Re: Update to MQM documentation and one question

Hi Dave,

That (obviously) isn't included in my sample declaration and schema, but there is no reason we couldn't add that as an optional section. I would make it optional because, in some cases, you might not know that information (or you might be proactively declaring the categories you use for evaluation but not the method, leaving that to the discretion of the reviewer). But that is obviously highly relevant information and I think it would make a lot of sense to have a way to declare that.

-Arle

On 2013 Jul 4, at 11:00 , Dave Lewis <dave.lewis@cs.tcd.ie> wrote:

> Hi Arle,
> Further on the referenced profile document, would you expect that the content of the referred document should have information on _how_ the quality assessment was conducted, i.e. detailing the guidelines that were followed by a QA worker or the algorithms that were applied by an automated QA tool?
> 
> cheers,
> Dave
> 
> 
> 
> On 02/07/2013 12:38, Arle Lommel wrote:
>> Hi Phil,
>> 
>> No, this would not be the only format for the profileRef, but I would say that it (or at least some version of it) should be the only one for an MQM metric. Maybe we could talk about a more generic mechanism to describe *any* quality profile. I would like to see such a specification.
>> 
>> But what I developed is only for MQM, with no thought for other things.
>> 
>> Best,
>> 
>> -Arle
>> 
>> 
>> On 2013 Jul 2, at 12:54 , Phil Ritchie <philr@vistatec.ie> wrote:
>> 
>>> Arle 
>>> 
>>> I'm happy to join a call. 
>>> 
>>> I agree with the principles that what is at the end of a profileRef should be well-defined (self-describing) and machine readable. Your example looks clear to me. We are not saying that this is the only format though, right? 
>>> 
>>> Phil.
>>> 
>>> 
>>> 
>>> 
>>> 
>>> From:        Arle Lommel <arle.lommel@dfki.de> 
>>> To:        David Lewis <dave.lewis@cs.tcd.ie>, Christian Lieske <christian.lieske@sap.com>, "public-i18n-its-ig@w3.org" <public-i18n-its-ig@w3.org>, Aljoscha Burchardt <aljoscha.burchardt@dfki.de>, Yves Savourel <ysavourel@enlaso.com>, Phil Ritchie <philr@vistatec.ie>, 
>>> Date:        02/07/2013 10:50 
>>> Subject:        Re: Update to MQM documentation and one question 
>>> 
>>> 
>>> 
>>> Hi all,
>>> 
>>> Further to our discussions about MQM and ITS, I have proposed a preliminary XML schema for representing MQM metrics. It is pretty straight-forward and I have put in some internal commenting to explain it:
>>> 
>>> https://dl.dropboxusercontent.com/u/223919/qtlaunchpad/mqmMetric.xsd
>>> 
>>> I have also created a sample metrics definition file containing the ITS 2.0 types plus one user-defined issue type (just to show how MQM can be extended as needed):
>>> 
>>> https://dl.dropboxusercontent.com/u/223919/qtlaunchpad/sampleMetric.xml
>>> 
>>> I put these out as a starting point for a discussion about how to better integrate MQM and ITS 2.0 at the formal level. As you will see in the schema, I have changed the MQM token names to conform to ITS 2.0 locQualityIssueType values where this can be done. It shows the default mapping between the (full) MQM set and ITS 2.0 issue types. It also shows those values that are more (or less) granular than ITS 2.0 in a separate section. The mapping for those values is not shown.
>>> 
>>> I think the next step may be to have a call with interested parties (Yves and Phil, I'm hoping you are interested, so I'm adding you to this mail) to discuss what makes sense and how best to ensure that there is an easy path from existing ITS 2.0 support to MQM support. I hope that this sort of formal representation will help in that discussion by giving a more concrete form to the discussion.
>>> 
>>> Felix, feel free to schedule some time in an upcoming ITS Interest Group meeting when it is appropriate.
>>> 
>>> Best,
>>> 
>>> Arle
>>> 
>>> 
>>> ************************************************************
>>> VistaTEC Ltd. Registered in Ireland 268483. 
>>> Registered Office, VistaTEC House, 700, South Circular Road, 
>>> Kilmainham. Dublin 8. Ireland. 
>>> 
>>> The information contained in this message, including any accompanying 
>>> documents, is confidential and is intended only for the addressee(s). 
>>> The unauthorized use, disclosure, copying, or alteration of this 
>>> message is strictly forbidden. If you have received this message in
>>> error please notify the sender immediately.
>>> ************************************************************
>>> 
>> 
>