementation response, and people use inverses even when the syntax does support inverted predicates

<Zakim> manu, you wanted to ask about hooking JSON-LD issue tracker up to public-rdf-comments (new topic)

<gavinc> rev is NOT a good example of people using it :P

<davidwood> manu, we will discuss that topic next

<manu> Also, the WHATWG /hates/ @rev

ivan: RDFa has inverses, but it is rarely used

<PatH> I use it all the time. Maybe I am wierd.

ivan: I found it useful in a few cases

<manu> and Digital Bazaar has found that @rev is very useful.

gavin: I can't explain @rev in RDFa

<davidwood> 3 Round Stones and Callimachus users make great use of @rev

<manu> Their reason is that "regular authors" don't understand what @rev does...

<manu> (which is true)

ivan: so it's rarely used, and then only by experts

<manu> also... "regular authors" don't understand... WebGL, Web Audio, Canvas, <i> vs. <em>, etc.

<MacTed> "when you want it, you *really* want it..."

<PatH> We have three hard use cases versus a mess of anecdote. Sorry, but thats the way it seems.

gavin: I don't see any examples of reverse in N3 or Manchester

<SteveH> it's sugar, so there's no situation where you can /need/ it

<SteveH> though it may make your text look neater

<AndyS> :x^:p is [:p :x] and is in the subject or object position.

david: syntactic sugar for reverse can be essential if you can't write them the wright way around

<PatH> lol

SteveH: you can always (in Turtle) flip things around

david: you can't always do this if you don't control the data

<AndyS> I note this does not answer the original commenters request nor adopts N3.

SteveH: I don't see this, but let's discuss off-line

ivan: what's the route to the sea

david: waiting for gavin's proposal

<PatH> downhill, usually.

andy: adding reverse doesn't answer the original comment

david: let's await gavin's proposal

<AndyS> See "Paths" in http://www.w3.org/TeamSubmission/n3/

<gavinc> WG said ^, TimBL said is .. of

ivan: it is the WG's responsibility to decide what to do

<sandro> echo ":a is :b of :c." | python swap/cwm.py -quiet

<sandro> @prefix : <#> .

<sandro> :c :b :a .

<davidwood> http://www.w3.org/TeamSubmission/n3/#path

andy: N3 has something different - paths

sandro: caret is different from ... is ... of

<PatH> caret?

<SteveH> ^

<gavinc> Don't need @ "If no @keywords directive is given, qualified names all have colons, and unquoted alphanumerics are all keywords. Only the keywords is, of, and a may be used naked."

david: andy, does TimBL want both bang and caret?

andy: don't know, I believe that it is is/of

gavin: the WG didn't like is/of

sandro: caret is very close to is/of where it can occur

andy: yes

<PatH> So it has come to this, exploding carrots.

gavin: no implementation experience, even N3 is different

andy: is/of has a NL flavour, which caret doesn't, so there is a big difference

sandro: the WG didn't like an NL approach

<gavinc> "The reverse traversal, x^p means [ p x ] . For either forward or backward traversal, p is a property, and x can be a whole path with both ! and ^ in it."

<AndyS> :x^:p!:q :prop 123 .

andy: [N3 is much more complex]

ivan: our caret is part of N3 reverse traversal

<gavinc> "I would like to formalize the request I have made at intervals to include the is ... of syntax as in N3 in Turtle."

sandro: I would like to respond that caret does something similar to SPARQL

gavin: can we first respond directly to the commenter to see if our solution is acceptable

david: there are two different solutions here

<PatH> My understanding is that none of these syntaxes actually *create* the inverse property, they are all syntactic ways to indicate a p b while writing some thing with the a and b in the reverse order.

<PatH> DO I have that wrong?

<AndyS> yes

<PatH> OK, ta.

sandro: they are both syntactic sugar and neither creates a new property

<AndyS> :a ^:p :b .

<AndyS> :b :p :a .

andy: which means that Turtle is further from the triples

<PatH> That seems like an argument against any kind of parsing at all.

gavin: there is no way to regenerate the carets

<SteveH> the store could tag triples that were inverse, but lets not go there

<PatH> indeed.

<zwu2> no tagging please

david: what to tell Gavin?

ivan: let's tell Tim that the NL proposal won't fly; that we are considering using caret, which clashes with N3

<cygri> +1 to ivan's proposal

david: so let's wait for a proposal until we get information back from Tim
... gavin, can you respond to Tim

gavin: OK

JSON LD comments

david: can we hook up the Jason LD tracker to rdf comments?

??: it should be possible using some automated filter

??: we are using git-hub currently

sandro: will this push all JSON-LD emails into the WG mailing list

Gregg: no, we'll have a filter in the middle
... right now comments are supposed to go to rdf-comments, but people can directly comment on the issues in the tracker

sandro: a manual approach would be to just tell posters to also send to rdf-comments

<gavinc> No, it doesn't work if it goes via a mailing list

<mlnt> you have to have a GitHub account to be able to reply by mail

gregg: we would have to test the various situations

<gavinc> do we have an -archive for the RDF-WG?

sandro: the W3C process is that the everything ends up on the email list

<sandro> sandro: I'd suggest once a week summarizing the github activity to public-rdf-comments

manu: we could do a summary, but an automated link might be closer and better

<PatH> JSON really makes things so very simple. I love it.

sandro: if you need to follow the JSON-LD stuff you really need to be on github, so a summary might be best

<gkellogg> +0

<cygri> -1

<ivan> -1

<sandro> strawpoll: +1 if you want the mailing lsits bridged -1 if not

<manu> +1 to have it reflected

<davidwood> +0

<gavinc> -0.1 to more github emails

<mlnt> +1

<AndyS> 0

<SteveH> -1

<ivan> to be clear: it may be good, but it leads to too much confusions

gregg: OK, we'll do the summary method

sandro: the -comments list needs to get anything that is non-editorial

gregg: OK

<sandro> sandro: excellent

AOB

david: done
... 22 Aug, "Return to Named Place"

gavin: agenda request, n-triples

Summary of Action Items

[NEW] ACTION: david to start wiki page for F2F [recorded in http://www.w3.org/2012/08/08-rdf-wg-minutes.html#action03]
[NEW] ACTION: Gavin to create a proposal for inverse property syntax in Turtle [recorded in http://www.w3.org/2012/08/08-rdf-wg-minutes.html#action02]
[NEW] ACTION: Gavinc to create a proposal for inverse property syntax in Turtle [recorded in http://www.w3.org/2012/08/08-rdf-wg-minutes.html#action01]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.136 (CVS log)
$Date: 2012/08/08 16:12:48 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.136  of Date: 2011/05/12 12:01:43  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/77/177/
Succeeded: s/weakly/weekly/
Succeeded: s/??/Gavin/
Succeeded: s/??/SteveH/
Succeeded: s/??/SteveH/
Succeeded: s/carrot is different/caret is different/
Succeeded: s/Time/TimBL/
Succeeded: s/Jason/JSON/
Succeeded: s/??/Gregg/
Found ScribeNick: pfps
Inferring Scribes: pfps

WARNING: No "Present: ... " found!
Possibly Present: AZ AlexHall AndyS Arnaud Gavin IBM Ivan MacTed OpenLink P17 P28 P3 P35 P8 P9 PatH Proposed SteveH SteveH__ SubTopic aaaa aabb aacc aadd andy cgreer cygri david davidwood ericP gavinc gkellogg gregg https joined manu mhausenblas mlnt pat rdf-wg sandro scribenick strawpoll trackbot zwu2
You can indicate people for the Present list like this:
        <dbooth> Present: dbooth jonathan mary
        <dbooth> Present+ amy


WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Found Date: 08 Aug 2012
Guessing minutes URL: http://www.w3.org/2012/08/08-rdf-wg-minutes.html
People with action items: david gavin gavinc

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.


[End of scribe.perl diagnostic output]