[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [Interoperability Now] XLIFF:doc Version 0.85
Ah, so that's your point. Yes. Here's my problem: I think we need something small and compact. I suggested a slimmed down mapping from TBX (in XML), but got strong feedback from all 3 reviewers that we shouldn't re-invent the wheel, and should use an existing standard. The only standard for terminology I know of that is smaller than TBX is UTX. UTX is tab-separated (not exactly CSV, but not any more bulletproof of course). UTX has downsides, of course:
- it's not XML, and we'd be embedding it in the middle of an XML file.
- it has arbitrary number of columns
- it is tab-seperated, which is hard for humans to read, and very easy to have accidental data problems.
- you can't use xpath or regular xml tools to parse it.
- it doesn't provide enough columns to suit everyone. (But we're all about compromise for the greater good here, so...)
If it's acceptable to create an XML "version" of UTX, then I'll do it in DX, but I don't know how much "better" that is than just using the version I originally proposed. Translating UTX into XML makes it a completely different format imho.
Global Translation Solutions<http://mitintra.corp.medtronic.com/mdt-ops/procure/Global_Translation_Solutions.html>
Micah Bly
Localization Manager | Medtronic, Inc.
710 Medtronic Parkway LS260 | Minneapolis, MN 55432
Office: +1 763-505-1515 micah.j.bly@medtronic.com<mailto:micah.j.bly@medtronic.com>
On Aug 19, 2011, at 8:53 AM, Dr. David Filip wrote:
Hi Jörg, I am with you, including a csv in a 21st century standard format seems a joke.. I did not respond to Micah when he annonced that he plans to use UTX. But then, I thought he would make some simple XML mapping. IMHO, there is no reason why the UTX community should not include your proposed XML mapping in their standardization, and at the same time, I think that unless you are going to propose it to them, it is not going to happen any soon..
I think that in terms of categories, UTX is just fine as the minimum format. But csv is not a sufficiently stable/unambiguous standard format and it would introduce unneded loose ends into your profile unless it is XML mapped.. I think that Alan Melby and others (including me) tried to persuade Yuji in Danvers that going for XML would be a good idea, but he did not seem to like it too much..
In case you are not in touch with UTX people, here goes [ admin edit: YAMAMOTO Yuji via LinkedIn http://www.linkedin,com/yyuji ]
I told him that IN! might need an XML mapping of UTX :-)mailto: david.filip@ul.ie<mailto:david.filip@ul.ie>
Rgds
dF
Dr. David Filip
=======================
LRC | CNGL | LT-Web | CSIS
University of Limerick, Ireland
telephone: +353-6120-2781
mobile: +353-86-049-34-68
facsimile: +353-6120-2734
2011/8/19 "Prof. Dr. Jörg Schütz" <joerg@bioloom.de<mailto:joerg@bioloom.de>>
Hi Micah,Office: +1 763-505-1515<tel:%2B1%20763-505-1515> micah.j.bly@medtronic.com<mailto:micah.j.bly@medtronic.com><mailto:micah.j.bly@medtronic.com<mailto:micah.j.bly@medtronic.com>>
I like your UTX approach although for me the column specification can only be seen as an inbetween temporary solution... Nevertheless, we should follow how the UTX ecosystem and community will evolve because the format is also (or primarily) intended to support machine translation applications.
For xliff:doc we should aim at a quite compact representation, i.e. limited number of columns.
All the best,
Jörg
On 08/18/2011 01:00 AM, Micah Bly wrote:
DX: bumped to 0.85
DX: revised the statuses and the table with the sample status usage, based on reviewer feedback.
DX: added some limitations to the match-quality number. It still allows for decimal numbers, but only 1 digit to the right of the decimal. (eg, 99.5, but not 99.52).
DX: based on feedback from beta reviewers, I added an alternative for the proposed scheme for embedded terminology. This one uses the UTX standard rather than a new standard, and also switches from embedding each term once per TU, to a look-up model. I have not removed references to the other model yet, and have not added full reference info to the new model. Those changes will come when a final decision has been made.
Chase: does the change to match-quality work for you now?
Everyone: what do you think of the new proposal for embedding terminology? Better? Should we dump the old one?
Micah Bly
Localization Manager | Medtronic, Inc.
Global Translation Solutions
710 Medtronic Parkway LS260 | Minneapolis, MN 55432
To view this notice in other languages you can either select the following link or manually copy and paste the link into the address bar of a web browser: http://emaildisclaimer.medtronic.com<http://emaildisclaimer.medtronic.com/>
[CONFIDENTIALITY AND PRIVACY NOTICE]
Information transmitted by this email is proprietary to Medtronic and is intended for use only by the individual or entity to which it is addressed, and may contain information that is private, privileged, confidential or exempt from disclosure under applicable law. If you are not the intended recipient or it appears that this mail has been forwarded to you without proper authority, you are notified that any use or dissemination of this information in any manner is strictly prohibited. In such cases, please delete this mail from your records.
_______________________________________________
Core mailing list
Core@interoperability-now.org<mailto:Core@interoperability-now.org>
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]