< Return to Calendar

* 4th Tuesday - Regular XLIFF OMOS TC telecenference (Conference Call)
Name * 4th Tuesday - Regular XLIFF OMOS TC telecenference (Conference Call)
Time Tuesday, 26 June 2018, 05:00pm to 06:00pm WEST
(Tuesday, 26 June 2018, 04:00pm to 05:00pm UTC)
Description

See the private action item for dial in details

https://www.oasis-open.org/apps/org/workgroup/xliff-omos/members/action_item.php?action_item_id=3822
Minutes # OMOS 2018-06-26 #

Present: phil (minutes), robert, david, james

Approve minutes from 12/6
rve seconded

df: I can see that chanes were made in the repo
...I updated phil on the change of boolean values to yes/no enumeration
...did you check phil?

pr: no sorry I didn't have a chance

df: the reasons for this change are agreed amongst others

rve: in principle not much difference
...but in programming languages generally bool defaults to false
...but in omos we want default to by `yes`

df: extensions separator is colon
...latest commit refers to addition of lang
...so we haven't yet introduced teh underscore separators

rve: json-ld would try to resolve colon separated names

df: would we use underscore or dash

rve: it does not matter in schema but for implementers?
...in some languages you can reference properties as language identifiers
...strings are dictionary keys

df: AI for Phil to check as an implementer whether dash or underscore are preferable

rve: .NET tends to use underscores
...underscores are used very frequently in wsdl schemas
...I think underscore will be better

df: `annotatorsRef` of its, Robert did you make any progress on this?

rve: Not yet

df: stephen mentioned to approach as an array
...or as a regex
...we agreed on dictionary

rve: I don't recall if the dictionary keys are fixed

df: annotatorsRef can have multiple values where bar separates category and value
...consists of space separated lists of category|string

rve: actually I think I implemented this already
...if you look in the schema at line 867
...pattern is a sequence of characters except space or bar

df: the first part is always teh data category, the string pattern is the same for all values

rve: must be non-empty
...nice and simple

df: yes works well
...domain is the same

rve: I don't think there's much else to do
...toolsAnnotation is done
...lqi is done
...text analysis is done

df: on localization note there is nothing
...terminology we have partially in core
...in core there is sm of type term

rve: I think that is done

df: so we have covered the core but there is also ITS part
...this is its specific
...for full support of terminology
...termConfidence, this is again can be on `sm`

rve: same as taConfidence

df: also mtConfidence

rve: I can move forward with these changes

df: probably you cannot capture that termConfidence requires cooccurance constraint of annotatorsRef must be given

rve: some can be represented but it is limited
...a dependence constraint can be added to the schema

df: so last remaining is mtConfidence

rve: should I start using underscores for modules?

df: please change the version number

## TBX ##
df: TAPICC XLIFF extracion paper will go for public review soon through GALA on 28
...its on GitHub
[draft](https://galaglobal.github.io)
...it covers a gap

## Next Meeting ##
10th July 2018

Agenda

Agenda

A. Admin

 

1- Roll call

 

? out of 5 voters:

aprove minutes from 12th June 2018

https://www.oasis-open.org/apps/org/workgroup/xliff-omos/email/archives/201806/msg00002.html

 

B. Material

 

1- XLIFF OM

 

OM wiki needs aligned with current JLIFF structure as per 0.9.7

 

https://github.com/oasis-tcs/xliff-omos-om/wiki

 

 

 

2- JLIFF

 

(https://github.com/oasis-tcs/xliff-omos-jliff)

 

Consensus on boolean vs a yes/no enumeration restated:

 

Stick with yes/no enumeration becuase boolean default is false, while XLIFF defaults are yes. 

 

Previous Consensus restated::

Don't use external context for extensions.

All extension context must be stated inline to avoid parsing external context files..

We also agreed that having a dedicated extension container is more validation friendly than just allowing additional properties on the root structure..

-Continue discussing pros and cons of the extensionsData approach

compare with XML and consider going back and forth between XML and JSON.

 

[ACTION ITEMS]: 

Robert: look into mtc:match: Add "domains" to translationCandidatesMatch. 

Robert: remove Tools Annotations from , add to the structural elements (file, group, unit)

Robert: create RegEx for value of annotatorsRef

Robert: Serialize annotatorsRef as an dictionary.

Robert: Serialize domains as a non-zero length array

Next Agenda:  Feedback on prefix solutions

David: Get feedback on the yes|no question

 

 

 

3- TBX Mapping

TBX-Basic mapping is in order, almost done on TBXInfo

 

 

C- Other Topics

 

 

3- Promotion

 

Also Phil JLIFF library open sourcing was announced in the GALA week 

 

https://twitter.com/VistatecGlobal/status/974538466565373952

 

https://twitter.com/merzbauer/status/974288543093854208

 

GALA TAPICC needs to work on launching JLIFF based Track 2

 

Will probably launch by end of June

 

4- AOB

 

1- Date of next meeting

 

Tentative 10th July 2018
Submitter Dr. David Filip
GroupOASIS XLIFF Object Model and Other Serializations (XLIFF OMOS) TC
Access This event is visible to OASIS XLIFF Object Model and Other Serializations (XLIFF OMOS) TC and shared with
  • OASIS Open (General Membership)
  • General Public