So this cannot be accomplished by sending
along two different XLIFF documents each contain different source-target
languages, e.g. en+es or en+de, in the same "package"? It has
to all be squeezed into the same file because the process meta-data has
no ability to establish that cross referencing?
From:
Ryan King <>
To:
Helena S Chapman/San
Jose/IBM@IBMUS, Shirley Coady <>
Cc:
"Dr. David Filip"
<>, "" <>,
Yves Savourel <>
Date:
01/17/2013 12:54 PM
Subject:
RE: [xliff]
RE: Reference Language <was: [xliff] 1.2 to 2.0 Gaps and Proposals>
Sent by:
<>
Let me give you a use case
for a Reference Language. We localize our products and services into Luxembourgish.
We usually do this simultaneous with German, however, Luxembourgish translators
will benefit from German translations, so being able to provide them with
German reference as it becomes available, as they translate, will help
increase their efficiency and quality. Another use case are languages such
as Quiche where it is difficult to find a translator who speaks English
well enough. In this case, we translate into Spanish first, and then provide
Spanish as a Reference language for translators, while reviewers and editors
who typically speak better English, can benefit from the English source.
Thanks,
ryan
From:
[
mailto:
]
On Behalf Of Helena S Chapman
Sent: Tuesday, December 18, 2012 7:44 AM
To: Shirley Coady
Cc: Dr. David Filip; Ryan King; ; Yves Savourel
Subject: RE: [xliff] RE: Reference Language <was: [xliff] 1.2 to
2.0 Gaps and Proposals>
Maybe I am missing something but can someone
explain to me why a 3rd target language (or 4th, 5th, 6th, 100th for that
matter) for the segment has to reside in the same XLIFF file and cannot
be managed through other outside structure? (file system, directory etc.)
From: Shirley
Coady <
>
To: Helena
S Chapman/San Jose/IBM@IBMUS, "Dr. David Filip" <
>
Cc: "Dr.
David Filip" <
>,
Ryan King <
>,
"
"
<
>,
Yves Savourel <
>
Date: 12/18/2012
10:28 AM
Subject: RE:
[xliff] RE: Reference Language <was: [xliff] 1.2 to 2.0 Gaps and Proposals>
Sent by: <
>
I am personally in favor of having the third target language specified
for matches – I understand at IBM the situation is a bit different, but
as a tool provider, I need to accept that the XLIFF created by my tool
is not necessarily going to be handled or processed within my tool. Being
able to ship reference matches – and read other tools’ reference matches
because they are in a standardized format my tool can accept – would be
ideal for this type of cross-tool situation.
Shirley
From:
[
mailto:
]
On Behalf Of Helena S Chapman
Sent: December-17-12 11:11 AM
To: Dr. David Filip
Cc: Dr. David Filip; Ryan King;
;
Yves Savourel
Subject: Re: [xliff] RE: Reference Language <was: [xliff] 1.2 to
2.0 Gaps and Proposals>
I personally disagree with having the third target language specified for
the matches. We do not do so at IBM and I believe this unnecessarily complicates
the XLIFF, an interchange format, file with added complexity. We manage
those complexity outside of XLIFF interchange process. XLIFF is not a container,
it is merely an interchange format which can be used by some other container
specification.
From: "Dr.
David Filip" <
>
To: "Dr.
David Filip" <
>
Cc: Ryan King
<
>,
Yves Savourel <
>,
"
"
<
>
Date: 12/16/2012
08:56 AM
Subject: Re:
[xliff] RE: Reference Language <was: [xliff] 1.2 to 2.0 Gaps and Proposals>
Sent by: <
>
Everyone, there were no comments on this one for some time.
Can we assume consensus and change the spec accordingly?
Summary:
Reference language was approved as an additional feature for the matches
module by a TC ballot.
In this thread stakeholders have agreed that source will be mandatory also
for reference matches
Matches will remain the same structurally but will be able to carry matches
with a third target language if marked by the optional reference flag.
Thanks
dF
Dr. David Filip
=======================
LRC
CNGL
LT-Web
CSIS
University of Limerick, Ireland
telephone: +353-6120-2781
cellphone: +353-86-0222-158
facsimile: +353-6120-2734
mailto:
On Tue, Dec 11, 2012 at 11:59 PM, Dr. David Filip <
>
wrote:
Thanks Ryan, I am OK with the whole proposal now.
Cheers
dF
Dr. David Filip
=======================
LRC
CNGL
LT-Web
CSIS
University of Limerick, Ireland
telephone:
+353-6120-2781
cellphone:
+353-86-0222-158
facsimile:
+353-6120-2734
mailto:
On Tue, Dec 11, 2012 at 11:10 PM, Ryan King <
>
wrote:
Thanks David for the clarifications. I agree there is no need to have more
than one <matches> module then as long as the following is allowed:
<unit id="1">
<segment>
<source>hello
world</source>
</segment>
<mtc:matches>
<mtc:match>
<segment>
<source xml:lang=”en-us”>hello world</source>
<target xml:lang=”ca-es”>hola món</target>
</segment>
</mtc:match>
<mtc:match
reference=”yes”>
<segment>
<source
xml:lang=”en-us”>hello world</target>
<target
xml:lang=”es-es”>hola mundo</target>
</segment>
</mtc:match>
</mtc:matches>
</unit>
As for process requirements,
it states this for <target> in the core spec:
·
When a <target>
element is a child of <segment> or <ignorable> and the optional
xml:lang attribute is present, its value must be equal to the value of
the tgtLang attribute of the enclosing <xliff> element.
My comment on a needed processing
requirement was to override this for <mtc:matches>
.
E.g.
·
When a <target>
element is a child of <segment> contained in an <mtc:match>
element whose reference attribute is set to “yes”, and the optional xml:lang
attribute is present, its value is not required to be equal to the value
of the tgtLang attribute of the enclosing <xliff> element.
Thanks,
ryan
From: Dr. David Filip [mailto:
]
Sent: Tuesday, December 11, 2012 1:11 PM
To: Ryan King
Cc: Yves Savourel;
Subject: Re: Reference Language <was: [xliff] 1.2 to 2.0 Gaps and
Proposals>
Ryan, all
I do support reference matches in
matches module. It [the whole module] will however need more PRs than just
saying that th ereference can be in a different language, which is even
NOT a PR :-)
Provided that source remains obligatory
as so far..
+---<match> +
+---<xlf:source>
1
+---<xlf:target>
1
It was possible upfront to have
+ = one or more matches in <matches>
<matches> 1
+---<match> +
Or do you mean something else? I
do not think that the top element should be duplicated, you can mix reference
and leverage matches in one IMHO, what would be the advantage of having
more top level elements?
Cheers
dF
Dr. David Filip
=======================
LRC
CNGL
LT-Web
CSIS
University of Limerick, Ireland
telephone:
+353-6120-2781
cellphone:
+353-86-0222-158
facsimile:
+353-6120-2734
mailto:
On Tue, Dec 11, 2012 at 8:08 PM,
Ryan King <
>
wrote:
Do we have consensus on this proposal? E.g.
Add an optional attribute reference=”yes no” with no as default. PR for
a “reference match” would be to allow an xml:lang on the target different
from the document. Additionally, we’d need to a llow
more than one <mtc:matches> where we currently only allow one sine
we could have both recycling and reference language present at the same
time.
<mtc:matches>
<mtc:match reference=”yes”>
<segment>
<source xml:lang=”en-us”>hello
world</target>
<target xml:lang=”es-es”>hola
mundo</target>
</segment>
</mtc:match>
</match>
Thanks,
ryan
From:
[mailto:
]
On Behalf Of Ryan King
Sent: Monday, December 3, 2012 11:46 AM
To: Dr. David Filip; Yves Savourel
Cc:
Subject: RE: [xliff] 1.2 to 2.0 Gaps and Proposals
Sounds good. Let’s keep source
in Reference Language.
From: Dr. David Filip [
mailto:
]
Sent: Saturday, December 1, 2012 11:17 AM
To: Yves Savourel
Cc: Ryan King;
Subject: Re: [xliff] 1.2 to 2.0 Gaps and Proposals
Yves, Ryan,
the source should be required in
all matches, reference or not. This was one very specific piece of feedback
from the toolmakers on the 2nd XLIFF Symposium in Warsaw. SDL, Kilgray,
Atril, and more agreed on that having no source in alt-trans complicated
the processing unnecessarily and said that they would provide better support
to an XLIFF local matching mechanism if it had mandatory source. We should
honot this wish in the matches module IMHO
So it might seem as redundancy but
actually is not so bad and explicitly supported by the voice of an important
constituency..
Cheers
dF
Dr. David Filip
=======================
LRC
CNGL
LT-Web
CSIS
University of Limerick, Ireland
telephone:
+353-6120-2781
cellphone:
+353-86-0222-158
facsimile:
+353-6120-2734
mailto:
On Thu, Nov 29, 2012 at 3:47 AM,
Yves Savourel <
>
wrote:
Hi Ryan, all,
Sorry for the delay: I'm just swamped and can't find the time to read emails
anymore.
> 1. Be able to specify optional custom values
> for match type in <mtc:matches>
I suppose some mechanism similar
to the subType we're using in inline codes and other places could allow
for custom values while making sure a top-level category is also declared.
Since we are discussing values for match type: I'm still not convinced
that the latest list makes sense:
am - Assembled Match
ebm - Example-based Machine Translation
idm - ID-based Match
ice - In-Context Exact Match
mt - Machine Translation
tm - Translation Memory Match
- 'Example-based Machine Translation' should not be there IMO: it's just
MT, what type of MT is not relevant (but could be a candidate for the subtype)
- 'In-Context Exact Match' IMO should be 'in-context' only: the fact that's
an exact one is captured in the similarity (and it could be an in-context
fuzzy too).
> 2. Support Reference Language in <mtc:matches>
> • Allow zero, one or more <mtc:matches> at each extension point,
because
> you might have both recycling and reference language data.
I assume you mean: allow more than
one <mtc:matches> where we currently allow one? Not in *all* extensions
point. right?
> • Add an optional attribute reference=”yes no” with no as default.
> Additionally, PR for a “reference
match” would be to allow an xml:lang on the target
> different from the document and allow the <source> not to be
present
> as it would be redundant information with the core <source>,
e.g. Spanish
> reference for Quechua might look like this:
- reference='yes
o' and allowing
a different language for xml:lang in those with reference='yes' seems ok
to me.
- source not being present... I don't know. If we do that for those 'matches'
why not for the normalmatches as well? If the source is the same.
I think we mandated the source originally that's to simplify processing:
testing for the presence of not of the source may be cumbersome for some
processors (XSLT maybe?).
We would need to update the definition of what a "match" is as
well.
hope this helps,
-ys
---------------------------------------------------------------------
To unsubscribe, e-mail:
For additional commands, e-mail:
