Next in thread →
Next in month →
RE: [xliff] Minutes of the XLIFF TC July 9th. 2002
Hi Christian, In the following untranslated XLIFF excerpt there are two terms, both "View Properties"; consider the case where the first means 'the properties of a view' and the second means and should be translated as 'look at the properties of a selected object'. Without the contextual information provided, translators and translation recovery applications would most probably translate both cases to the same translation, which would be wrong in most languages other than English. ...<trans-unit id="00343EA980256C06"> <source>View Properties</source> <target>View Properties</target> <context-group name="Term Context Information" > <context context-type="Database">Discussion - Notes & Web (6)</context> <context context-type="Element">Forms</context> <context context-type="ElementTitle">(ProfileDisplay)</context> <context context-type="Record">Field Default Value Formula</context> <context context-type="RecordTitle">Categories</context> </context-group> </trans-unit> <trans-unit id="00343BBB80256C06"> <source>View Properties</source> <target>View Properties</target> <context-group name="Term Context Information" > <context context-type="Database">Discussion - Notes & Web (6)</context> <context context-type="Element">Views</context> <context context-type="ElementTitle">Interest Profile</context> <context context-type="Record">Action Title</context> </context-group> </trans-unit>... I had a look at the working draft spec again and ultimately I think that the comment for <context> should suffice "The purpose of this context information is to allow certain pieces of text to have different translations depending on where they came from. The translation of a piece of text may differ if it is a web form or a dialog or an Oracle form or a Lotus form for example. This information is thus required by a translator when working on the file. Likewise, the information may be used by any tool proposing to automatically leverage the text successfully.". We could duplicate this for the 'context-type' attribute or add a referral back to the 'context' element. Regards, Mark "Lieske, Christian" <> 07/30/02 01:03 PM To: Mark Levins/Ireland/IBM@LOTUSINT cc: Subject: RE: [xliff] Minutes of the XLIFF TC July 9th. 2002 Hi Mark, You wrote 'Perhaps a simple example of the usage of <context> would prove more useful', and I share your feeling. Would it be possible for you, to provide a couple of examples? Best, Christian -----Original Message----- From: Mark Levins/Ireland/IBM [mailto:] Sent: Dienstag, 30. Juli 2002 13:03 To: Peter Reynolds <Peter.Reynolds Cc: Subject: Re: [xliff] Minutes of the XLIFF TC July 9th. 2002 Hi Peter et al, I missed the meeting but from the minutes see my action item. The minutes state that I need to supply more information for 'content-type', however, assuming a typo, I'll make a quick explanation for the 'context-type' attribute of the 'context' element which I requested in XLIFF. Within IBM, during the localisation of Domino applications, translatable terms may appear in many possible locations and without a definite resourcing strategy. The context element allows us to locate a term to a specific location in a database and therefore provide a different contextual translation to a similar term where needed. I'm also guessing that the context of the conversation during the last meeting was in relation to open or closed lists, and (if so) I would suggest that this list be left open altogether and remove the existant defaults. The original list was somewhat a compromised catch-all that was agreeable to both me and to the rest of the group, and I think they are probably a little misleading to others on reading the specification. Perhaps a simple example of the usage of <context> would prove more useful. If I've misunderstood what was needed of me in this mail, please let me know. Regards, Mark P.S. For anyone interested, the <context> element can also be useful when extracting localisable text from property bundle files withing JAR or CAB files to XLIFF i.e. the context being used to identify the bundle file and the resource ID for the translatable term Peter Reynolds <> 07/26/02 02:37 PM To: cc: Subject: [xliff] Minutes of the XLIFF TC July 9th. 2002 Hi all, Minutes of last Tuesday's meeting is below. Thanks to Tony for his help with these minutes. Peter. Minutes of the XLIFF TC July 9th. 2002 1/ Attendance: Tony Jewtushenko, Christian Lieske, Enda McDonnell, John Reid, Peter Reynolds, Yves Savourel, G廨ard Cattin des Bois, David Pooley. Apologies: Jonathan Clark, John Corrigan, Reinhard Schaler, Mark Levins, Mat Lovatt. 2/ Minutes of the last meeting were agreed as a correct record. 3/ Charter clarification. The proposed new wording of the charter was agreed by 12 votes for and 2 abstentions. Both abstentions were late votes. Tony pointed out that Oasis refer to this changing of the wording as clarification and not changing it. 3/ Migration policy Tony introduced a discussion on the remaining issues with the migration policy. We will publish a migration statement as part of the version 1.1 release. XSLTs will not be part of the 'official' release but may be provided as samples. Tony will send a revised version of the migration document to the group. This will go to an email vote for approval. Tony raised the issue of education and publicity and said he was willing to draft a presentation on behalf of the TC. Peter and Enda volunteered to help. 4/ XLIFF 1.1 work in progress: Formalized Extensibility Yves presented the proposal on Formal Extensibility. Christian will send a proposal for dealing with datatype and mimetype. (general comment): It was agreed that each of the presently open enumerated lists would be extended and left open. content-type - assumed to be needed by Mark Levins - the group was not certain of precisely how the list is used in real life - need input from Mark (and anyone else who uses this) as to what the list should look like. ctype, mtype lists : review, comment and add as much to this list as possible, and leave it open. state - state names in the spec are not complete, and must be extended, but must remain open. It was agreed that the list of states would not be possible to close due to the nature and complexity of a localisation process in particular and workflow in general. Gerard agreed to submit a "straw man" list of values, and Christian will investigate if LISA has a subcommittee that has produced a list. Embedded XLIFF - deferred until next meeting. Working Draft - deferred until next meeting. 5/ LISA European Conference: we will try to arrange a limited face-to-face meeting during our usual timeslot, if enough people attending the LISA conference would have time to do so. Please inform Tony J. if you have definite intentions to attend the LISA conference. 6/ Next meeting will be Tuesday, 6 August 2002 at 4PM Europe/London/Dublin / 8AM PST. Tony J. will be absent, and will arrange for Jonathan Clarke to chair the meeting, or if not available, Peter or Yves. John Reid made a motion to adjourn the meeting and Enda seconded. Peter Reynolds, Software Development Manager, Berlitz GlobalNET 3, West Pier Business Campus, Dun Laoghaire, Co. Dublin, Ireland Tel: +353-1-202-1280 Fax: +353-1- 202-1299 Web Site: http//www.berlitzglobalnet.com "Don't just translate it, BerlitzIT" http//www.berlitzIT.com ---------------------------------------------------------------- To subscribe or unsubscribe from this elist use the subscription manager: <http://lists.oasis-open.org/ob/adm.pl>
Next in thread →
Next in month →