<fsasaki> agenda at http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Jul/0048.html

previous minutes

<fsasaki> http://www.w3.org/2012/06/28-mlw-lt-minutes.html

<fsasaki> no comments

actions and events

<fsasaki> ACTION-115 and ACTION-127 - Des is not here, so postponed

<fsasaki> ACTION: Yves to work on targetPointer [recorded in http://www.w3.org/2012/07/05-mlw-lt-minutes.html#action01]

<trackbot> Created ACTION-153 - Work on targetPointer [on Yves Savourel - due 2012-07-12].

General reminder: our plan until September

<fsasaki> http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Jun/0133.html

pub for 26 of July

scribe: does not depend on the call fo consensus

ok with all

The Leipzig Semantic Web day

<fsasaki> http://aksw.org/Events/2012/LeipzigerSemanticWebDay

event overlapping face-to-face meeting

in Prague

should we move F2F 27-28 of sept?

Jirka: not possible (holidays)

Felix: so stay with 25-26 sept.

REMINDER - "call for consensus mails"

<fsasaki> http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Jul/0045.html

some confusion on how it works

an example email 0045 above

most easy to have the definition in the call

editors can just copy that later

Jirak or Felix can put it in the draft in ODD

ISSUES

some we can close

cache, ruby, etc.

can we close the issues? Ruby we have an action item

all agree

<fsasaki> can we close "cache" too?

all agree

issues still open with targetPointer, etc.

<fsasaki> ISSUE-16 Parameter for rules , ISSUE-25 Existing metadata fields in Dublin core, ISSUE-26 Post editing related datacats

daveL: can take post-editing

<fsasaki> ACTION: David to look into ISSUE-26 post editing data categories [recorded in http://www.w3.org/2012/07/05-mlw-lt-minutes.html#action02]

<trackbot> Sorry, amibiguous username (more than one match) - David

<trackbot> Try using a different identifier, such as family name or username (eg. dlewis6, dfilip)

<fsasaki> ACTION: dlewis6 to look into ISSUE-26 post editing data categories [recorded in http://www.w3.org/2012/07/05-mlw-lt-minutes.html#action03]

<trackbot> Created ACTION-154 - Look into ISSUE-26 post editing data categories [on David Lewis - due 2012-07-12].

<scribe> ACTION: Yves to work on parameters [recorded in http://www.w3.org/2012/07/05-mlw-lt-minutes.html#action04]

<trackbot> Created ACTION-155 - Work on parameters [on Yves Savourel - due 2012-07-12].

working mode

<fsasaki> http://www.w3.org/International/multilingualweb/lt/wiki/Main_Page#Draft_documents_and_time_line

any question on the process?

<dF> Zakim would not know as the voice is on GoTo :-)

daveL: is it worth to have people posting whenthey are available over the summer?

<scribe> ACTION: fsasaki to create page for july availability [recorded in http://www.w3.org/2012/07/05-mlw-lt-minutes.html#action05]

<trackbot> Created ACTION-156 - Create page for july availability [on Felix Sasaki - due 2012-07-12].

daveL will be out next week

dF: maybe a doodle page?

<scribe> ACTION: 5 to for july and august [recorded in http://www.w3.org/2012/07/05-mlw-lt-minutes.html#action06]

<trackbot> Sorry, couldn't find user - 5

<fsasaki> action-156: also for august

<trackbot> ACTION-156 Create page for july availability notes added

Call for consensus about externalPlaceholder

<fsasaki> http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Jul/0000.html

made by Shaun

yves replied

any objection or need time?

<dF> good here

<scribe> ACTION: fsasaki to put Shuan text in ODD file [recorded in http://www.w3.org/2012/07/05-mlw-lt-minutes.html#action07]

<trackbot> Created ACTION-157 - Put Shuan text in ODD file [on Felix Sasaki - due 2012-07-12].

Call for consensus about idValue

<fsasaki> http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Jul/0045.html

Yves posted that call

realted issue: id granularity

<dF> good here

daveL: yes, nothing to add. all fine

<scribe> ACTION: jirka to add the idValue to ODD [recorded in http://www.w3.org/2012/07/05-mlw-lt-minutes.html#action08]

<trackbot> Created ACTION-158 - Add the idValue to ODD [on Jirka Kosek - due 2012-07-12].

<fsasaki> http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Jul/att-0045/idValueDescription.pdf

Charter clarification

<fsasaki> http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Jul/0012.html

dificult to drop RDFa

don't have Tadej, maxim here

thanks to Jirka for the place holder. we need Tadej, maxim to fill in.

daveL: in Dublin we talked about it,

felix: outcome with w3C was that we can't drop this

it's part of the charter now

but we can do it only where it make sense

daveL: so we need implementations too

felix: need an algorithm to create the markup
... could adapt what i have, other too. no need for consumers

specialRequirements

felix: long thread about this

<fsasaki> scribe: fsasaki

yves: guiseppe came up with a proposal
... two aspects: length text restriction
... storage in a data base etc.
... and to restrict the type of character that can be used for content
... there was discussion whether that should be different data categories
... latest proposition from Guiseppe seems to be quite workable
... e.g. unit for storage
... but it seems implementable
... there was information about how to specifiy what character would be forbidden
... e.g. using a subset of a regular expression

felix: how does this relate to the XLIFF TC work?

david: frederick said he will submit the draft for discussion before he leaves for vacation

felix: can we also see the draft?

david: I can post it to our mailing list
... overlapping part is just storage and display not forbidden characters
... from the three storage is more stable and makes more sense
... I'd be happy to seperate the three isseus
... I would just support storage size
... latest proposal is implementable

felix: so you would not support restriction of characters?

david: it's not stable at this time
... there is work needed to explore what regex is interoperable etc.
... hard to do in a reasonable time frame
... it is a good idea
... but hard to do
... in time

felix: constraints via xml schema

<Yves_> felix: constrains in xml can be expressed with XSD

felix: or via schematron, path based constraint language

<Yves_> .. or schematron

<Yves_> .. those could be used for forbidden characters

<dF> +1

yves: not sure how that would work
... in extracted data
... you don't have the schema available when you process the document
... it doesn't work during the translation

david: so you need a representation mechanism for roundtripping?

felix: couldn't you have a schematron file for the extracted content

yves: yes
... forbidden characters is not a huge issue
... length constraint is more used

<Jirka> There is CREPDL schema language for restricting character repertoire http://www.asahi-net.or.jp/~eb2m-mrt/crepdl/ns/structure/1.0/index.xml

felix: couldn't that be in schematron too? e.g. via xpath

<Jirka> It's ISO standard

felix: if there is a solution in schematron, shouldn't we use that instead of re-inventing in ITS?

<scribe> ACTION: felix to make examples how length restriction could work in schematron or in xsd [recorded in http://www.w3.org/2012/07/05-mlw-lt-minutes.html#action09]

<trackbot> Created ACTION-159 - Make examples how length restriction could work in schematron or in xsd [on Felix Sasaki - due 2012-07-12].

action-159: take also CREPDL language into account

<trackbot> ACTION-159 Make examples how length restriction could work in schematron or in xsd notes added

<Yves_> Felix: seem storage aspect has consensus

<Yves_> Yves: yes

<Yves_> df: most useful and stable

<Yves_> felix: should it be its own data cat?

<dF> +1

<Yves_> +1

<Yves_> ACTION: yves to summarise this to giussepe? [recorded in http://www.w3.org/2012/07/05-mlw-lt-minutes.html#action10]

<trackbot> Created ACTION-160 - Summarise this to giussepe? [on Yves Savourel - due 2012-07-12].

Test suite design

<Yves_> action-168: ie. separation of length vs forbidden chars

<trackbot> Sorry... adding notes to ACTION-168 failed, please let sysreq know about it

<Yves_> daveL: will take this taks

<Yves_> ... looking at felix scripts

<Yves_> .. probaly we need more than the w3c framework

<Yves_> .. but we can use it

<Yves_> .. and add to it

<Yves_> .. it's a way to structure test repository

<Yves_> felix: yes. I'll continue to make use of the ITS 1.0 tests

Auto language processing rule

http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Jul/0010.html

<Yves_> More about a transliterate rule

<Yves_> anyone want to implement this?

yves: shaun has implemented something, I think
... including also machine translation

[Correction after the meeting about above statement, see http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Jul/0087.html].

<Yves_> .. how does it relate to BCP47

<Yves_> we should have same value

<Yves_> felix: BCP47 has additional decsriptor

yves: problem with bcp 47 extension, not used very widely
... haven't seen any real use case for that

david: agree with felix that we should try to be consistent

<Yves_> df: yes we should try to be consistent

<Yves_> .. should be mappable

<Yves_> ACTION: felix to talk to shaun about BCP47 compatibility [recorded in http://www.w3.org/2012/07/05-mlw-lt-minutes.html#action11]

<trackbot> Created ACTION-161 - Talk to shaun about BCP47 compatibility [on Felix Sasaki - due 2012-07-12].

<Yves_> daveL: broader question. should we look at language tags, etc. sourcelang, etc.

<Yves_> .. how those things work with BCP47?

<Yves_> issue would be: compatibility with BCP47 extensions

<Yves_> felix: any thoughts on this?

<dF> I agree that it is a more general issue

readiness

<Yves_> felix: long thread about that

<Yves_> daveL: summary:

<Yves_> .. from Dublin meeting, agreement seemed to say that it's about telling a file is ready to be processed

<Yves_> .. with some extra features

<Yves_> .. but also there are more attributes

<Yves_> .. closely related to translation jobs

<Yves_> .. they could be optional

<Yves_> .. are the more general than just for readiness?

<Yves_> .. also overlap with TS116049(?)

<Yves_> .. to specify translation job status, etc.

<Yves_> .. so should this be just in readiness or more general as translation parameters

<Yves_> .. it seems we were close to say we should ISO and Linport to do this

<Yves_> .. should it be seprated from readiness?

<Yves_> felix: other thoughts?

<Yves_> dF: related to business relations between parties

<Yves_> .. not sure all this is ready for standardization

<Yves_> felix: maybe like for specialRequirements

<Yves_> .. some parts could be extracted as useful and easy to implement

<Yves_> .. some atomic parts

<Yves_> .. a big beats for now.

<Yves_> .. but we don't want to kill those ideas either

<Yves_> daveL: pedro had some specific use cases.

<Yves_> .. maybe this could be addressing common use cases

<Yves_> .. i can follow up on that aspect

<Yves_> felix: could be also addition to the translate data cat

<Yves_> .. relates to conformance too

<Yves_> .. if we add to translate then implementation for translate would have to implement this too

<dF> Pedro is trying to speak, I told him to q here

<Yves_> daveL: readiness is more for real-time translation maybe

<Yves_> Pedro: we can provide requirements from real cases

<Yves_> .. we may have no time

<Yves_> .. our requirements may not be covered by readiness

<Yves_> .. priority, who does what, when, etc.

<Yves_> .. this is in the HTML page

<Yves_> .. and we can put it on the warpper

<Yves_> .. we have customers that need those metadata

<Yves_> felix: pedro, could you move on with those customer

<Yves_> .. and keep the WG in the loop

<Yves_> .. and provide example (like HTML code)

<Yves_> .. then we can put that either in the standard or as best practices

<Yves_> .. if we can't put it in 2.0

<Yves_> pedro: yes

<Yves_> .. we have to do anyway

<Yves_> .. our requirements apply to our customers

<Yves_> felix: can you present a partial showcase in september F2F?

<Yves_> pedro: for CMS of package 3 yes

<Yves_> .. for package 4, maybe some of it

<Yves_> .. we are implementing the solution for our customer

<philr> leave

<scribe> ACTION: pedro to present a readiness related partial showcase in september f2f, to see what is ready for standardization or what will be a best practice [recorded in http://www.w3.org/2012/07/05-mlw-lt-minutes.html#action12]

<trackbot> Created ACTION-162 - Present a readiness related partial showcase in september f2f, to see what is ready for standardization or what will be a best practice [on Pedro Luis Díez Orzas - due 2012-07-12].

<Yves_> pedro: yes. we'll have to talk with Cocomore too

<Yves_> felix: we do want to hear your requirements

<Yves_> .. without worrying about what ends up in standard

<Yves_> .. please continue the excitment

<Yves_> pedro: hope to have partial implementation in september

domain discussion

<Yves_> felix: for this we need feedback from declan and Thomas

<Yves_> .. for the mapping

<Yves_> .. proposal on the table

http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#domain-implementation

domainMapping="automotive auto, medical medicine, 'criminal law' law, 'property law' law"/>

<Yves_> felix: list of strangs pairs

<Yves_> .. should we use URI instead of strings?

<Yves_> daveL: URI or strings

<Yves_> using URI doesn't break the definition

<Yves_> Felix: ok then

<Yves_> daveL: more a usage case

<Yves_> felix: then what the implementation need to do?

<Yves_> .. to be conformant

<Yves_> daveL: you just matching a string to a selection

<Yves_> .. in that case you have the option to use a more structure way

<Yves_> .. of doing this

<Yves_> .. if there are overlap you have to have a solution

<Yves_> .. don't think it change sthe definition

<Yves_> felix: URI as a diferent type of string

<Yves_> daveL: yes

<Yves_> .. no fetching of the URI

<scribe> ACTION: felix to write an example of "string as URI" for domain [recorded in http://www.w3.org/2012/07/05-mlw-lt-minutes.html#action13]

<trackbot> Created ACTION-163 - Write an example of "string as URI" for domain [on Felix Sasaki - due 2012-07-12].

aob

<Yves_> dF: did we had a quorum?

<Yves_> felix: we could decide the rules

<Yves_> dF: qorum easier if missing people are put on bad standing

<Yves_> felix: we'll have a doodle pool to see how many people we can get during the summer

<Yves_> daveL: call for the LRC conference (20-21 sept)

<Yves_> .. should coordinate on submission

<Yves_> .. deadline is end of july

<Yves_> .. CNGL wa splanning demo there

<Yves_> .. some demo should be relevant for MLW-LT

lrc conference is on the wiki at https://www.w3.org/International/multilingualweb/lt/wiki/EventSchedule

<Yves_> .. showcase is on the day before

<Yves_> felix: no plan to submit yet

<Yves_> .. feel free to submit

<Yves_> .. notify about the showcase by email