OASIS XML Localisation Interchange File Format (XLIFF) TC

  • 1.  Translation status values for 2.0

    Posted 04-22-2011 15:12
    Talking about attributes: one information that is important and may be useful to have as part of the core is the status of the translation. In 1.2 for example we have a approved in <trans-unit> and state in <target>. I think this need to be re-work in a way that allows to know the status of the segment via one attribute. Looking at the different tools and their usage of state, it seems it varies a lot. But this is one specific list of values that is going to be important to get carefully defined so it has some logical progression but covers as much as possible the states required by the different tools. Each value needs to be carefully defined so misue risks are reduced. We may also want to talk directly to developers and see how they map their values to it. -ys


  • 2.  RE: [xliff] Translation status values for 2.0

    Posted 04-22-2011 15:52
    Hi Yves, I would move the "state" attribute from <target> to <segment> element and review the list of possible values and perhaps change its name. We still need to know if a segment is approved or not, regardless of its current status (a segment may be approved by a translator but require further proofreading in a later stage). We still need two separate attributes unless we define a value set that allows us to differentiate when the translation can be included in the final document. I can modify the schema and add the "state" attribute to <segment>, using a tentative list of values that will have to be reviewed later. Should I post an updated schema? Regards, Rodolfo -- Rodolfo M. Raya <rmraya@maxprograms.com> Maxprograms http://www.maxprograms.com >
    Original Message


  • 3.  RE: [xliff] Translation status values for 2.0

    Posted 04-22-2011 16:48
    Hi Rodolfo,

    I'm not sure (yet) that moving the state attribute from <target> is safe to do. Adding it to <segment> seems like a good enough idea. I would lean toward not moving the state attribute from <target> until we consider the use cases that could convince us that a collection of state attributes in several <segment> elements could always be interpreted to inform us of the overall state of the larger <target> (my stance is based on the assumption that a <target> can/will contain several <segment> elements).

    - Bryan


    Original Message


  • 4.  RE: [xliff] Translation status values for 2.0

    Posted 04-26-2011 14:20
    Hi Yves, all,

    True: an attribute related to "status" with regard to translation is definitely a requirement that exists.

    I would think that we may need it on a least two levels:

    1. file
    2. trans-unit/segment

    I also agree that a careful definition of the possible values (and correspondingly the processing requirements/expectations) is needed. Possibly, we will have to consider the idea of "phases" that was circulated at the XLIFF Symposium in Limerick. Looking at this, we would have "phases/processes" related to linguistic work such as "language editing", "translation", "transcreation", "transliteration", and "states" such as "aborted, executable, finished, not-executable, preliminary-finished, started, suspended/paused, resumed".

    Cheers,
    Christian


    Original Message


Related Content