SV_MEETING_TITLE -- 28 Nov 2012

<Yves_> I'm waiting for the GTM session to start

<Yves_> Do you know if Dave is joining us?

<Yves_> Or if anyone else can start the GTM session?

<Yves_> Hi Felix, we are waiting for the GTM to start

<Yves_> Ok I'll start a different one then.

<Yves_> https://www1.gotomeeting.com/join/745412769

<Yves_> Scribe: Yves_

LQ Rating/Precis

Phil: let's drop global
... for score I think Des was asking for 0 to 100
... and make them decimal

Y: what about threshold?

Jirka: can do a condition
... or create two threshold

Y: do we have example for the two types of score at the same time?

Jirka: currently it's not allowed

Phil: not one of my scenarios

Jirka: we can have two thresholds and keep the scores exclusive

<philr> Yes, two thresholds is fine with me.

Jirka: that way we can have a change if needed

<daveL> scribe: daveL

loc quality precis data category

yves: summarising, based on discussion with phil, arle
... keeping score and vote mutually exclusive and give them separate thresholds
... will update these offline
... later today

<Yves_> Scribe: yves_

Dave: no issue in ussing both provenance and LQ rating

<philr> Way back locQualityIssue/Rating had an agent but this wa dropped so I would want to use Translation Agent Provenance

<scribe> Scribe: Dave

another change for LQ Rating: use decimal from 0 to 100 for score.

<daveL> http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#its-tool-annotation

section 8 review

<daveL> ... christian has also sent a bunch of topic we will review later

<daveL> Yves: is a 'note' in a normative section itself nomartive

<daveL> jirka: no

<daveL> dave: do we need a table label?

<daveL> yves: good point, but how to do this?

<daveL> ACTION: felix to check if the data category table can be given a numbered caption in ODD [recorded in http://www.w3.org/2012/11/28-mlw-lt-minutes.html#action01]

<trackbot> Created ACTION-333 - Check if the data category table can be given a numbered caption in ODD [on Felix Sasaki - due 2012-12-05].

<daveL> karl: points out some issues in example 28 following the table

<daveL> ... with notes in the rule, but not in the content

<daveL> Fixed by introducing a type element in exmaple 28

translate

<daveL> yves: seems Ok

localsiaiton note

<daveL> should values be in code element or quote element.

<daveL> There seem to be an issue with quote element missing a space in the generated document.

<daveL> also seem to be val in the 'val' element in some places

<daveL> ACTION: yves to check consistency of use of quote and val [recorded in http://www.w3.org/2012/11/28-mlw-lt-minutes.html#action02]

<trackbot> Created ACTION-334 - Check consistency of use of quote and val [on Yves Savourel - due 2012-12-05].

<daveL> Yves: we need to know which is the right one to use and also the cause of this white space issue

<daveL> yves: will examples be validated

<daveL> ACTION: jirka to run examples through ITS schema validator [recorded in http://www.w3.org/2012/11/28-mlw-lt-minutes.html#action03]

<trackbot> Created ACTION-335 - Run examples through ITS schema validator [on Jirka Kosek - due 2012-12-05].

<daveL> Yves; another instance of missing white space for use of val element for description in locNoteType

terminology

<daveL> yves: should we have confidence score on this

<daveL> .. for global

<daveL> ... but it is there fore local

<daveL> ... so the issue is that we can't use confidence with attribute text

<daveL> ACTION: daveL check up on concensus on not using confidence on term global [recorded in http://www.w3.org/2012/11/28-mlw-lt-minutes.html#action04]

<trackbot> Created ACTION-336 - Check up on concensus on not using confidence on term global [on David Lewis - due 2012-12-05].

<daveL> yves: concerned 39 is a bit contrved, but is there from 1.0

<daveL> yves; should we state in LOCAL definition that 'term' is required

<daveL> ... possibly not since this was OK for 1.0 even though we tend to say this in the new 2.0 data categories

<daveL> ... so leave this as is

directionality

<daveL> dF; checked discussion on terminology, but its not clear there was support for confidence

<daveL> ACTION: marcis to confirm that the current local (but not global) is supported - as this is not clear from discussion [recorded in http://www.w3.org/2012/11/28-mlw-lt-minutes.html#action05]

<trackbot> Sorry, couldn't find marcis. You can review and register nicknames at <http://www.w3.org/International/multilingualweb/lt/track/users>.

<daveL> ACTION: daveL to check with marcis to confirm that the current local (but not global) is supported - as this is not clear from discussion [recorded in http://www.w3.org/2012/11/28-mlw-lt-minutes.html#action06]

<trackbot> Created ACTION-337 - Check with marcis to confirm that the current local (but not global) is supported - as this is not clear from discussion [on David Lewis - due 2012-12-05].

<daveL> yves: for directionality, we won't look in detail as note indicates it is a feature are risk, and the same for ruby

<daveL> jirka: there are some missing attribute in HTML5, but as this is unstable its difficult to determine this now

<daveL> ... recommend leaving it as it is and revisit against updates happending in HTML during the spec process

language information data category

<daveL> yves: there is an editor note on adding example for HTML5 and lang

<daveL> ... added some wording on use of lang in HTML

<daveL> ... and make some paragrpahs specific to XML

<daveL> ... but don't say anything about HTML since lang is mandated

<daveL> Yves; dropped the note about having a HTML example since it didn't seem very necessary

<daveL> karl; is the editor note looking for explanation of why we don't support HTML

<daveL> yves: Ok, add note explaining that lang is mandatory in HTML so it doesn't need to be supported in this data categories

<daveL> topic; elements within text

<daveL> yves: we did add a local mechanism that wasn't there before in 1.0

<daveL> ... seems Ok

domain

<daveL> yves: first of the new ones so needs careful examination

<daveL> Made explanatory text for example 54 clearer

<daveL> added 'val' elements that were missing

<daveL> topic;disambiguation

<daveL> ACTION: yves to add globally the disambigsource attribute [recorded in http://www.w3.org/2012/11/28-mlw-lt-minutes.html#action07]

<trackbot> Created ACTION-338 - Add globally the disambigsource attribute [on Yves Savourel - due 2012-12-05].

<daveL> removed ed note referring to need for test case. this was noted in general for all MUSTS in the last editing call

<daveL> daveL: do we need val elements for 0 and 1

<daveL> Yves: no since we are actually refernecing a range of values, plus it would need to be changed in a lot of places.

<daveL> fixed example 56 and 57

<daveL> topic; locale filter

<daveL> Yves: it was not clear why the line "Implementations MUST NOT combine lists of language ranges from multiple rules or local attributes." is there?

<daveL> ... since it would be addressed anyway by overriding

provenance

<daveL> remove notes column from table as it was empty

<daveL> changes refernce of external provenance data to 'information'

<daveL> improved wording on example 61 text

<daveL> ACTION: daveL to check with flix about provRef now only having one IRI rather than a list of them [recorded in http://www.w3.org/2012/11/28-mlw-lt-minutes.html#action08]

<trackbot> Created ACTION-339 - Check with flix about provRef now only having one IRI rather than a list of them [on David Lewis - due 2012-12-05].

<daveL> daveL: had an action to provide an XLIFF mapping version of the global ref to standoff data

<daveL> Yves: but this probably doesn't make sense right now since this would be for XLIFF2.0 which is not yet stable

<daveL> daveL; agreed, this should be something to pick up later in best practice

- DRAFT -

SV_MEETING_TITLE

28 Nov 2012

Attendees

Contents