Minutes from XLIFF TC Public Info Sessions on June 7 2016

Hi all,

[this also went as BCC to Non-OASIS attendees of the info sessions..

Non-OASIS attendees will not be able to reply to the list, you can send your reaction through the XLIFF TC commenting facility

https://www.oasis-open.org/committees/comments/index.php?wg_abbrev=xliff 

or

FEISGILTT 2016 <> (on CC)]

as promised please find copy pasted and attached as docx the meeting minutes of Dublin OASIS XLIFF TC Info Sessions

I am also exposing a pdf version of the same on the FEISGILTT webpage

http://locworld.com/wp-content/uploads/2015/11/FEISGILTT-2016-XLIFF-Public-Info-sessions.pdf

I invite every attendee to post more detail or amendments to the draft minutes I posted.

The plan is to approve these minutes as minutes of record in the next XLIFF TC conference call on 2nd August 2016.

Public details of the call to appear here by next Monday:

https://www.oasis-open.org/committees/calendar.php

Finally, these are just the XLIFF TC related info sessions, the XLIFF OMOS TC list is addressed only FYI (on CC). Earlier this month I sent the OMOS related minutes, CCing the XLIFF TC. The OMOS minutes were approved in the meantime.

Cheers and thanks for your attention

dF

Dr. David Filip
===========

OASIS XLIFF OMOS TC Chair

OASIS XLIFF TC Secretary, Editor, Liaison Officer

Spokes Research Fellow

ADAPT Centre

KDEG, Trinity College Dublin

Mobile: +420-777-218-122

FEISGILTT 2016 XLIFF Public Info sessions

Draft meeting
minutes                Tue June 7 2016

Participants:

[participation in particular sessions was
not tracked, this list is based on FEISGILTT 2016 participation]

XLIFF TC members

Chair

N/A

Secretary

David Filip (ADAPT Centre)

Members

Ryan King (Microsoft), Kevin O’Donnell
(Microsoft), Soroush Saadatfar (ADAPT Centre @ LRC @ University of Limerick), Felix
Sasaki (Individual), Phil Ritchie (Vistatec), 

Public

Victor Alves (Individual), Jan Bareš (Moravia), George
Bina (Syncro Soft), Konrad Chmielewski (XTRF), David Clarke (Welocalize),
Mariza Flores (Thousand Words Translations), Andreas Galambos (Transmission Übersetzungne), Andrew
Gibbons (Welocalize), Ján
Husarčík (Moravia), Łukasz Kaleta (XTRF), Gary Lefman
(Cisco), David Lewis (ADAPT Centre),
Ian Morris (Individual), Dimitris Orfanoudakis (TAUS), Achraf Oumghar
(Lionbridge), Felix Sasaki (DFKI), Uta Seewald-Heeg (Anhalt University of
Applied Sciences), Richard Sikes (Content Rules), Vinod Sudharshan (TAUS), Chase Tingley (Spartan Software), Angelika Zerfaß (zaac).

 

XLIFF TC Public Info Sessions started at 10.15am
WEST on June 7th 2016.

XLIFF 2.1 Work in Progress: ITS
Module

Chair: dF

Panelists: XLIFF TC membership

 

This was introduced by a presentation by
Soroush and dF on what it means for ITS to become a module rather than an
extension as of XLIFF 2.1 http://lanyrd.com/2016/feisgiltt/sfbtzf/

During the info session, the TC members
informed the public on what is the current state of the ITS nodule development.
The audience and TC members expressed interest mainly in the Text Analytics and
Provenance metadata categories

XLIFF 2.1 Work in Progress:
Module Fixes

Chair:
dF

Panelists:
XLIFF TC membership

dF inquired with Phil about the state of
the Change Tracking module fix that is being led by Phil. Phil said there was
unfortunately no progress on that so far. dF said that the module fix should be
done by the end of August, should it be included in the 2.1 release as the ITS
and Advanced Validation features are near completion and the 1st
public review shouldn’t start later than in August 2016.

Other module fixes should include
explicitly allowing the metadata module on most of the other modules. This is
now explicitly allowed only on the Translation candidates module.

 

XLIFF 2.1 Work in Progress: Advanced
Validation

Chair: dF

Reporters: Soroush Saadatfar (ADAPT Centre @ LRC @
University of Limerick) & Felix Sasaki (DFKI)

 

Soroush and Felix demonstrated some of the
Advanced XLIFF Constraints that are not expressible using XML Schema.

dF explained that the Advanced Constraints
had be validated by custom code in the available XLIFF 2.0 validators such as
OKAPI Lynx http://okapi-lynx.appspot.com/validation,
Xmarker http://xmarker.com/node/13, or
the Microsoft XLIFF Object Model https://github.com/Microsoft/XLIFF2-Object-Model

XLIFF TC Public Info Sessions broke for
lunch at 12:30 WEST.

XLIFF TC Public Info Sessions resumed at
16:00 WEST [after lunch and XLIFF OMOS Info Sessions].

 

XLIFF 2.2 Requirements Gathering

Chair: dF

Reporter: Chase Tingley

Panel: All

Plurals?

Chase reported on a plurals support
discussion that took place on the OKAPI developer forum.

dF reminded Chase and others that there was
a discussion about plurals support in the early stages of XLIFF 2.0
development. David Walters from IBM said then that we don’t need to support the
old limited notion of plurals and that now conditional plural structures are
part of CLDR data.

dF also reported on a ULI discussion that
basically agreed that the proper I18n approach is to have different strings
based on the CLDR locale specifics.

Chase and dF discussed how this could be
handled during the XLIFF roundtrip either on group or subunit level. The
advantage of the subunit handling would be that the modifiers would be able to
create the conditional strings w/o violating XLIFF structural Constraints. The
subunit conditional segments should be protected from re-segmenting so that
agents not understanding the plural handling module couldn’t break the feature.

In either case new module metadata would
need to be developed to describe the conditional multiplicity groups and placed
either on the unit or the group extension point.

dF suggested that this should be proposed as
a 2.2 feature by roughly end of October 2016 and that chase better joined the
XLIFF TC so suggest that. Otherwise, this would need to be suggested over the
commenting facility and Chase couldn’t drive the feature.

Segmentation information?

Including segmentation rule information in
XLIFF 2.2 was also discussed. That was suggested by ULI representatives in
Berlin 2015. However, no one proposed the feature to XLIFF TC formally. dF
suggested that this could be done via inclusion of SRX as module data. Chase
argued that not all rule were SRX expressible. A short discussion revealed that
what Chase meant that statistically driven methods of segmentation cannot be
described with rules. dF suggested that rule based segmentation should be
always encoded as SRX, whereas there should be an option record just provenance
of the tool if the method wasn’t rules driven.

dF also informed that the ULI stakeholders
would be in favor of using the BCP47 t-extension to roundtrip this data via
XLIFF. Nevertheless, dF recalled a discussion with Mark Davis (video presence)
during ITS 2.0 Requirements gathering in Dublin 2012 http://videolectures.net/w3cworkshop2012_davis_data/,
where Davis explicitly said that the t and u extensions where not suitable for
structured XML content that should store this metadata using its own markup. dF
said that one of the main issues of the extension based segmentation
information that it was not stateful and it wouldn’t be possible to tell if the
method was applied, had to be applied, if it was preserved etc.

This could also be formally submitted to XLIFF
TC as a feature for XLIFF 2.2 consideration.

Register Information?

Ján Husračík from Moravia suggested that language register variants (polite,
informal, archaic, intimate etc.) should be also treated in XLIFF at some
point. dF explained that this would need to be first codified for various
locales in the CLDR. XLIFF as a transfer format would overstep its scope if it
tried to codify locale specifics in this area in a material way. dF reported on
a similar discussion that took place during the W3C ITS 2.0 Requirements
Gathering. The ITS 2.0 stakeholders agreed back in 2012 that this would need to
be addressed at the CLDR level first.

XLIFF TC Public Info Sessions were
adjourned on 7th June 2016 at 5:15pm WEST

Attachment:
FEISGILTT 2016 XLIFF Public Info sessions.docx

Description: application/vnd.openxmlformats-officedocument.wordprocessingml.document