RDF Working Group Teleconference -- 18 May 2011

<trackbot> Date: 18 May 2011

<hsbauer> zakim 1.507.261.aacc is Scott

<MacTed> :-)

admin

<zwu2> Guus: PROPOSED to accept the minutes of the 11 May telecon:

<cygri> +1

<pfps> minutes look ok to me

<zwu2> ... http://www.w3.org/2011/rdf-wg/meeting/2011-05-11

<zwu2> Accepted.

Action item review

<zwu2> pending review items: http://www.w3.org/2011/rdf-wg/track/actions/pendingreview

<zwu2> Guss claim victory on Action 46

<ww> respec++

<PatH> I put a brief text on the wiki re. action 26

action items: http://www.w3.org/2011/rdf-wg/track/actions/open

<zwu2> action-26?

<trackbot> ACTION-26 -- Patrick Hayes to write an description of action-21 -- due 2011-04-13 -- OPEN

<trackbot> http://www.w3.org/2011/rdf-wg/track/actions/26

<zwu2> who is speaking?

<zwu2> thanks

<zwu2> guus. keep action 26 open

<gavin> ACTION-26?

<trackbot> ACTION-26 -- Patrick Hayes to write an description of action-21 -- due 2011-04-13 -- OPEN

<trackbot> http://www.w3.org/2011/rdf-wg/track/actions/26

<zwu2> ACTION-33?

<trackbot> ACTION-33 -- Dan Brickley to danbri, you wanted to note a bug in RDFS spec; it references Primer example 16 -- an example that doesn't even use rdf:value. -- due 2011-08-21 -- OPEN

<trackbot> http://www.w3.org/2011/rdf-wg/track/actions/33

<zwu2> Dan is not here

<zwu2> Guss. keep Action 33 open

<gavin> close ACTION-43

<trackbot> ACTION-43 Compare/contrast respec vs mediawiki for spec authoring closed

<zwu2> Action-43?

<trackbot> ACTION-43 -- Gavin Carothers to compare/contrast respec vs mediawiki for spec authoring -- due 2011-05-11 -- CLOSED

<trackbot> http://www.w3.org/2011/rdf-wg/track/actions/43

<danbri> oh i'm sorry :( regrets mostly, i'll only be in irc

<zwu2> Guss: asking Gavin for confirmation. can mark the actions closed

<zwu2> Action-44?

<trackbot> ACTION-44 -- William Waites to compare/contrast respec vs mediawiki for spec authoring -- due 2011-05-11 -- OPEN

<trackbot> http://www.w3.org/2011/rdf-wg/track/actions/44

<danbri> (but noted/acked. i'll fix when i put rdfs into merurial repo)

<ww> details to follow

<ww> sorry for lateness

<pchampin> same here, I'm affraid :-(

<zwu2> Guss: we might as well close it

<zwu2> ... we will close it if no objection

<ww> briefly, respec++ :)

<zwu2> Action-45?

<trackbot> ACTION-45 -- Pierre-Antoine Champin to compare/contrast respec vs mediawiki for spec authoring -- due 2011-05-11 -- OPEN

<trackbot> http://www.w3.org/2011/rdf-wg/track/actions/45

<zwu2> Guus: let's take a decision and close it

<PatH> the noise level just increased.

<PatH> Look under issue-21

<zwu2> Guus. done with action item review

F2F Poll

<zwu2> Guus: the poll is now closed.

<zwu2> ... MIT is the winner

<zwu2> ... date wise, very slight preference for Oct 12

<zwu2> ... almost the same video link wise

<zwu2> ... I tend to propose Oct 12 at MIT

<zwu2> ... anyone want to discuss?

<zwu2> PROPOSED: second F2F to be held at MIT in Oct 12

<davidwood> seconded

<zwu2> no objections heard

<pfps> Given that the F2F is going to be at MIT, can we get some information on reasonably priced hotels?

<zwu2> ACCEPTED: second F2F to be held at MIT in Oct 12

<davidwood> Yes, thanks to Peter.

<zwu2> thanks peter for provding an alternate venue

<zwu2> Guss. we will dicuss F2F3 in the near future

editing documents

<zwu2> Guus. tools have been looked at, any winner?

<ww> hg + respec

<zwu2> david: we are defining the default

<zwu2> Guus: one is mediawiki

<zwu2> .. the other is respec

<zwu2> ... which has some builtin javascript to do some useful stuff, html5 based

<zwu2> richard: there is a strong case for making a binding decision on editors for version control

<ww> problem with wiki is, "how do i use my favourite ${EDITOR}"

<zwu2> ... edit in html is reasonable

<zwu2> ... can do it in a text editor or other tools

<PatH> +q

<zwu2> gavin: from the discussions, leaning towards to html5 and javascript

<zwu2> guus: version control is a good point

<zwu2> ... sandro metioned that handling cross references is hard

<zwu2> ... he wants to agree on reference support

<Zakim> Guus, you wanted to discuss reference support

<zwu2> PatH: I don't like any of these ideas

<zwu2> ... I will prodcue legal HTML my way

<zwu2> Guss. I don't think that is a problem

<zwu2> ... it is good to keep citations consistent

<zwu2> Davidwood: sandro will be unhappy to tidy up all citations

<zwu2> ... on other people's behalf

<zwu2> Guss. that is the only reason for standardization

<zwu2> ... if we use version control, shall we use the same?

<zwu2> richard: it makes sense to use a single repository

<PatH> Why does this make sense? Seems to me that it is just an extra burden on editors.

<zwu2> ... I would like to know the location to fetch the latest versions

<ww> +1 for common repository - but mind that people can still use what they want, git-hg, hg-svn, whatever

<zwu2> Guss: PatH can you live with that?

<gavin> http://mercurial.selenic.com/

<gavin> Yes.

<zwu2> PatH: sure. can someone show me how to do it on a Mac?

<gavin> Runs just fine on a mac

<zwu2> Davidwood: it runs on Mac

<zwu2> Guus: maybe we can ask someone to provide a tutorial or send a pointer

<zwu2> PatH: we did the first version of RDF via emails...

<gavin> Pat, you are of course welcome to simply do all your work outside of version control and then commit it all at the end ;)

<zwu2> Guus: go with HTML, if you want Sandro to do the citations, please use the standard way

<PatH> I am installing Mercurial now. Sigh.

<zwu2> :_

<zwu2> :)

<zwu2> one sec,

<zwu2> what is the passcode?

<AZ> 73394

<ww> scribenick: ww

guus: one reason to settle this this week...because we need to start editing these documents
... as long as this delay doesn't prevent us...

<zwu2> Guus: should not prevent us from editing this week

<zwu2> ... concept document

<scribe> scribenick: zwu2

Guus: as a general rule, I'd like to replace the names of all editors
... how did OWL2 do this?

peter: pretty much all documents are fresh

Guss: how is it done in SPARQL?

too much noise...

<AndyS> SPARQL uses xmlspec

<AndyS> and that has prev editor

<davidwood> See http://www.w3.org/2001/06/manual/#Editors

mainly the question is do we have to convert the old docs to respec,

scribe: before we should put in version control?

<davidwood> W3C Manual of Style, Section 5.2.1 Managing Changing Affiliations

scribe: I understand that converting may be too much work

Guus: that is not a big deal according Gavin

Gavin: no idea how to do the grammar section of Turtle
... otherwise, it is easy

Guus: would be nice to do the conversion, as a token to move forward,
... PatH please do the same
... create a new version of RDF semantics
... exactly the same as the previous version

PatH: I can do it

<PatH> I have absolutely no idea what the speaker is talking about.

Davidwood: do we need to change javascript to do it?

who is speaking?

<davidwood> Peter

which peter?

<ww> http://dev.w3.org/2009/dap/ReSpec.js/template.html

<davidwood> Peter Patel-Schneider

<ww> around line 48 for editors setting

<cygri> is http://www.w3.org/2011/rdf-wg/wiki/Editors up-to-date?

Guus: if you don't use respec, then you don't need to worry about it
... as far as I know, the list is up to date

davidwood: PathH, you agree to edit the RDF semantics?

<PatH> Right, 2 editors, PatH and pfps.

PatH: yes

Guus: apart from PatH, the doc looks ok

SPARQL Last Call

http://lists.w3.org/Archives/Public/public-rdf-wg/2011May/0192.html

Guus: normally we have two kinds of reviews for LC
... 1) individual member can comment
... 2) the working group can review. have a couple of designated members to review and comment on behalf of the whole WG
... I prefer to also do the 2)
... is that necessary?

<gavin> Graph terminology http://www.w3.org/TR/2011/WD-sparql11-http-rdf-update-20110512/#terminology

<davidwood> I've updated http://www.w3.org/2011/rdf-wg/wiki/Editors

Guus: who wants to do it?
... should be people not in the SPARQL WG

<pchampin> what's the deadline?

<scribe> ACTION: Guus to send a message to the mailing list [recorded in http://www.w3.org/2011/05/18-rdf-wg-minutes.html#action01]

<trackbot> Created ACTION-47 - Send a message to the mailing list [on Guus Schreiber - due 2011-05-25].

<yvesr> pchampin, +1, what's the deadline?

<Guus> http://lists.w3.org/Archives/Public/public-rdf-wg/2011May/0192.html

<ww> [SPARQL] Review comments welcome through 29 July.

<pchampin> ok thanks

guus: it is a good practice trying give reviews asap
... so that they can keep their schedule

<pchampin> +1 to involve Sandro

<SteveH> I'll join

guus: anyone wants to join the telecon to discuss graph

<pchampin> I'm interested too

<AndyS> Please avoid semtech dates

<AndyS> I'm interested (on both sides)

guus: we have three people agreed. expect sandro to join as well
... that is all we can do for now.

ISSUE-40?

<trackbot> ISSUE-40 -- Skolemization advice in the RDF dcocument -- raised

<trackbot> http://www.w3.org/2011/rdf-wg/track/issues/40

Guus: can we make a agreement?
... let's see if we can move forward

<cygri> http://www.w3.org/2011/rdf-wg/wiki/Skolemization

cygri: gone through a few iterations
... by different people,
... the latest version has quite wide support
... I don't recall anyone raising objections

thanks ww

Guus: 6.6.1 in that document is the one we are talking about

cygri: your proposal is right under

<cygri> ISSUE-12?

<trackbot> ISSUE-12 -- Reconcile various forms of string literals (time permitting) -- open

<trackbot> http://www.w3.org/2011/rdf-wg/track/issues/12

Guus: can the issue owner make a proposal?

<cygri> PROPOSAL: Resolve ISSUE-40 by adding text to RDF Concepts, per the “Updated Proposal” from http://www.w3.org/2011/rdf-wg/wiki/Skolemization#Updated_Proposal

<gavin> ISSUE-40?

<trackbot> ISSUE-40 -- Skolemization advice in the RDF dcocument -- raised

<trackbot> http://www.w3.org/2011/rdf-wg/track/issues/40

<yvesr> cygri, are we considering PatH's one or the one above?

<yvesr> there are two different formulations in this doc

<cygri> yvesr, the first one

<yvesr> ok

Guus: any more discussion?

pfps: I worry about the wording.
... having trouble find name, I don't think name should be in the text at all

guus: is there an easy patch?

<cygri> In the RDF abstract syntax, a blank node is just a unique node that can be used in one or more RDF statements, but has no intrinsic name.

cygri: I quote something from rdf concept now
... I repeat the same thing here

<davidwood> +1

cygri: it is consistent (probably not the best wording)

<cygri> http://www.w3.org/TR/rdf-concepts/#section-URI-Vocabulary

pfps: not sure which RDF concepts doc you are looking at

cygri: section 3.2 of the same doc

pfps: seciton 3 is all informal

cygri: pfps, can you propose a fix?

pfps: it is not just that one setence, the whole paragraph has to be carefully crafted

guus: pfps, can you come up with something

pfps: sure. will take a bit time
... technical details are correct

guus: we can accept the resolution with an action to polish the wording

pfps: yes.
... I will produce edits to Section 3.2 as well

<pchampin> you can't ear me :-(

pchampin?

<pchampin> sorry about that

<gavin> Pointless possibly annoying question, "Given two blank nodes, it is possible to determine whether or not they are the same." ... is that true if the two blank nodes come from diffrent graphs?

<pchampin> it was just about the "SteveH" part of the proposal

<pchampin> that needs to be sorted out

<SteveH> pchampin, Note: “SteveH” is a placeholder. Names currently under discussion are “genid”, “bnode”, “skolem”.

<pchampin> I know

noisy

<pchampin> I agree with Richard

<pchampin> of course

cygri: the SteveH is just a placeholder

<PatH> Gavin, if they come from different graphs they must be different.

<pchampin> ok with me

<Zakim> pchampin, you wanted to notice that the wiki page still lets open the issue of replacing "SteveH" by something

<AndyS> PatH, subgraph? (not two doc read in)

<cygri> +1

PROPOSAL: Resolve ISSUE-40 by adding text to RDF Concepts, per the “Updated Proposal” from http://www.w3.org/2011/rdf-wg/wiki/Skolemization#Updated_Proposal

<MacTed> +1

<SteveH> +1

<AndyS> +1

<pchampin> +1

<gavin> +0

<AZ> +1

<yvesr> +0

<PatH> Andy, yes.

<PatH> +1

<pfps> +1

<OlivierCorby> +1

<cmatheus> +1

Guus: we can close issue-40 now

<ww> PatH: so bnode in subgraph shadows bnode in supergraph?

issue-12?

<trackbot> ISSUE-12 -- Reconcile various forms of string literals (time permitting) -- open

<trackbot> http://www.w3.org/2011/rdf-wg/track/issues/12

guus: where to start?

<PatH> Take this bnode thing offline, guys.

<cygri> +1 PatH

<PatH> Yes please

<PatH> What do we want the type of "foo" to be: plainliteral, xsd:striong, something else, or no type?

<Guus> http://lists.w3.org/Archives/Public/public-rdf-wg/2011May/0230.html

PatH: looking at all the debates, I think people have different positions on which is the desrible solution
... we should get it clear and take a binding decision

guus: I think you are saying that once we are clear on this, the other issues will follow

cygri: kind of agree. we should start to agree on the behavior, and then on the machinary
... maybe we should go back a bit
... if we have "" and ""^^xsd:string, are they the same?

<SteveH> I don't think there's a clear message from SPARQL

<PatH> Yes, sparql clearly treats seems to have decided on xsd:string.

<gavin> No objection to Lee's answer from me

Guus: can we live with that?

PatH: there are reasons for people to tream "chat"@en and "chat"@fr as different strings

<AndyS> Currently datatype("foo") = xsd:string datatype("foo"@en) = error [and an extension?]

PatH: it is odd to see datatype changes when there is a language tag
... added

<gavin> mmmm

<gavin> Hey, err, what about just forcing "" == ""@und?

<pchampin> @gavin: that would make "" != ""^^xsd:string, then

<SteveH> AndyS, does it explicitly say it's an error? http://www.w3.org/TR/rdf-sparql-query/#func-datatype

<ww> what about datatypes as used with skos:notation - a little like "language" no?

PatH: plain literal is now neutral in RDF

<AndyS> SteveH - yes - not mentioned in dispatch => no dispatch => error in basic SPARQL

<SteveH> AndyS, ok

PatH: i like strings too

<pfps> But plain literals without language tags are semantically the same as xsd:string.

pfps: strings are strings, 01 is syntatically different from 1, I don't care

<AndyS> Could be expressed more clearly but there are lots and lots of such cases. Editorial.

<gavin> @pchampin Yeah, I know. But why the heck is ""@en != ""^^xsd:string then?

<pchampin> @gavin, because it has a language tag. Strings don't.

<MacTed> it seems that ""@en is subtype of ""^^xsd:string ...

<ww> MacTed: +1

<gavin> @pchampin Yeah :\ Ugh, hard to explain to programers

<PatH> +q

<MacTed> in other words, ""@en is ""^^xsd:string plus a lang property

<ww> @en - syntactic sugar for ^^englishString

guus: we can take Lee's position for the moment

<MacTed> changing datatypes is NOT aesthetic...

PatH: we can leave the current syntax as is

<davidwood> The problem with xsd:string is that XSD (*all* of XSD) is RDF's extension mechanism for types.

<SteveH> <literal datatype="http://www.w3.org/2001/XMLSchema#string" xml:lang="en-GB">foo</literal> (SPARQL Result format) matters

<ww> too revolutionary to suggest that there are no languages, they're just certain string-derived datatypes?

<Guus> AndY?

PatH: we should decide on which one we like and update the syntax

<AndyS> Languages matter - we see this used for which label to display in apps

<SteveH> strawpoll?

a strawpoll is a good idea

<SteveH> prefer plain literal

<pfps> prefer rdf:plainLiteral

guus: +1 if you prefer plain literal or xsd:string

<PatH> Im am unclear what we ar voting on.

<cygri> prefer plain literal or xsd:string, but not rdf:PlainLiteral

<AndyS> Either no change, or simple literal

<PatH> still unclear.

<PatH> IS the question, what should be the type of "foo" ?

<davidwood> Unclear to me, due to the relation of xsd:string to the rest of XSD.

sorry

<cygri> Straw poll on: in the abstract syntax, if we want to have a single representation for strings (with and without language tag), which would it be?

<AZ> bye

<pchampin> bye

<ww> by all

thanks

bye

<MacTed> trackbot, end meeting

<gavin> BNods offline conversation at some point would be helpful!

- DRAFT -

RDF Working Group Teleconference

18 May 2011

Attendees

Contents