Yves, we're running out of time on this
The recorded resolution
https://issues.oasis-open.org/browse/XLIFF-53says
Remove the editorial uppercase comment and improve the Warning wording if
necessary.
Since it is a minor editorial issue, I suggest that we just drop the
UPPERCASE editorial comment for now and keep the Warnings as they are..
You can raise the issue of removing or improving the warnings again in the
4th public review. No danger that this one would send us for another round.
Cheers and thanks
dF
Dr. David Filip
===========
OASIS XLIFF OMOS TC Chair
OASIS XLIFF TC Secretary, Editor, Liaison Officer
Spokes Research Fellow
ADAPT Centre
KDEG, Trinity College Dublin
Mobile: +420-777-218-122
On Thu, May 18, 2017 at 7:00 PM, David Filip <
david.filip@adaptcentre.ie>
wrote:
> I agree that there are lot of cases where the syntactically correct
> doesn't give the intended info and in general we don't need to warn against
> them.
>
> I think this case is different, we say that the annotatorsRef is optional
> on that annotation when normatively defining the annotation, unless the
> annotation is in scope of an annotatorsRef for xyz datacat. We say that
> because this is what can be syntactically checked. But the intent of the
> notation is to give the intended value not just any syntactically matching
> value.
>
> The warning is informative (non-normative, as any other warning or note in
> the spec) and it helps the implementers who are coming from the XLIFF end.
>
> I think in this particular case the normative text has big potential to be
> misleading (at least for the "naive implementer" - and we agreed to be
> targeting the naive implementers, not to require "tribal knowledge" to
> successfully consume the spec), so the warning is warranted IMHO.
>
>
>
> Dr. David Filip
> ===========
> OASIS XLIFF OMOS TC Chair
> OASIS XLIFF TC Secretary, Editor, Liaison Officer
> Spokes Research Fellow
> ADAPT Centre
> KDEG, Trinity College Dublin
> Mobile: +420-777-218-122 <+420%20777%20218%20122>
>
>
> On Thu, May 18, 2017 at 6:32 PM, Yves <
yves@opentag.com> wrote:
>
>> Hi David, all,
>>
>>
>>
>> Ø In other words, valid syntax can give the wrong information for
>> reasons they haven't foreseen and they should double check. That's all.
>>
>>
>>
>> I’m more and more confused I’m afraid.
>>
>>
>>
>> Maybe you can give a concrete example of how one can give/get the wrong
>> information?
>>
>> (And how do you know it’s wrong).
>>
>>
>>
>> Would it be something like this? (From example 20):
>>
>>
>>
>> <unit id="u1" its:annotatorsRef="mt-confidence|MTServices-XYZ">
>>
>> <segment>
>>
>> <mrk id="m1" type="its:generic" its:mtConfidence="0.8982">Some
>> Machine
>>
>> Translated text.</mrk>
>>
>>
>>
>> And the tool “MTServices-XYZ” would not really be the tool that MTed the
>> source text?
>>
>>
>>
>> I don’t see how that could happen, except if some tool does not its job
>> conforming to the normal ITS constraints.
>>
>> But we don’t want such warnings: There are a lot of cases where the
>> syntax can be right but the info wrong.
>>
>>
>>
>> Also, how can you “double check” such info?
>>
>> And if you guess it is wrong, what can you do about it?
>>
>>
>>
>> It’s like having a warning saying: “The state attribute may be
>> syntactically correct but not have the intended value.” (Because a tool did
>> not updated it properly).
>>
>> Such warnings are not really helping in my opinion.
>>
>>
>>
>> Or maybe I’m missing completely the meaning of that paragraph (which is
>> quite possible).
>>
>>
>>
>> Cheers,
>>
>> -yves
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> *From:* David Filip [mailto:
david.filip@adaptcentre.ie]
>> *Sent:* Thursday, May 18, 2017 2:32 AM
>> *To:* Yves <
yves@opentag.com>
>> *Cc:* XLIFF Main List <
xliff@lists.oasis-open.org>;
>>
xliff-comment@lists.oasis-open.org>> *Subject:* [xliff-comment] Re: [xliff] RE: [xliff-comment] Re: [xliff]
>> csprd03 - remaining inline comment to resolve
>>
>>
>>
>> The reason why this is on exactly these three is that they must be in
>> scope of a *relevant* annotatorsRef (some of them conditionally, but this
>> is clear from the annotation description).
>>
>>
>>
>> This warning makes the implementer aware that being in scope is enough
>> for validity but might not be enough to give the intended annotatorsRef
>> information, I think that's an important warning and is XLIFF specific. In
>> other words, valid syntax can give the wrong information for reasons they
>> haven't foreseen and they should double check. That's all..
>>
>>
>> Dr. David Filip
>>
>> ===========
>>
>> OASIS XLIFF OMOS TC Chair
>>
>> OASIS XLIFF TC Secretary, Editor, Liaison Officer
>>
>> Spokes Research Fellow
>>
>> ADAPT Centre
>>
>> KDEG, Trinity College Dublin
>>
>> Mobile: +420-777-218-122 <+420%20777%20218%20122>
>>
>>
>>
>>
>>
>> On Wed, May 17, 2017 at 11:08 PM, Yves <
yves@opentag.com> wrote:
>>
>> Hi David, all,
>>
>>
>>
>> There are actually 3 places where that warning exists: For mtConfidence,
>> for taConfidence and for termConfidence.
>>
>>
>>
>> The first problem I have is that I don’t understand it, or at least I’m
>> not sure I understand it correctly. And since I’m not sure I understand it,
>> I can’t provide you with a better wording J
>>
>> The second issue is that--if I understand it correctly--it seems to say:
>> “Your confidence attribute may be in the scope of an annotatorsRef with the
>> proper a xyz data category tool reference, but that reference may have been
>> put there by another tool and has nothing to do with your value.”
>>
>>
>>
>> If this is the meaning of the warning, then I’m not sure I get it: A
>> confidence provider is supposed to maintain the annotatorsRef for the
>> confidence, so it should always make sure the confidence and the
>> tool-reference info are set properly. Is this a warning for the case where
>> a tool does not follow the ITS rules?
>>
>>
>>
>> In my humble opinion: Either the reader understand how annotatorsRef
>> works and doesn’t need the warning, or the reader does not and needs a lot
>> more information and should read the ITS specification.
>>
>>
>>
>> In other words: This is an ITS problem, we should leave ITS education to
>> the ITS specification. I would recommend to keep things simple and just
>> drop the 3 warnings.
>>
>>
>>
>> Cheers,
>>
>> -yves
>>
>>
>>
>> *From:* David Filip [mailto:
david.filip@adaptcentre.ie]
>> *Sent:* Wednesday, May 17, 2017 11:10 AM
>> *To:* Yves Savourel <
ysavourel@enlaso.com>; XLIFF Main List <
>>
xliff@lists.oasis-open.org>
>> *Cc:*
xliff-comment@lists.oasis-open.org>> *Subject:* [xliff-comment] Re: [xliff] csprd03 - remaining inline
>> comment to resolve
>>
>>
>>
>> Hi Yves,
>>
>>
>>
>> would you please advise an improved wording for the warning?
>>
>>
>> Warning
>>
>> This annotation can be syntactically in scope of a relevant
>> its:annotatorsRef
>> <http://docs.oasis-open.org/xliff/xliff-core/v2.1/csprd03/xliff-core-v2.1-csprd03.html#itsm_annotatorsRef>
>> attribute, while it still fails to resolve with the intended value.
>> This can happen if more then one terminology providers were used.
>>
>>
>>
>> It seems fine to me ;-)
>>
>>
>>
>> Cheers and thanks
>>
>> dF
>>
>>
>> Dr. David Filip
>>
>> ===========
>>
>> OASIS XLIFF OMOS TC Chair
>>
>> OASIS XLIFF TC Secretary, Editor, Liaison Officer
>>
>> Spokes Research Fellow
>>
>> ADAPT Centre
>>
>> KDEG, Trinity College Dublin
>>
>> Mobile: +420-777-218-122 <+420%20777%20218%20122>
>>
>>
>>
>>
>>
>> On Tue, May 16, 2017 at 12:09 PM, David Filip <
david.filip@adaptcentre.ie>
>> wrote:
>>
>> Hi Yves,
>>
>>
>>
>> I created a csprd03 issue from this
>>
>>
https://issues.oasis-open.org/browse/XLIFF-53>>
>>
>>
>> The associated thread is
>>
>>
http://markmail.org/thread/m7izjxihm74dfyge>>
>>
>>
>> Cheers
>>
>> dF
>>
>>
>>
>>
>> Dr. David Filip
>>
>> ===========
>>
>> OASIS XLIFF OMOS TC Chair
>>
>> OASIS XLIFF TC Secretary, Editor, Liaison Officer
>>
>> Spokes Research Fellow
>>
>> ADAPT Centre
>>
>> KDEG, Trinity College Dublin
>>
>> Mobile: +420-777-218-122 <+420%20777%20218%20122>
>>
>>
>>
>>
>>
>> On Thu, May 11, 2017 at 4:17 PM, Yves Savourel <
ysavourel@enlaso.com>
>> wrote:
>>
>> Hi all,
>>
>>
>>
>> There is at least one remaining comment in the draft: In section
>> “5.9.7.2.3 ITS Terminology Annotation” we have a warning with an inline
>> comment: “COMMENT: HARD TO UNDERSTAND THE INTENTION OF THE WARNING, NEEDS
>> REWRITING OR SHOULD BE DROPPED.”
>>
>>
>>
>> I guess this need to be resolved before csprd04.
>>
>>
>>
>> Cheers,
>>
>> -yves
>>
>>
>>
>>
>>
>> Yves Savourel
>> Localization Solutions Architect | ENLASO®
>> 4888 Pearl East Circle | Suite 300E | Boulder | Colorado 80301
>> t: 303.945.3759 <(303)%20945-3759> | f: 303.516.1701 <(303)%20516-1701>
>> An ISO 9001:2015 certified company
>>
>>
>>
>> *Confidentiality Notice*
>> The information in this transmittal may be privileged and confidential
>> and is intended only for the recipient(s) listed above. Any review, use,
>> disclosure, distribution or copying of this transmittal, in any form, is
>> prohibited except by or on behalf of the intended recipient. If you have
>> received this transmittal in error, please notify me immediately by reply
>> email and destroy all copies of the transmittal.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>
>