<daveL> previous minutes;

<daveL> http://www.w3.org/2012/08/09-mlw-lt-minutes.html

agenda review

Agenda approved

daveL: are there comments on previous minutes?

no comments

daveL: Felix, what about the interaction with the HTML5 WG?

fsasaki: I gave background last time, are you asking what are the next steps?

daveL: I guess our categories should be digestible for HTML5 WG

fsasaki: each category will need an HTML5 example
... I have a long standing AI to make the draft more HTML5 friendly, I am not sure how to procede, addind HTML5 examples is obvious
... I posted today a quality example with javascript
... it demonstrates ITS interoperability with a browser

philr: This scenario seems to resonate with us. We should be able to make a lightweight implementation

fsasaki: It would be good to have it open source

philr: Let me consider it..

fsasaki: We need uptake from the HTML community

daveL: This is also interesting for the CMS-L10n platform
... Felix, do you think it is selling category by category or persuade them that ITS is generally a good idea

Felix: We need to specify lists
... What I would push internally in W3C, have examples, but do not have HTML5 WG redo every category

<scribe> ACTION: Felix to folow up with W3C management on the approach [recorded in http://www.w3.org/2012/08/23-mlw-lt-minutes.html#action01]

<trackbot> Created ACTION-199 - Folow up with W3C management on the approach [on Felix Sasaki - due 2012-08-30].

<daveL> http://microformats.org/wiki/existing-rel-values#HTML5_link_type_extensions

daveL; wouldn't be the microdata registration approach useful?

fsasaki: should not be difficult

<fsasaki> http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#datacategories-defaults-etc

<scribe> ACTION: Felix to create a new column for HTML5 in the above table [recorded in http://www.w3.org/2012/08/23-mlw-lt-minutes.html#action02]

<trackbot> Created ACTION-200 - Create a new column for HTML5 in the above table [on Felix Sasaki - due 2012-08-30].

Issue 42 - inheritance of agents info related to confidence type of data across categories

fsasaki: I did not get to this action item, recalling what was discussed two weeks ago
... the overlap might not be as big

tadej: In my mail I exposed that mtConfidence and text analytics confidence have different targets
... selector for a consolidated data cetagory?

<fsasaki> scribe: fsasaki

david: I said: either the info goes to provenance, or it goes to categories
... the real issue arle + felix have seen:
... sometimes the information is in conflict
... you cannot deal with it with standard inheritance
... for mtconfidence, that is not an issue
... it is about content
... agree with what tadej said:
... the target of text analytices confidence
... and mtconfidence are different

ta confidence is metadata about metadata

scribe: mtconfidence is metadata about data
... hence there is no inheriteance conflict with mtconfidence
... I concluded for mtconfidence that I would wait for felix to propose general inheritance formulation
... as part of issue-42
... issue-42 has also other parts. that is conflicting information
... mtconfidence doesn't has dependence on that

<dF> fsasaki: yves separated a general confidence

<dF> .. no more inheritance issue with quality

<scribe> scribe: dF

.daveL: global and local for mtConfidence

daveL: it might be good to stick with a specific translation agent, not introduce an mt Confidence agent.

<scribe> ACTION: daveL to produce a strawman how to deal with mtConfidence using the general translation agent [recorded in http://www.w3.org/2012/08/23-mlw-lt-minutes.html#action03]

<trackbot> Created ACTION-201 - Produce a strawman how to deal with mtConfidence using the general translation agent [on David Lewis - due 2012-08-30].

Joerg: what is the import, proces or data?

daveL: we need to tie the info to the producing software as agent

Joerg: I got you

fsasaki: should be done by next call

Data Category Specifications

daveL: Micha, do you wnat to give update on splitting the special requirement into three?

Micha: not sure about the purpose
... display size, storage size, and forbidden characters

<scribe> ACTION: Micha to follow up on display size data category [recorded in http://www.w3.org/2012/08/23-mlw-lt-minutes.html#action04]

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

<scribe> ACTION: Micha to follow up on storage size data category [recorded in http://www.w3.org/2012/08/23-mlw-lt-minutes.html#action05]

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

dF: Would give chance to Pedro and/or Arle to pick up the forbidden character data category

fsasaki: I suggest to drop forbidden character if the solution is not found by the end of the next week

<scribe> ACTION: Micha: To post to the list that owner is needed for the forbidden character otherwise it will be dropped [recorded in http://www.w3.org/2012/08/23-mlw-lt-minutes.html#action06]

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

Micha: is it characters or bytes
... ?

Publication

<fsasaki> Update change log. Update table with overview of data categories. Change html5 examples (are borken, e.g. empty "span" element). Check editor notes for new data categories. Add placeholder for mtconfidence. add "html5 column" to sec. 6.1

fsasaki: I need some time to agree on it, so that it can go out next week

<fsasaki> http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html

fsasaki: we will need placehoder for mtConfidence
... we already have some other placeholders
... I propose to publish by Thursday

daveL: what about quality?

<fsasaki> http://www.w3.org/International/multilingualweb/lt/wiki/LocQuality#Localization_Quality_Precis

fsasaki: part of quality already there
... with warning that it is in flux

daveL: are we happy to publish on that basis? Objections?

tadej: sounds good

fsasaki: Thanks

<fsasaki> ACTION: felix to publish next draft by next thursday, with the edits described above [recorded in http://www.w3.org/2012/08/23-mlw-lt-minutes.html#action07]

<trackbot> Created ACTION-202 - Publish next draft by next thursday, with the edits described above [on Felix Sasaki - due 2012-08-30].

data categories

<daveL> scribe: daveL

dF: mtconfidence need some minor editorial, but also dependencie to issue 42
... on inheritence info which felix is looking at
... and strawman on translationAgent from DaveL
... so won't be mature till these issues are closed

<joerg> What about naming conventions: Localization Quality Precis (mixed en-US + fr-FR) vs. Localization Quality Precise (en-US)?

<dF> daveL: now named entity, Tadej, do you wnat to comment?

<dF> Tadej: RDF people do no see the need to specify disambiguation source

<dF> .. but we need to include it, trying to look for a compromise

<dF> tadej: maybe realxing requirements for a pointer, inlcude options for local identifiers, or RDF URIs

<dF> .. this should be a good way to reconcile

<dF> .. this should be ready by tomorrow

<dF> daveL: translation agent and translation revision agent

<dF> .. should be done in a day or two

<dF> daveL: Provenance

<dF> .. I need to add pointer

<dF> .. still need to look at the provenance WG work, but this will need to go into non-normative

<dF> .. Provenance WG are making final call in September, they contacted us

<dF> .. provenance URL, we shoudl probaly require URI

<dF> .. I will write to them to that effect

<dF> Tpic: Test suite

Test suite

<fsasaki_> see test suite discussion at http://lists.w3.org/Archives/Public/public-multilingualweb-lt-tests/

<dF> tahnks, felix

<dF> daveL: any suggestions to the test suite development? Formats?

<dF> Declan: more formally

<dF> daveL: there is a technical discussion on the list

<dF> .. we need to keep up with the spec

<dF> we need help with use case development, not always the sample files conversion is sufficiently comprehensive

<dF> daveL: meeting adjourned