W3C home > Mailing lists > Public > public-media-annotation@w3.org > February 2012

Re: XMP example and XMP JSON response files don't seem to be in sync.

From: Florian Stegmaier <stegmai@dimis.fim.uni-passau.de>
Date: Fri, 10 Feb 2012 09:23:32 +0100
Cc: "Bailer, Werner" <werner.bailer@joanneum.at>, "Höffernig, Martin" <Martin.Hoeffernig@joanneum.at>, "tmichel@w3.org" <tmichel@w3.org>, "public-media-annotation@w3.org" <public-media-annotation@w3.org>
Message-Id: <69368F35-667B-4594-A69A-55B16E93B470@dimis.fim.uni-passau.de>
To: Felix Sasaki <fsasaki@w3.org>
Hi Felix!

I can provide the JSON file. I am not quiet sure if a will make it today, but on Monday at latest.

Cheers.
_____________________________
Dipl. Inf. Florian Stegmaier
Chair of Distributed Information Systems
University of Passau
Innstr. 43
94032 Passau

Room 248 ITZ

Tel.: +49 851 509 3063
Fax: +49 851 509 3062

stegmai@dimis.fim.uni-passau.de
https://www.dimis.fim.uni-passau.de/iris/
http://twitter.com/fstegmai
_____________________________

Am 10.02.2012 um 08:47 schrieb Felix Sasaki:

> Werner Thierry, Florian, all,
> 
> I have implemented the identifier issue that Martin found and that Werner explained to me. At Thierry. the updated files are at 
> 
> http://fabday.fh-potsdam.de/~sasaki/mawg/xmp/xslt/
> 
> http://fabday.fh-potsdam.de/~sasaki/mawg/xmp/xslt/MAWG_test_canonical.xml (original XMP, not changed)
> http://fabday.fh-potsdam.de/~sasaki/mawg/xmp/xslt/xmp-canonical-mapped-to-mawg-by-xslt.rdf (new RDF)
> http://fabday.fh-potsdam.de/~sasaki/mawg/xmp/xslt/xmp-canonical-mapped-to-mawg-by-xslt.rdf (ttl version of RDF file)
> 
> I have not re-generated the JSON file. Florian, can you do that?
> 
> Thierry, about
> 
> "Please also send me the XMP atomic test for following properties
> location, keywork, genre,rating, policy, framesze, compression,duration,
> samplingrate, frame rate numtraks
> 
> X8,X10, X11,X12,X16,X21,X22, X23,X25, X26,,X28,
> 
> (marked with @@ in the API testsuite.")
> 
> If I understand you correctly you want to have the input data for the tests (that is an XMP file that contains just the data relevant for the test) and the JSON response, correct?
> 
> I am in a meeting the whole day and then will drive to Prague tonight. So it's likely that I get to the atomic tests next week. Anyway, it would be great to have first the JSON re-generated from Florian so that I'm sure that I am using the right JSON for the atomic tests.
> 
> Regards,
> 
> Felix
> 
> 2012/2/2 Bailer, Werner <werner.bailer@joanneum.at>
> Felix, to clarify this: ma:identifer is part of the spec, but there is no corresponding element in the RDF vocabulary. The reference [1] cited by Martin states:
> 
>  
> 
> “(1) The identifier of a media resource is represented in RDF by the URI of the node representing that media resource. If a resource is identified by several URI, owl:sameAs should be used.”
> 
>  
> 
> The identifier can be expressed by rdf:about on the ma:MediaResource node (see other examples, e.g. the mapping of dc:identifier). As the note above states, if you have several identifiers, use owl:sameAs (see e.g. the MPEG-7 example).
> 
>  
> 
> Best regards,
> Werner
> 
>  
> 
> [1] http://www.w3.org/TR/2011/PR-mediaont-10-20111129/#correspondance-id1
> 
>  
> 
> Von: Felix Sasaki [mailto:fsasaki@w3.org] 
> Gesendet: Donnerstag, 02. Februar 2012 13:42
> An: Höffernig, Martin
> Cc: Florian Stegmaier; tmichel@w3.org; Bailer, Werner; public-media-annotation@w3.org
> 
> 
> Betreff: Re: XMP example and XMP JSON response files don't seem to be in sync.
> 
>  
> 
> Just to make sure I understand: ma:identifier is at
> 
> http://www.w3.org/TR/2011/PR-mediaont-10-20111129/#core-property-definitions
> 
> and in the mapping tables e.g. at
> 
> http://www.w3.org/TR/2011/PR-mediaont-10-20111129/#xmp-table
> 
> but not part of the spec?
> 
>  
> 
> I am happy to remove ma:identifier from the XMP example(s), but want to be sure I am doing the right thing.
> 
>  
> 
> Felix
> 
>  
> 
> 2012/2/2 Höffernig, Martin <Martin.Hoeffernig@joanneum.at>
> 
> Dear Felix, all,
> 
> I have checked the RDF and TTL version of the XMP example.
> Both ontology files (RDF and TTL version) contain the property ma:identifier in connection with 3 different identifier values (xmp:Identifier 1, xmp:Identifier 2, xmp:Identifier 3).
> However, property ma:identifier is no longer part of ma ontology spec [1].
> 
> [1]: http://www.w3.org/TR/2011/PR-mediaont-10-20111129/#correspondance-id1
> 
> Best,
> Martin
> 
> 
> > -----Ursprüngliche Nachricht-----
> > Von: Florian Stegmaier [mailto:stegmai@dimis.fim.uni-passau.de]
> > Gesendet: Donnerstag, 02. Februar 2012 10:40
> > An: tmichel@w3.org
> > Cc: Felix Sasaki; Bailer, Werner; public-media-annotation@w3.org
> 
> > Betreff: Re: XMP example and XMP JSON response files don't seem to be
> > in sync.
> >
> 
> > Hi Thierry, all,
> >
> > at the moment, Martin is checking the RDF file. In case the RDF file is
> > correct (which i think it will be), i will check the JSON response, if
> > it fits to the current RDF version. I will send you my findings later.
> >
> > Cheers.
> > _____________________________
> > Dipl. Inf. Florian Stegmaier
> > Chair of Distributed Information Systems University of Passau Innstr.
> > 43
> > 94032 Passau
> >
> > Room 248 ITZ
> >
> > Tel.: +49 851 509 3063
> > Fax: +49 851 509 3062
> >
> > stegmai@dimis.fim.uni-passau.de
> > https://www.dimis.fim.uni-passau.de/iris/
> > http://twitter.com/fstegmai
> > _____________________________
> >
> > Am 02.02.2012 um 10:38 schrieb Thierry MICHEL:
> >
> > > Felix,
> > >
> > > I have republished the  XMP .rdf and .ttl file from
> > > http://fabday.fh-potsdam.de/~sasaki/mawg/xmp/xslt/xmp-canonical-
> > mapped
> > > -to-mawg-by-xslt.ttl
> > > and
> > > http://fabday.fh-potsdam.de/~sasaki/mawg/xmp/xslt/xmp-canonical-
> > mapped
> > > -to-mawg-by-xslt.rdf
> > >
> > > to
> > >
> > http://www.w3.org/2008/WebVideo/Annotations/drafts/metadata_formats/xm
> > > p-canonical-mapped-to-mawg.ttl
> > >
> > http://www.w3.org/2008/WebVideo/Annotations/drafts/metadata_formats/xm
> > > p-canonical-mapped-to-mawg.rdf
> > >
> > > these RDf and TTL files seem identical to the former ones.
> > >
> > >
> > > I hope it is also the case for the corresponding JSON response
> > >
> > http://www.w3.org/2008/WebVideo/Annotations/drafts/API10/JSON/normativ
> > > e_json_ma_xmp.json
> > >
> > > Because I have used this file to make the testsuite ...
> > >
> > > Thierry.
> > >
> > >
> > >
> > > Le 02/02/2012 09:31, Felix Sasaki a écrit :
> > >> Thierry, Florian, all,
> > >>
> > >> I have generated the .rdf and .ttl file from the .xml file
> > >> automatically, these are definitely in sync. Florian, could you
> > >> re-generate the JSON file from
> > >>  <http://fabday.fh-potsdam.de/%**7Esasaki/mawg/xmp/xslt/xmp-**
> > >> canonical-mapped-to-mawg-by-**xslt.rdf<http://fabday.fh-
> > potsdam.de/~s
> > >> asaki/mawg/xmp/xslt/xmp-canonical-mapped-to-mawg-by-xslt.rdf>
> > >>>
> > >> ?
> > >>
> > >> Then we are sure that everything is OK.
> > >>
> > >> Felix
> > >>
> > >> 2012/1/31 Thierry Michel<tmichel@w3.org>
> > >>
> > >>> Felix,
> > >>>
> > >>> could you please check that the 4 files are indeed in sync
> > >>>
> > >>>
> > >>> http://www.w3.org/2008/**WebVideo/Annotations/drafts/**
> > >>> metadata_formats/xmp-**canonical-
> > original.xml<http://www.w3.org/2008
> > >>> /WebVideo/Annotations/drafts/metadata_formats/xmp-canonical-
> > original
> > >>> .xml><
> > >>> http://fabday.fh-potsdam.de/%**7Esasaki/mawg/xmp/xslt/MAWG_**
> > >>> test_canonical.xml<http://fabday.fh-
> > potsdam.de/%7Esasaki/mawg/xmp/xs
> > >>> lt/MAWG_test_canonical.xml>
> > >>>>
> > >>>
> > >>> http://www.w3.org/2008/**WebVideo/Annotations/drafts/**
> > >>> metadata_formats/xmp-**canonical-mapped-to-
> > mawg.rdf<http://www.w3.or
> > >>> g/2008/WebVideo/Annotations/drafts/metadata_formats/xmp-canonical-
> > ma
> > >>> pped-to-mawg.rdf><
> > >>> http://fabday.fh-potsdam.de/%**7Esasaki/mawg/xmp/xslt/xmp-**
> > >>> canonical-mapped-to-mawg-by-**xslt.rdf<http://fabday.fh-
> > potsdam.de/%
> > >>> 7Esasaki/mawg/xmp/xslt/xmp-canonical-mapped-to-mawg-by-xslt.rdf>
> > >>>>
> > >>>
> > >>> http://www.w3.org/2008/**WebVideo/Annotations/drafts/**
> > >>> metadata_formats/xmp-**canonical-mapped-to-
> > mawg.ttl<http://www.w3.or
> > >>> g/2008/WebVideo/Annotations/drafts/metadata_formats/xmp-canonical-
> > ma
> > >>> pped-to-mawg.ttl><
> > >>> http://fabday.fh-potsdam.de/%**7Esasaki/mawg/xmp/xslt/xmp-**
> > >>> canonical-mapped-to-mawg-by-**xslt.ttl<http://fabday.fh-
> > potsdam.de/%
> > >>> 7Esasaki/mawg/xmp/xslt/xmp-canonical-mapped-to-mawg-by-xslt.ttl>
> > >>>>
> > >>>
> > >>> http://www.w3.org/2008/**WebVideo/Annotations/drafts/**
> > >>>
> > API10/JSON/normative_json_ma_**xmp.json<http://www.w3.org/2008/WebVi
> > >>> deo/Annotations/drafts/API10/JSON/normative_json_ma_xmp.json>
> > >>>
> > >>> Thanks,
> > >>> I would not want to have to redo the alll test suite ...
> > >>>
> > >>> Thierry
> > >>>
> > >>>
> > >>>
> > >>> Felix Sasaki a écrit :
> > >>>
> > >>>>
> > >>>>
> > >>>> 2012/1/31 Thierry Michel<tmichel@w3.org<mailto:tmichel@w3.org>>
> > >>>>
> > >>>>
> > >>>>    Felix,
> > >>>>
> > >>>>    I have just published the RDF file you have sent.
> > >>>>    http://www.w3.org/2008/**WebVideo/Annotations/drafts/**
> > >>>> metadata_formats/xmp-**canonical-
> > original.xml<http://www.w3.org/200
> > >>>> 8/WebVideo/Annotations/drafts/metadata_formats/xmp-canonical-
> > origin
> > >>>> al.xml>
> > >>>>
> > >>>>    But I am getting lost now with your new wording.
> > >>>>
> > >>>>    Could you please send me the apropriate 4 files in sync ASAP
> > >>>>
> > >>>>    XMP- xml
> > >>>>
> > >>>>
> > >>>> See http://fabday.fh-potsdam.de/~**sasaki/mawg/xmp/xslt/MAWG_**
> > >>>> test_canonical.xml<http://fabday.fh-
> > potsdam.de/~sasaki/mawg/xmp/xsl
> > >>>> t/MAWG_test_canonical.xml><
> > >>>> http://fabday.fh-potsdam.de/%**7Esasaki/mawg/xmp/xslt/MAWG_**
> > >>>> test_canonical.xml<http://fabday.fh-
> > potsdam.de/%7Esasaki/mawg/xmp/x
> > >>>> slt/MAWG_test_canonical.xml>
> > >>>>>
> > >>>>
> > >>>>    XMP- rdf
> > >>>>
> > >>>>
> > >>>> See http://fabday.fh-potsdam.de/~**sasaki/mawg/xmp/xslt/xmp-**
> > >>>> canonical-mapped-to-mawg-by-**xslt.rdf<http://fabday.fh-
> > potsdam.de/
> > >>>> ~sasaki/mawg/xmp/xslt/xmp-canonical-mapped-to-mawg-by-xslt.rdf><
> > >>>> http://fabday.fh-potsdam.de/%**7Esasaki/mawg/xmp/xslt/xmp-**
> > >>>> canonical-mapped-to-mawg-by-**xslt.rdf<http://fabday.fh-
> > potsdam.de/
> > >>>> %7Esasaki/mawg/xmp/xslt/xmp-canonical-mapped-to-mawg-by-xslt.rdf>
> > >>>>>
> > >>>>
> > >>>>    XMP- ttl
> > >>>>
> > >>>>
> > >>>> See http://fabday.fh-potsdam.de/~**sasaki/mawg/xmp/xslt/xmp-**
> > >>>> canonical-mapped-to-mawg-by-**xslt.ttl<http://fabday.fh-
> > potsdam.de/
> > >>>> ~sasaki/mawg/xmp/xslt/xmp-canonical-mapped-to-mawg-by-xslt.ttl><
> > >>>> http://fabday.fh-potsdam.de/%**7Esasaki/mawg/xmp/xslt/xmp-**
> > >>>> canonical-mapped-to-mawg-by-**xslt.ttl<http://fabday.fh-
> > potsdam.de/
> > >>>> %7Esasaki/mawg/xmp/xslt/xmp-canonical-mapped-to-mawg-by-xslt.ttl>
> > >>>>>
> > >>>>
> > >>>>
> > >>>>    XMP- JSON
> > >>>>
> > >>>>
> > >>>> As Florian pointed out, this file has not changed. It has been
> > >>>> generated by Florian from the RDF, so I cannot re-generate it.
> > >>>>
> > >>>> Best,
> > >>>>
> > >>>> Felix
> > >>>>
> > >>>>
> > >>>>    I am working on the implementation report and I must have the
> > >>>>    final files to work on these.
> > >>>>
> > >>>>    thanks you.
> > >>>>
> > >>>>    thierry
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>>    Felix Sasaki a écrit :
> > >>>>
> > >>>>        Yes indeed :)
> > >>>>
> > >>>>        Felix
> > >>>>
> > >>>>        2012/1/31 Florian Stegmaier<stegmai@dimis.fim.uni-
> > passau.**de<stegmai@dimis.fim.uni-passau.de>
> > >>>>
> > >>>> <mailto:stegmai@dimis.fim.uni-**passau.de<stegmai@dimis.fim.uni-
> > pas
> > >>>> sau.de>
> > >>>>>
> > >>>>
> > >>>> <mailto:stegmai@dimis.fim.uni-**passau.de<stegmai@dimis.fim.uni-
> > pas
> > >>>> sau.de>
> > >>>>
> > >>>>
> > >>>> <mailto:stegmai@dimis.fim.uni-**passau.de<stegmai@dimis.fim.uni-
> > pas
> > >>>> sau.de>
> > >>>>>>>
> > >>>>
> > >>>>
> > >>>>           Hi Felix, all!
> > >>>>
> > >>>>           In that case, Thierry should use the file as it was ;)
> > >>>>
> > >>>>           Cheers.
> > >>>>           _____________________________
> > >>>>           Dipl. Inf. Florian Stegmaier
> > >>>>           Chair of Distributed Information Systems
> > >>>>           University of Passau
> > >>>>           Innstr. 43
> > >>>>           94032 Passau
> > >>>>
> > >>>>           Room 248 ITZ
> > >>>>
> > >>>>           Tel.: +49 851 509 3063<tel:%2B49%20851%20509%203063>
> > >>>>        <tel:%2B49%20851%20509%203063>
> > >>>>           Fax: +49 851 509 3062<tel:%2B49%20851%20509%203062>
> > >>>>        <tel:%2B49%20851%20509%203062>
> > >>>>
> > >>>>           stegmai@dimis.fim.uni-passau.**de<stegmai@dimis.fim.uni-
> > passau.de>
> > >>>>
> > >>>> <mailto:stegmai@dimis.fim.uni-**passau.de<stegmai@dimis.fim.uni-
> > pas
> > >>>> sau.de>
> > >>>>>
> > >>>>
> > >>>> <mailto:stegmai@dimis.fim.uni-**passau.de<stegmai@dimis.fim.uni-
> > pas
> > >>>> sau.de>
> > >>>>
> > >>>>
> > >>>> <mailto:stegmai@dimis.fim.uni-**passau.de<stegmai@dimis.fim.uni-
> > pas
> > >>>> sau.de>
> > >>>>>>
> > >>>>
> > >>>>           https://www.dimis.fim.uni-
> > **passau.de/iris/<https://www.dimis.fim.uni-passau.de/iris/>
> > >>>>           http://twitter.com/fstegmai
> > >>>>           _____________________________
> > >>>>
> > >>>>           Am 31.01.2012 um 13:02 schrieb Felix Sasaki:
> > >>>>
> > >>>>           >  This was an error in the XSLT that generates the RDF
> > from the
> > >>>>           original XMP. I have fixed this now, see
> > >>>>           >
> > >>>>           >  http://fabday.fh-
> > potsdam.de/~**sasaki/mawg/xmp/xslt/<http://fabday.fh-
> > potsdam.de/~sasaki/mawg/xmp/xslt/>
> > >>>>
> > >>>> <http://fabday.fh-
> > potsdam.de/%**7Esasaki/mawg/xmp/xslt/<http://fabd
> > >>>> ay.fh-potsdam.de/%7Esasaki/mawg/xmp/xslt/>
> > >>>>>
> > >>>>
> > >>>> <http://fabday.fh-
> > potsdam.de/%**7Esasaki/mawg/xmp/xslt/<http://fabd
> > >>>> ay.fh-potsdam.de/%7Esasaki/mawg/xmp/xslt/>
> > >>>>>
> > >>>>           >
> > >>>>
> > >>>> http://fabday.fh-potsdam.de/~**sasaki/mawg/xmp/xslt/xmp-
> > >>>> **canonical-mapped-to-mawg-by-**xslt.rdf<http://fabday.fh-
> > potsdam.de/~sasaki/mawg/xmp/xslt/xmp-canonical-mapped-to-mawg-by-
> > xslt.rdf>
> > >>>>
> > >>>> <http://fabday.fh-potsdam.de/%**7Esasaki/mawg/xmp/xslt/xmp-**
> > >>>> canonical-mapped-to-mawg-by-**xslt.rdf<http://fabday.fh-
> > potsdam.de/
> > >>>> %7Esasaki/mawg/xmp/xslt/xmp-canonical-mapped-to-mawg-by-xslt.rdf>
> > >>>>>
> > >>>>                  <http://fabday.fh-potsdam.de/%**
> > >>>> 7Esasaki/mawg/xmp/xslt/xmp-**canonical-mapped-to-mawg-by-
> > **xslt.rdf
> > >>>> <http://fabday.fh-potsdam.de/%7Esasaki/mawg/xmp/xslt/xmp-
> > canonical-
> > >>>> mapped-to-mawg-by-xslt.rdf>
> > >>>>>
> > >>>>
> > >>>>
> > >>>>           >
> > >>>>           >  Florian, I guess you when need to regenerate the JSON
> > >>>>        from above
> > >>>>           RDF.
> > >>>>           >
> > >>>>           >  Felix
> > >>>>           >
> > >>>>           >  2012/1/31 Bailer, Werner<werner.bailer@joanneum.at
> > >>>>
> > <mailto:werner.bailer@**joanneum.at<werner.bailer@joanneum.at>>
> > >>>>
> > >>>> <mailto:werner.bailer@**joanneum.at<werner.bailer@joanneum.at>
> > >>>>
> > >>>>
> > >>>> <mailto:werner.bailer@**joanneum.at<werner.bailer@joanneum.at>>>>
> > >>>>
> > >>>>           >  For some reason, the additional identifiers from the
> > original
> > >>>>           XMP files were dropped in the RDF rather than having
> > multiple
> > >>>>           identifiers, there is a comment instead
> > >>>>        <!--xmp:Identifier-->  in
> > >>>>           the RDF.
> > >>>>           >
> > >>>>           >  Best regards,
> > >>>>           >  Werner
> > >>>>           >
> > >>>>           >  >  -----Ursprüngliche Nachricht-----
> > >>>>           >  >  Von: Florian Stegmaier
> > >>>>        [mailto:stegmai@dimis.fim.uni-
> > **passau.de<stegmai@dimis.fim.uni-passau.de>
> > >>>>
> > >>>> <mailto:stegmai@dimis.fim.uni-**passau.de<stegmai@dimis.fim.uni-
> > pas
> > >>>> sau.de>
> > >>>>>
> > >>>>           <mailto:stegmai@dimis.fim.uni-
> > **passau.de<stegmai@dimis.fim.uni-passau.de>
> > >>>>
> > >>>> <mailto:stegmai@dimis.fim.uni-**passau.de<stegmai@dimis.fim.uni-
> > pas
> > >>>> sau.de>
> > >>>>>> ]
> > >>>>           >  >  Gesendet: Dienstag, 31. Jänner 2012 12:27
> > >>>>           >  >  An: tmichel@w3.org<mailto:tmichel@w3.org>
> > >>>>        <mailto:tmichel@w3.org<mailto:tmichel@w3.org>>
> > >>>>
> > >>>>           >  >  Cc: Felix Sasaki; public-media-annotation@w3.org
> > >>>>
> > >>>> <mailto:public-media-**annotation@w3.org<public-media-
> > annotation@w3
> > >>>> .org>
> > >>>>>
> > >>>>
> > >>>> <mailto:public-media-**annotation@w3.org<public-media-
> > annotation@w3
> > >>>> .org>
> > >>>>
> > >>>>
> > >>>> <mailto:public-media-**annotation@w3.org<public-media-
> > annotation@w3
> > >>>> .org>
> > >>>>>>
> > >>>>
> > >>>>           >  >  Betreff: Re: XMP example and XMP JSON response
> > files don't
> > >>>>           seem to be in
> > >>>>           >  >  sync.
> > >>>>           >  >
> > >>>>           >  >  Hi Thierry,
> > >>>>           >  >
> > >>>>           >  >  yes, i have already told you that this is is a
> > relict
> > >>>>        before
> > >>>>           our decision to use
> > >>>>           >  >  the RDF data set as basis. Please use the attached
> > >>>>        version of
> > >>>>           the JSON
> > >>>>           >  >  instead. In any case, the information of the RDF
> > shall be
> > >>>>           reflected in the
> > >>>>           >  >  JSON.
> > >>>>           >  >
> > >>>>           >  >  Are there also difficulties because of *-a/*-b
> > test cases?
> > >>>>           >  >
> > >>>>           >  >  Cheers.
> > >>>>           >
> > >>>>           >
> > >>>>           >
> > >>>>           >  --
> > >>>>           >  Felix Sasaki
> > >>>>           >  DFKI / W3C Fellow
> > >>>>           >
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>>        --         Felix Sasaki
> > >>>>        DFKI / W3C Fellow
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>> --
> > >>>> Felix Sasaki
> > >>>> DFKI / W3C Fellow
> > >>>>
> > >>>>
> > >>>
> > >>
> > >>
> >
> 
> 
> 
> 
>  
> 
> -- 
> Felix Sasaki
> 
> DFKI / W3C Fellow
> 
>  
> 
> 
> 
> 
> -- 
> Felix Sasaki
> DFKI / W3C Fellow
> 
Received on Friday, 10 February 2012 08:24:02 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 10 February 2012 08:24:03 GMT