<trackbot> Date: 10 October 2012

(probably)

I can scribe

<scribe> scribenick: AndyS

admin

minutes of last meeting -- http://www.w3.org/2011/rdf-wg/meeting/2012-10-03

<pfps> minutes look ok

guus: accepted

RESOLVED to accept the minutes of 2012-10-03 (http://www.w3.org/2011/rdf-wg/meeting/2012-10-03)

subtopic: action items

no telecon 31 Oct (same week as face-to-face)

Turtle Last Call

next meeting next week : 17 Oct

<ericP> http://www.w3.org/2011/rdf-wg/wiki/I18n-Comments

<ericP> http://www.w3.org/2011/rdf-wg/wiki/I18n-Comments

Guus: issue numbers are in I18N tracker.

(the proposals are linked to by Nack/ack)

ericP: 178 -- BCP 47 reference

<Guus> PROPOSAL: wedge in a not-too-contrived ref in section 7 Parsing: "and an optional language tag [BCP47] or datatype IRI" per editor's draft.

<gavinc> +1 already completed in ED

Guus: any discussion? no ...

<pfps> go for it

<MacTed> issue-178?

<trackbot> ISSUE-178 does not exist

Guus: resolved

<gavinc> I18N-ISSUE-178?

RESOLUTION: wedge in a not-too-contrived ref in section 7 Parsing: "and an optional language tag [BCP47] or datatype IRI" per editor's draft.

Guus: 179 ...

gavinc: scope of language e.g. common langauge for all doc.

<gavinc> PROPOSAL: Turtle will not add a method to declare the language of the document.

gavinc: proposal - not to add

<pfps> good idea to not do it

RESOLUTION: Turtle will not add a method to declare the language of the document.

<ericP> http://dvcs.w3.org/hg/rdf/raw-file/default/rdf-turtle/index.html#turtle-literals

guus: next ... 180 ...

PROPOSAL: add this markup [[ (<span class="codepoint">#005E</span>)"]] to the first reference to each character per the editor's draft <http://dvcs.w3.org/hg/rdf/raw-file/default/rdf-turtle/index.html#turtle-literals>.

<pfps> obnoxious, but probably a good idea

RESOLUTION: add this markup [[ (<span class="codepoint">#005E</span>)"]] to the first reference to each character per the editor's draft <http://dvcs.w3.org/hg/rdf/raw-file/default/rdf-turtle/index.html#turtle-literals>.

<patH> good+obnoxious = medicinal?

guus: ... 181 ...

<ericP> http://dvcs.w3.org/hg/rdf/raw-file/default/rdf-turtle/index.html#turtle-literals

<pfps> sounds reasonable

PROPOSAL: add example of non-ASCII IRI

<yvesr> +1

RESOLUTION: add example of non-ASCII IRI

guus: ... 182 ...

ericP: need a sublanguage example. Don't see how it can be better.

<pchampin> well for the record, we had this TV show in France :)

<pfps> ok by me

<patH> agree this is a good example.

pchampin: could use colour/color in @en

PROPOSAL: 182 -- No change

RESOLUTION: 182 -- No change

<ericP> http://dvcs.w3.org/hg/rdf/raw-file/default/rdf-turtle/index.html#abbrev

ericP: 183 -- work in progress

<AZ> my:productInStock :hasQuantity 10

<pfps> you could always cheat and have the property be something like usdebtindollars

gavinc: good relations as a source of examples? Need xsd:double as well.

<gkellogg> Avagadro's Number?

<patH> engineering data uses doubles somewhere?

<patH> Check out ISO definition of a mole.

<patH> Aaaarghhhh

<patH> I will attempt it.

<pfps> along the same lines, one could use dimensionless constants

gavinc: 183 still open at the moment

guus: need a plan

ericP: ping with nack unless you (i18n) have an example

<pfps> fine, as long as we close the issue pending further input

<ivan> +1 to pfps

PROPOSED: i18n-183 -- EricP to respond with "no change unless you have further input"

RESOLUTION: i18n-183 -- EricP to respond with "no change unless you have further input"

guus: ... 184 ...

<ericP> http://www.w3.org/mid/20121002044103.GD27799@w3.org

<pfps> +1

<ivan> +1

RESOLUTION: i18n-184 -- Use the text U+xx form [[ U+B7, U+300 to U+36F and U+203F to U+2040 are permitted anywhere except the first character. ]] per the editor's draft <http://dvcs.w3.org/hg/rdf/raw-file/default/rdf-turtle/index.html#BNodes>

ACTION EricP Respond to i18n for issue 183

<trackbot> Created ACTION-188 - Respond to i18n for issue 183 [on Eric Prud'hommeaux - due 2012-10-17].

ericP: line terminators 185

PROPOSAL: Alter the text introducing the 4th example in §3 to say " The line breaks in this example are LINE FEED characters (U+0A)." per editor's draft <http://dvcs.w3.org/hg/rdf/raw-file/default/rdf-turtle/index.html#sec-examples>

<pfps> picky, but works for me

RESOLUTION: Alter the text introducing the 4th example in §3 to say " The line breaks in this example are LINE FEED characters (U+0A)." per editor's draft <http://dvcs.w3.org/hg/rdf/raw-file/default/rdf-turtle/index.html#sec-examples>

<gavinc> PROPOSAL: Turtle will continue to use \u \U escape syntax.

gavinc: 187 -- \u escape

<pfps> agreed, we are not going to fix this issue

<patH> +1

RESOLUTION: Turtle will continue to use \u \U escape syntax.

<pfps> well, we are following the crowd, which may not be going in the "right" direction

ericP: 188 - special handling of % in IRI
... what we have is correct - discussion has happened already.

<gavinc> damn, ISSUE-186 is for us

<ericP> http://www.w3.org/2011/rdf-wg/wiki/I18n-Comments#189:_.5BS.5D_reference_obs-language-tag_instead_of_defining_your_own

RESOLUTION: i18n-188 no change -- special handling of % in IRI

ericP: ... 189 ... advice from BCP 47 for language tags
... ABNF is quite complicated and much more than RFC.
... historical was 8 char/numbers per section between the "-"

PROPOSAL: Use {1,8} in lang tag production and {..} for \u \U escapes

(variation to SPARQL) Minor

<gavinc> +q to say nothing is wrong

gavinc: adding {1,8} does not make full checking happen so nothing changes really.
... already have to do further checking so little value in adding to Turtle

<patH> could pont that out in the prose, though.

gavinc: no need to complicate the grammar

(c.f. RD/XML)

<patH> since apparently they missed it.

(c.f. RDF/XML)

<Zakim> gavinc, you wanted to say nothing is wrong

guus: one more week to agree and on agenda next week

sandro: what moves this on?

ericP: ericP and gavinc to email on list

guus: editors to agree
... i18n 190
... " attempting to erase combining marks?"

PROPOSAL: i18n-190 -- No change - follow XML ed 5

RESOLUTION: i18n-190 -- No change - follow XML ed 5

guus: ... 191 ...

ericP: editorial matters
... one change proposed otherwise already covered

RESOLUTION: i18n-191 as per http://lists.w3.org/Archives/Public/public-rdf-wg/2012Oct/0017.html

<ericP> https://github.com/darobin/respec/blob/f7f242642de460e73dcbbb509efb582925b4c6ac/bibref/biblio.js

guus: ... 192 ...

<gkellogg> Issue a pull request against robin's Git repo

gavinc: will do it (ReSpec and reference issue)

RESOLUTION: i18n-192 Update ReSpec for doc.

<gavinc> ACTION Gavin to finish i18n-192

<trackbot> Created ACTION-189 - Finish i18n-192 [on Gavin Carothers - due 2012-10-17].

<scribe> ACTION: gavinc - send an ack to the comments list an di18n for 192 [recorded in http://www.w3.org/2012/10/10-rdf-wg-minutes.html#action01]

<trackbot> Sorry, couldn't find gavinc. You can review and register nicknames at <http://www.w3.org/2011/rdf-wg/track/users>.

<patH> "anack" (noun): a state of total confusion about what to do next.

ericP: 193 - no change
... good to have someone double check.
... the escape processing text

gavinc: 3rd party confirms it works

RESOLUTION: 18n-193 - no change

guus: ... missed 186 ...

gavinc: i18n suggests not just UTF-8
... XML is not a good example.

<gavinc> http://www.w3.org/International/track/issues/186

ericP: agree - no change

PROPOSED: i18n-186 : keep Turtle as UTF-8

andys: a perma-issue on support lists

ivan: deployed practice is UTF-8

RESOLUTION: i18n-186 : keep Turtle as UTF-8

<scribe> ACTION: gavin -- respond for i18n 187 [recorded in http://www.w3.org/2012/10/10-rdf-wg-minutes.html#action02]

<trackbot> Created ACTION-190 - -- respond for i18n 187 [on Gavin Carothers - due 2012-10-17].

ACTION-190: respond for i18n 186

<trackbot> ACTION-190 -- respond for i18n 187 notes added

guus: other comments

gavinc: already done .. no response
... rest minor editorial
... except TTL+HTML section to separate Note

guus: prefer appendix

sandro: use of script tag?

gavinc: problem is Turtle with an end script tag in content.

<gkellogg> Or CDATA?

<gkellogg> :-P

sandro: useful to put into the spec as good practice

gkellogg: JSON-LD does not talk about this - not TTL specific - generic for embedding in <script>
... e.g. n-triples

<patH> Not following all of the detail, but I hear violent agreement and people saying the spec should draw attention to it. So, editorial. Case closed?

sandro: give an escaping example

gavinc: I can add it

guus: propose to be an appendix

<gavinc> RESOLVE: Leave Turtle in HTML in appendix, add stuff about escaping

<gavinc> ACTION Gavin add escaping to Turtle in HTML

<trackbot> Created ACTION-191 - Add escaping to Turtle in HTML [on Gavin Carothers - due 2012-10-17].

<gavinc> ISSUE Turtle in HTML doesn't explain escaping

<gavinc> ISSUE: Turtle in HTML doesn't explain escaping

<trackbot> Created ISSUE-100 - Turtle in HTML doesn't explain escaping ; please complete additional details at http://www.w3.org/2011/rdf-wg/track/issues/100/edit .

guus: What is then the status of Turtle?
... do we need 2LC or can we go to CR?

andys: test suite?

guus: I believe we do not need 2LC
... we are in a position to ask for CR when responds come in.
... test suite
... needed for CR

ericP: test suite -- original submission had some tests.
... in SPARQL - separate test doc - inputs, outputs, there is N-triples as a reference
... in SPARQL - separate test doc . For TTL, inputs, outputs, there is N-triples as a reference

ivan: may be tricky to get an impl report because already widely deployed so not really watching closely. They will update later, not during W3C process

<Zakim> AndyS, you wanted to suggest splitting into syntax (yes/no) and triple output

<gavinc> +q to NOT specify in the tests what to do for invalidity

andys: Jena will run the tests.

<Zakim> gavinc, you wanted to NOT specify in the tests what to do for invalidity

<gavinc> +q to NOT specify in the tests what to do for invalidity

<Zakim> gavinc, you wanted to NOT specify in the tests what to do for invalidity

gavinc: want to be loose to be outside bad TTL and do recovery or graceful non-spec usage

<patH> Whoa. I think having optional recovery is a BAD idea.

<patH> There are lots of published warnings about how this ruins interoperability and leads to nonstandard-creep.

<patH> I vote AGAINST this.

sandro: describe in the conformance section - just need to be clear what the spec says

<patH> PLease record my negative vote.

path - we do not specify what the recovery is -- this is not fortran 4

<ericP> patH, i think you'd be content with the mechanics

<ericP> that is, the negative syntax tests are specifically not turtle, but you don't have to reject them all to be a conformant turtle parser

sandro: timescale? CR start end Oct? For ISWC

<ericP> (having uttered that, i'm not confident that you'd agree with that)

ivan: exit crtieria? test organisation?
... TPAC non-publish period

guus: chairs+team will sort out the timing
... ericP and gavinc to complete actions in a timely fashion.

gavinc: will need help on test suite

sandro: do we approve en mass?

<gkellogg> do you have a reference to the test suite?

ADJOURNED

<Guus> trackbot, end meeting

This is scribe.perl Revision: 1.137  of Date: 2012/09/20 20:19:01  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/??/pchampin/
Succeeded: s/manu: JSON-LD/gkellogg: JSON-LD/
Succeeded: s/manu/gkellogg/
Found ScribeNick: AndyS
Inferring Scribes: AndyS
Default Present: AndyS, tbaker, gavinc, +31.20.598.aaaa, Guus, Sandro, yvesr, pchampin, cgreer, MacTed, ivan, AZ, pfps, Arnaud, +1.443.212.aabb, AlexHall, +1.603.897.aacc, Souri, EricP, gkellogg, zwu2, PatH
Present: AndyS tbaker gavinc +31.20.598.aaaa Guus Sandro yvesr pchampin cgreer MacTed ivan AZ pfps Arnaud +1.443.212.aabb AlexHall +1.603.897.aacc Souri EricP gkellogg zwu2 PatH
Found Date: 10 Oct 2012
Guessing minutes URL: http://www.w3.org/2012/10/10-rdf-wg-minutes.html
People with action items: - ack an gavin gavinc send