xliff-inline — archive

Teleconference - Aug-10-2010 - 13:30 UTC

From: Yves Savourel <[email protected]>
To: [email protected]
Date: Wed, 11 Aug 2010 21:54:00 -0000













We can certainly make sure to use a more XML-like notation for
the example.







From: [email protected]
[mailto:[email protected]]

Sent: Wednesday, August 11, 2010 3:08 PM

To: [email protected]; Yves Savourel; [email protected]

Subject: RE: [xliff-inline] Teleconference - Aug-10-2010 - 13:30 UTC -
Minutes







Hi all,



I like the new wording for a). 



And I know the example was not meant to be included in the
actual requirements. But if we were going to include an example, I would make a
case to make it also comply with requirement 18 (which I recently proposed),
that says for well-formed XML input that does not span segments (like this
sample, "I like <i>italics</i>."), it is best practice to
preserve the inline XML element as an XML node.



So I would rather see an example of the XLIFF that preserves the
node.



Not like this:

XLIFF: I like <XLIFFtagOfSomeSort/>italics<XLIFFtagOfSomeSort/>.



But rather like this:

XLIFF: I like
<XLIFFtagOfSomeSort>italics</XLIFFtagOfSomeSort>.



Now if the source was, say, RTF:

{\insrsid12611904 \hich\af35\dbch\af11\loch\f35 I
like }

{\i\insrsid12611904\charrsid12611904 \hich\af35\dbch\af11\loch\f35
italics}

{\insrsid11469172 \par }



I'd be less picky about preserving the inline as an XML node.



I know my hair-splitting is off-topic for requirement 12.
Forgive me for trying to set the stage for requirement 18.



Thanks,



Bryan









From:
Arle Lommel [mailto:[email protected]] On Behalf Of Arle Lommel

Sent: Wednesday, August 11, 2010 8:04 AM

To: Yves Savourel; [email protected]

Subject: Re: [xliff-inline] Teleconference - Aug-10-2010 - 13:30 UTC -
Minutes







Yves, on further consideration, how
about we rephrase point A as:


a) to store only the XLIFF representation, discarding the native data


I think this is clearer, because it's not really that we're just discarding everything,
as in the following:


Source: I like <i>italics</i>.

XLIFF: I like italics.


But rather going for something like this:


XLIFF: I like <XLIFFtagOfSomeSort/>italics<XLIFFtagOfSomeSort/>.


The existing text could imply the former as well as the latter possibility.


-Arle


On 8/10/10 10:51 AM, sic scripsit Yves Savourel: 

Requirement #12: (was #10, #11, and #12)
After discussion, the three requirements related to the representation of the native data were grouped into a single one (#12), with the following wording:
[
There must be three ways to deal with the native data corresponding to an XLIFF inline code:
a) not to store it at all
b) to store it along with its XLIFF representation
c) to store a pointer to it along with its XLIFF representation
]