OASIS Open

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

A. Admin

1- Roll call

dF, Chase, Felix, James, Phil, Soroush, Yves, Robert (quorum with 7 out of 8 voters)

2- Minutes from the last meeting 24 Jan 2017

dF: I move that minutes at the following URI are the official record of our previous meeting from 14th Feb, 2017

(https://www.oasis-open.org/apps/org/workgroup/xliff-omos/email/archives/201702/msg00013.html)

Chase: I second.

[no objections]

dF: It’s approved.

B. Material

1- XLIFF OM

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

dF: Felix and I are both very busy with XLIFF 2.1 public review and unfortunately there is no progress on OM, we probably would need more mebership participation on the -om Github repo.

2- JLIFF

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

Mailing list feedback from Phil (https://www.oasis-open.org/apps/org/workgroup/xliff-omos/email/archives/201702/msg00011.html, https://www.oasis-open.org/apps/org/workgroup/xliff-omos/email/archives/201702/msg00012.html)

See progress of representing modules vs. general extensibility with the metagroup method

look into inheritance of global preperties from OM structural levels, liff or file levels, or group recursivity?

Robert: The progress is based on the received feedback. We added metadata elements which are small deviation from the OM with no id. The next thing is on the wiki page. Chase and Phil gave very helpful comments there. The latest draft schema is 0.9.3. Segment is now marked as default, then an ignorable needs to be annotated. The examples are also changed accordingly. It is essentially choice of two different objects with different attributes. 

dF: The ignorable shouldn’t have the ‘state’ attribute, it has only an optional ‘id’.

Robert: Sure, I willmake this change in 0.9.3. won't make a next version

dF: On the other hand, ‘ignorable’ can sort of inherit ‘canResegment’ form a higher level. We need to consider instantiation of propreties from higher up in xliff when dealing with JLIFF that only represents a LIFF fragment.

Robert: The other issue we will need to think of is the ‘id' and ‘idref’ mechanism with no standard representation in JSON which are left off so far.

dF: as mentioned previously, XLIFF is also not using the id / idref that is standard in XML, we are using a custom mechanism based on NMTOKEN mostly..

dF: Phil sent his feedback based on prototype imeplementations of the serialization mandated with teh JSON schema. What's you rtake on that, Robert? Phil, are you able to serialize and deserialize acording the current schema?

Phil: I think I am able to implement it. But the main question is how simple we would want to make it for people to implement this; I use a library for JSON serilization and  deserialization and it reflects easily on my classes. So if JSON is explicit, the deserializer can almost automatically match up the characteristics of the model with my classes. But when we deal with an array of objects and string values, then you have to interfere with the serialization pipeline much more.

Robert: It is a very important point on JSON disability to provide stream of objects with properties. I think the question is if we want to keep the schema simple or the serialization/deserialization.

dF: We might need more examples. Robert, do you have any progress on instance templates  (besides examples and schema)?

Robert: I think we could start working on that once schema is in a good shape.

dF: We should bear in mind that the goal is seamless mapping between XML and JSON pipelines and all the data model properties, even from higher up in LIFF should be instantiated in JSON/JLIFF. We should provide processing requirements at some point that will guarantee lossless swiching between the XML and JSON pipelines.

Robert: I will update the schema with adding id to subunit.

dF: The other topic we should work on is extensibility.

Robert & Chase: Not yet.

Chase: this is my pending AI..

dF: As discussed last time, we could start with the MDA module which is JSON-friendly. So we concluded to use this model for any modularity and extensibility (more generic ‘metadata’, as a class). ‘subState’ attribute could be also a good example of extensibility by design (custom values introduced by authorities prefixes such as msft:, the authority and custom value are separated by ":"). Is it possible to define such properties in JSON and validate them?

Robert: I will take a look at that and crosscheck. Should be possible

3- TBX Mapping

Report from editorial 1-2-1 meeting

current editor's draft: https://tools.oasis-open.org/version-control/browse/wsvn/xliff-omos/trunk/XLIFF-TBX/xliff-tbx-v1.0.pdf

dF: No meeting with James today unfortunately. But I made the print out of editors’ draft with the changes agreed in previous weeks applied. The latest PDF in now on our SVN. TBX progress on ISO front is very slow and we actually are not in rush for this item, as it doesn't make sense to have the specific mapping before the new TBX structure is stable.  James, would you please share the progress on namespaces and validation discussions at the TBX SC?

James: They are looking to get a namespace for TBX for the first time. ISO namespace is being considered. Also, the root element is changed to <tbx> instead of <martif>. Other changes: term entry is going to be changed to concept entry, langSec (language section) will be langGroup, tig will be termGroup. The Basic dialect will also need to go through the name changes. This will effect the mapping as well as all utilities.

C. Other Topics

1- Date of next meeting

regular date: 14th March

dF: Since several voters cannot attend due to spring break, we will skip March 14th and the next meeting will be  on 28th March.

2- Liaisons

dF: GALA TAPICC : will report on OMOS JLIFF and OM progress at GALA Amsterdam as part of the TAPICC session

TBX Steering Committee / ISO TC37/SC3 [covered as part of TBX mapping]

XLIFF TC

dF: 2nd public review closed

Many comments, some require matreial changes, so there will be a 3rd public review, hopefully opens still in March.

(https://www.oasis-open.org/news/announcements/public-review-for-xliff-v2-1-ends-february-24th, You can follow the new comments / issues handling at https://issues.oasis-open.org/browse/XLIFF/?selectedTab=com.atlassian.jira.jira-projects-plugin:issues-panel)

At this point you SHOULD NOT log new comments, but you can follow up on open issues via the "Environment" threads

E.g. the comment list thread for https://issues.oasis-open.org/browse/XLIFF-32 is http://markmail.org/thread/udmpeio52di5p44s [link from the "Environment" field of the JIRA issue]

This way all related feedback will be gathered in the relevant thread referenced from the JIRA issue.

dF: Chase, as you sent quite heavy comments on CTR, you should cosnider joining the sister TC if only to help finalize CTRM, the indurect discussion becomes too difficult and lengthy..

3- Promotion

dF: Start planning for November in Silicon Valley https://locworld.com/events/locworld35-silicon-valley-2017/

The main conf Nov 1-Nov 3, FEISGILTT likely Oct 31-Nov 1, 2017. Will be posting CFP, now just save the dates..

GALA Amsterdam 2017 https://www.gala-global.org/all-events/gala-2017-annual-conference March 26-29

Planned JIMACATT workshop in The Hague, May 10-12, 2017

http://www.jiamcatt.org/Home register: https://ec.europa.eu/eusurvey/runner/JIAMCATT-2017-Registration 

4- AOB

dF: If no other business, the meeting is adjourned. 

A. Admin

1- Roll call

8 voters total

2- Minutes from the last meeting 24 Jan 2017 

Approve minutes from 14th Feb, 2017

https://www.oasis-open.org/apps/org/workgroup/xliff-omos/email/archives/201702/msg00013.html

B. Material

1- XLIFF OM

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

2- JLIFF

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

Mailing list feedback from Phil

https://www.oasis-open.org/apps/org/workgroup/xliff-omos/email/archives/201702/msg00011.html

https://www.oasis-open.org/apps/org/workgroup/xliff-omos/email/archives/201702/msg00012.html

See progress of representing modules vs. general extensibility with the metagroup method

look into inheritance of global preporties from OM structural levels, liff or file levels, or group recursivity?

3- TBX Mapping

Report from editorial 1-2-1 meeting 

current editor's draft: https://tools.oasis-open.org/version-control/browse/wsvn/xliff-omos/trunk/XLIFF-TBX/xliff-tbx-v1.0.pdf

C. Other Topics

1- Date of next meeting

regular date: 14th March, BUT US will be already on DST and Europe not yet.. 

2- Liaisons

dF: GALA TAPICC

TBX Steering Committee / ISO TC37/SC3

XLIFF TC

2nd public review closed

Many comments, there will be a 3rd public review, hopefullz opens within March.

(https://www.oasis-open.org/news/announcements/public-review-for-xliff-v2-1-ends-february-24th,

You can follow the new comments / issues handling at

https://issues.oasis-open.org/browse/XLIFF/?selectedTab=com.atlassian.jira.jira-projects-plugin:issues-panel

At this point you SHOULD NOT log new comments, but zou can follow up on open issues via the "Environment" threads

E.g. the comment list thread for https://issues.oasis-open.org/browse/XLIFF-32

is http://markmail.org/thread/udmpeio52di5p44s

This way all related feedback will be gathered in the relevant thread referenced from the JIRA issue.

3- Promotion

Start planning for November in Silicon Valley https://locworld.com/events/locworld35-silicon-valley-2017/

The main conf Nov 1-Nov 3, FEISGILTT likely Oct 31-Nov 1, 2017. Will be posting CFP, now just save the dates..

GALA Amsterdam 2017 https://www.gala-global.org/all-events/gala-2017-annual-conference March 26-29

Planned JIMACATT workshop in The Hague, May 10-12, 2017

http://www.jiamcatt.org/Home register: https://ec.europa.eu/eusurvey/runner/JIAMCATT-2017-Registration

4- AOB

• OASIS Open (General Membership)
• General Public