OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

xliff message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: RE: [xliff] Type values for the match element (Y22)

As you know, there is a difference between a 100% match and a 100% match which comes from a part of a document with the same surrounding context. The latter is typically no more checked by translators, because it is assumed to be the correct translation. A 100% match can however be completely inappropriate in a certain context. So it will sometimes be paid to be checked for correctness by translators. I therefore highly support the context match value.


I am actually the one who coined the term and implemented context matches first in 2001, and because of the absence of a proper attribute I simply made them 101% matches, which still makes people with mathematical sensibility shiver. So, better we have a value which expresses this property :-)


Best regards,



From: xliff@lists.oasis-open.org [mailto:xliff@lists.oasis-open.org] On Behalf Of Helena S Chapman
Sent: Montag, 29. Oktober 2012 14:27
To: Yves Savourel
Cc: xliff@lists.oasis-open.org
Subject: Re: [xliff] Type values for the match element (Y22)


Is the tm to 100% similarity not close to ICE enough? I still don't quite get the need of it.

Also, I would suggest not using "ibm" for ID based match because that is too close to be confused as a match from IBM memory.

Best regards,

Helena Shih Chapman
Globalization Technologies and Architecture
+1-720-396-6323 or T/L 938-6323
Waltham, Massachusetts

From:        Yves Savourel <ysavourel@enlaso.com>
To:        <xliff@lists.oasis-open.org>
Date:        10/29/2012 07:25 AM
Subject:        [xliff] Type values for the match element (Y22)
Sent by:        <xliff@lists.oasis-open.org>

Hi all,

I think a few of the proposed values for the type attribute of the match element should be changed:

We have:

am - Assembled Match
ebm - Example-based Machine Translation
ibm - ID-based Match
ice - In-Context Exact Match
mt - Machine Translation
tm - Translation Memory Match

- a) It seem to me that 'ebm' should be removed: Knowing the kind of MT system produced the candidate is not the purpose here, or if it is then we should also have values for rule-based MT, hybrid-MT, etc. IMO 'mt' is enough. If some specific information is needed about the type of MT that should be a different attribute.
I propose to remove 'ebm'.

- b) I think 'ice' (In-Context Exact Match) gives too much information: the fact that a match is exact should be provided by the 'similarity' attribute. You can have in-context match that are not exact.
I propose to change 'ice' to 'icm' and the definition to "In-Context Match". We should also probably expand a bit the definitions, for example a new comer may have hard time understanding what 'context' means here.


To unsubscribe, e-mail: xliff-unsubscribe@lists.oasis-open.org
For additional commands, e-mail: xliff-help@lists.oasis-open.org

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]