[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [xliff] Groups - Event "XLIFF TC Call" added
I Administration (0:00 - 0:10)
A. Roll call
Rollcall taken 12 voters out of 16 (quorum reached), Yves may join later [he did] (13).
[Attendance recorded:] Victor Amaya, Helena Chapman, Shirley Cody, Tom Comerford, Fredrik Estreen, DavidF Filip, Ryan King, Kevin O'Donnell, Yves Savourel, Bryan Schnabel, Joachim Schurig, Uwe Stahlschmidt, DavidW Walters.
B. Approve previous meeting minutes, 13 February 2013 https://lists.oasis-open.org/archives/xliff/201303/msg00005.html
Joachim seconds
No objections
The above become meeting minutes of record.
C. Charter clarification proposed and seconded. Ready for Roll Call ballot https://lists.oasis-open.org/archives/xliff/201302/msg00020.html
Bryan: There was lot of discussion
.. the proposal for ballot was seconded by Bryan on mailing list.
dF: Helena asked about confoirmance test suite after this was seconded, I answered
.. anyone wish to weigh in?
Bryan: [reads proposal as suggested by dF on the mailing list][ballot starts]
dF: [takes votes from eligible voters..]
[11 yes, Helena no, motion carries]Bryan: The charter will be updated in the main TC page.
[AI on Bryan, to update the main TC webpage with the clarified charter]
II XLIFF 2.0 (0:10 - 0:45)
A. Review Approved Features in WIKI (https://wiki.oasis-open.org/xliff/XLIFF2.0/FeatureTracking ) (0:10 - 0:20)
1. Take a quick look at statuses
Bryan: [reviews statuses of features as per wiki https://wiki.oasis-open.org/xliff/XLIFF2.0/FeatureTracking]
2. See if any features need “live” input from TC
D15 Mime type for XLIFF
dF: Will move forward once we have CD
[AI on dF, edit D15 accordingly]
C3 native support for ITS
Bryan: currently owned by Yves and dF, do you want to explain?
dF: The dedicated module/profile is slated for XLIFF 2.1, now we are looking at native XLIFF capabilities to represent ITS. Now we are looking at basic semantic match between the standards, ITS group works on the mapping here http://www.w3.org/International/multilingualweb/lt/wiki/XLIFF_Mapping
There are two pending issues across features:
1) Translation candidates <match> is not extensible and the current set of attributes does not allow for mapping MT Confidence scores, or any other quality related scores, it has just similarity..
[See recent discussion on <match> attributes
https://www.oasis-open.org/apps/org/workgroup/xliff/email/archives/201303/msg00011.html]
dF: I appreciate that there is now standard for similarity or quality scores, but this shouldn't be an argument for dropping them, rather to allow to carry them with tools info to allow for meaingful comparisons of values.
In the above e-mail I proposed two possible approaches, one simple - having just one general suitability attribute or having three along the lines of Yves description in the wiki https://wiki.oasis-open.org/xliff/XLIFF2.0/Feature/Translation%20Proposals. This basically means to replace XLIFF 1.2 match-quality attribute from <alt-trans> with a set of three attributes, 1) similarity [between sources] as currently specified, 2) quality [of target] that could be used to carry MT Confidence or other quality score info mapped onto a 0-100 score, 3) a composite suitability score that is intended for ordering [Yves: and legacy mapping] purposes.
.. It seems important to note that XLIFF TC is not in business of defining any of the algorithms that are needed to produce meaningful and comparable score, yet it must not fail to provide sufficiently expressive set of attributes to carry around state of the art metadata. Otherwise XLIFF local leveraging mechanism will be ignored by toolmakers.
Joachim: We should go for the more complex solution
dF: I will follow up on the above discussion
[AI on dF, finalize a 3 attributes based proposal and propose it on the mailing list]
2) Markers are not extensible. This is one of the two pending inline issues that were identified at he Inline SC conclusion.
Yves and I have argued for making markers extensible here
https://lists.oasis-open.org/archives/xliff/201210/msg00101.html
here http://markmail.org/message/4wckgif4iuq3vato and elsewhere including the inline SC list.
Yves and Fredrik agreed that the markers should be extensible.
Fredrik: I agreed with placing ITS modules attributes on markers, but not with general extensibility.
dF: I see. The issue is that the module is only planned for XLIFF 2.1 and there would be no vehicle for placing many ITS metadata items legally until the module makes it into the official spec. This seems unnatural. It is kind of OK with ITS as its elements will be allowed. But I think that markers should be extensible to allow for organic developments of inline annotation modules. How could the modules possibly develop in practice if it was only possible to extend by approved modules' namespaces?
Bryan: The argument for modules development from extension resonates with me.
dF: I guess this was now extenively discussed and will formally propose extensibility of markers on the mailing list.
[AI on dF, propose ballot for markers extensibility on mailing list]
J33 Glossaries
dF: This was marked as DONE, but reference language should be added
Joachim: Yes, this in my bucket, will update
[AI on Joachim, to add reference language functionality to Glossaries module]
B26 Add optional format attribute for quick at-a-glance review
dF: This is marked as done, yet it does not appear in the spec, as it has not yet been included. In general if you finish a module, either include it via xi:include mechanism in the xliff20.xml on svn or contact me on inclusion.
.. Word of warning, while working on modules and features, please note that ids across the spec must be unique
Ryan: It does not help to put files into subfolders
Fredrik: It should be enough to make the ids unique. Anyway, I will test print before uploading my includes onto svn. It is surprising how many files must be touched to keep modular.
dF: Yes, it should be just the ids, the thing is that the issues only surface when you attempt to include and merge..
R43 Change track module
Ryan: Will be done by next meeting.
[AI on Ryan, to finalize module]
F35 Length and Size restrictions
Fredrik: Not by today but say Thursday
[AI on Fredrik, to finalize module]
R36 Resource Data Module
Ryan: The spec was just printed with it
dF: Still finalizing some include issues.
R37 Validations Module
Ryan: Will be done by next meeting.
[AI on Ryan, to finalize module]
E. Charter
[see above ballot]
III Sub Committee Reports (0:45 - 0:55)
dF: one good news, one bad news
1) Microsoft donates space in central London http://www.microsoft.com/en-gb/about/offices/london.aspx for XLIFF TC F2F meeting, Thanks, Kevin, for making this happen. This will be a very important meeting. If CD is pushed on time, we will be handling public comments in this meeting.
2) 4th XLIFF Symposium CFP: http://www.localizationworld.com/lwlon2013/feisgiltt/xliffcall.html attracted only a few proposals so far. Let us hope that all are working on their proposals. The CFP is open till early April.
PC members who wish to help with Symposium organization, please join P&L SC
https://www.oasis-open.org/apps/org/workgroup/xliff-promotion/
Even if you don't please approach possible speakers you'd like to see speak with CFP details, propose talks yourself. Kevin's idea that we embraced is to keep it production focused this year round and also 1.2 to 2.0 transition oriented.
IV Current Business (0:55 -->)
V New Business
Agenda
I Administration (0:00 - 0:10)
A. Roll call
B. Approve previous meeting minutes, 13 February 2013 https://lists.oasis-open.org/archives/xliff/201303/msg00005.html
C. Charter clarification proposed and seconded. Ready for Roll Call ballot https://lists.oasis-open.org/archives/xliff/201302/msg00020.html
II XLIFF 2.0 (0:10 - 0:45)
A. Review Approved Features in WIKI (https://wiki.oasis-open.org/xliff/XLIFF2.0/FeatureTracking ) (0:10 - 0:20)
1. Take a quick look at statuses
2. See if any features need “live” input from TC
B. Feature Spotlight
1. How we view XLIFF 2.0 as entirely as an exchange format (Fredrik)a. If we expect XLIFF 2.0 to be a pure exchange format or if we also expect it to be usable for direct native processing? Many features will be incompatible, for example segmentation changes. ..that would not be an issue if we expect tools to convert into own format and export result of processing back to XLIFF. But could be an issue if processing is done directly on XLIFF
2. Match type and subtype (Yves, Ryan) https://lists.oasis-open.org/archives/xliff/201301/msg00025.html
C. Mailing List Summary
1. Resource Data Module (Ryan) https://lists.oasis-open.org/archives/xliff/201301/msg00031.htmla. DavidW comment on Resource Data Module https://lists.oasis-open.org/archives/xliff/201302/msg00009.html
2. Change Track Module (Ryan) https://lists.oasis-open.org/archives/xliff/201301/msg00032.html
3. dF sanity check of translation candidates module (DavidF, Yves) https://lists.oasis-open.org/archives/xliff/201303/msg00003.html
D. Conformance criteria
E. Charter (see above ballot)
III Sub Committee Reports (0:45 - 0:55)
IV Current Business (0:55 -->)
V New Business
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]