Comment on RDF Model Theory

[Jeremy Carroll]

We suggest that the following text is in error:
http://www.w3.org/TR/2004/REC-rdf-mt-20040210/#DTYPEINTERP
[[
If D is a datatype map, a D-interpretation of a vocabulary V is any
rdfs-interpretation I of V union {aaa: < aaa, x > in D for some x }
which satisfies the following extra conditions for every pair < aaa, x >
in D:
]]

we suggest that this text is better:
[[
If D is a datatype map, a D-interpretation of a vocabulary V is any
rdfs-interpretation I of V union {aaa: < aaa, x > in D for some x }
union { "sss"^^aaa : < aaa, x > in D for some x and "sss" in the lexical
space of x }
which satisfies the following extra conditions for every pair < aaa, x >
in D:
]]

This is motivated so that the D-entailments recorded in RDF Test Cases
do in fact hold.
http://www.w3.org/TR/2004/REC-rdf-testcases-20040210/#tc_cert
e.g.
tests
http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/Manifest.rdf#semantic-equivalence-within-type-1
http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/Manifest.rdf#semantic-equivalence-within-type-2

involving files
datatypes/test003a.nt
and
datatypes/test003b.nt

With the currently worded RDF Semantics these entailments do not hold.

Given that the spec is no longer in active development, we would be
happy with a note in the erratum document to this effect, for formal
consideration when this specification is next reviewed by a working group.

Jeremy Carroll
David Turner


--
Hewlett-Packard Limited
registered Office: Cain Road, Bracknell, Berks RG12 1HN
Registered No: 690597 England

[Pat Hayes]

>We suggest that the following text is in error:
>http://www.w3.org/TR/2004/REC-rdf-mt-20040210/#DTYPEINTERP
>[[
>If D is a datatype map, a D-interpretation of a vocabulary V is any
>rdfs-interpretation I of V union {aaa: < aaa, x > in D for some x }
>which satisfies the following extra conditions for every pair < aaa,
>x > in D:
>]]
>
>we suggest that this text is better:
>[[
>If D is a datatype map, a D-interpretation of a vocabulary V is any
>rdfs-interpretation I of V union {aaa: < aaa, x > in D for some x }
>union { "sss"^^aaa : < aaa, x > in D for some x and "sss" in the
>lexical space of x }
>which satisfies the following extra conditions for every pair < aaa, x > in D:
>]]
>
>This is motivated so that the D-entailments recorded in RDF Test
>Cases do in fact hold.
>http://www.w3.org/TR/2004/REC-rdf-testcases-20040210/#tc_cert
>e.g.
>tests
>http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/Manifest.rdf#semantic-equivalence-within-type-1
>http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/Manifest.rdf#semantic-equivalence-within-type-2

As editor of the RDF Semantics document, I concur with the above
change. The text as presently written does not handle cases where the
conclusion of an entailment contains a typed literal which is not in
the premises. There are no such entailments for RDF(S) but they can
occur for OWL. Thanks to Ivan Herman for noticing this.

Pat Hayes
--
---------------------------------------------------------------------
IHMC [phone] or [phone]   home
40 South Alcaniz St. [phone]   office
Pensacola [phone]   fax
FL 32502 [phone]    cell
phayesAT-SIGNihmc.us       http://www.ihmc.us/users/phayes

[Ivan Herman-2]

Thanks Jeremy. I have added the error to:

http://www.w3.org/2001/sw/RDFCore/errata.html

Thanks again.

Ivan

Jeremy Carroll wrote:

>
> We suggest that the following text is in error:
> http://www.w3.org/TR/2004/REC-rdf-mt-20040210/#DTYPEINTERP
> [[
> If D is a datatype map, a D-interpretation of a vocabulary V is any
> rdfs-interpretation I of V union {aaa: < aaa, x > in D for some x }
> which satisfies the following extra conditions for every pair < aaa, x >
> in D:
> ]]
>
> we suggest that this text is better:
> [[
> If D is a datatype map, a D-interpretation of a vocabulary V is any
> rdfs-interpretation I of V union {aaa: < aaa, x > in D for some x }
> union { "sss"^^aaa : < aaa, x > in D for some x and "sss" in the lexical
> space of x }
> which satisfies the following extra conditions for every pair < aaa, x >
> in D:
> ]]
>
> This is motivated so that the D-entailments recorded in RDF Test Cases
> do in fact hold.
> http://www.w3.org/TR/2004/REC-rdf-testcases-20040210/#tc_cert
> e.g.
> tests
> http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/Manifest.rdf#semantic-equivalence-within-type-1
>
> http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/Manifest.rdf#semantic-equivalence-within-type-2
>
>
> involving files
> datatypes/test003a.nt
> and
> datatypes/test003b.nt
>
> With the currently worded RDF Semantics these entailments do not hold.
>
> Given that the spec is no longer in active development, we would be
> happy with a note in the erratum document to this effect, for formal
> consideration when this specification is next reviewed by a working group.
>
> Jeremy Carroll
> David Turner
>
>

--

Ivan Herman, W3C Semantic Web Activity Lead
URL: http://www.w3.org/People/Ivan/
PGP Key: http://www.cwi.nl/%7Eivan/AboutMe/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf

[Jeremy Carroll]

Thank you

Jeremy

Ivan Herman wrote:

> Thanks Jeremy. I have added the error to:
>
> http://www.w3.org/2001/sw/RDFCore/errata.html
>
> Thanks again.
>
> Ivan
>
> Jeremy Carroll wrote:
>> We suggest that the following text is in error:
>> http://www.w3.org/TR/2004/REC-rdf-mt-20040210/#DTYPEINTERP
>> [[
>> If D is a datatype map, a D-interpretation of a vocabulary V is any
>> rdfs-interpretation I of V union {aaa: < aaa, x > in D for some x }
>> which satisfies the following extra conditions for every pair < aaa, x >
>> in D:
>> ]]
>>
>> we suggest that this text is better:
>> [[
>> If D is a datatype map, a D-interpretation of a vocabulary V is any
>> rdfs-interpretation I of V union {aaa: < aaa, x > in D for some x }
>> union { "sss"^^aaa : < aaa, x > in D for some x and "sss" in the lexical
>> space of x }
>> which satisfies the following extra conditions for every pair < aaa, x >
>> in D:
>> ]]
>>
>> This is motivated so that the D-entailments recorded in RDF Test Cases
>> do in fact hold.
>> http://www.w3.org/TR/2004/REC-rdf-testcases-20040210/#tc_cert
>> e.g.
>> tests
>> http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/Manifest.rdf#semantic-equivalence-within-type-1
>>
>> http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/Manifest.rdf#semantic-equivalence-within-type-2
>>
>>
>> involving files
>> datatypes/test003a.nt
>> and
>> datatypes/test003b.nt
>>
>> With the currently worded RDF Semantics these entailments do not hold.
>>
>> Given that the spec is no longer in active development, we would be
>> happy with a note in the erratum document to this effect, for formal
>> consideration when this specification is next reviewed by a working group.
>>
>> Jeremy Carroll
>> David Turner
>>
>>
>

--
Hewlett-Packard Limited
registered Office: Cain Road, Bracknell, Berks RG12 1HN
Registered No: 690597 England

[Jeremy Carroll]

Oh, one other thing, the errata page says near the top:
"approved fixes"

Given that (to my knowledge) no one is currently chartered to approve
any fixes, this could lead to the state where no errata are possible.
Or perhaps, in the absence of a WG, it's part of your job to approve
fixes (this part of the process is not clear to me), in which case the
page in its current state is accurate.

I thought DanBri's comment
This is a draft errata note, for review. ----danbri
was appropriate and helpful, and a similar comment would more accurately
reflect the status of the latest erratum.

Jeremy

Ivan Herman wrote:

> Thanks Jeremy. I have added the error to:
>
> http://www.w3.org/2001/sw/RDFCore/errata.html
>
> Thanks again.
>
> Ivan
>
> Jeremy Carroll wrote:
>> We suggest that the following text is in error:
>> http://www.w3.org/TR/2004/REC-rdf-mt-20040210/#DTYPEINTERP
>> [[
>> If D is a datatype map, a D-interpretation of a vocabulary V is any
>> rdfs-interpretation I of V union {aaa: < aaa, x > in D for some x }
>> which satisfies the following extra conditions for every pair < aaa, x >
>> in D:
>> ]]
>>
>> we suggest that this text is better:
>> [[
>> If D is a datatype map, a D-interpretation of a vocabulary V is any
>> rdfs-interpretation I of V union {aaa: < aaa, x > in D for some x }
>> union { "sss"^^aaa : < aaa, x > in D for some x and "sss" in the lexical
>> space of x }
>> which satisfies the following extra conditions for every pair < aaa, x >
>> in D:
>> ]]
>>
>> This is motivated so that the D-entailments recorded in RDF Test Cases
>> do in fact hold.
>> http://www.w3.org/TR/2004/REC-rdf-testcases-20040210/#tc_cert
>> e.g.
>> tests
>> http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/Manifest.rdf#semantic-equivalence-within-type-1
>>
>> http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/Manifest.rdf#semantic-equivalence-within-type-2
>>
>>
>> involving files
>> datatypes/test003a.nt
>> and
>> datatypes/test003b.nt
>>
>> With the currently worded RDF Semantics these entailments do not hold.
>>
>> Given that the spec is no longer in active development, we would be
>> happy with a note in the erratum document to this effect, for formal
>> consideration when this specification is next reviewed by a working group.
>>
>> Jeremy Carroll
>> David Turner
>>
>>
>

--
Hewlett-Packard Limited
registered Office: Cain Road, Bracknell, Berks RG12 1HN
Registered No: 690597 England

[Ivan Herman-2]

Good point.

I will try to clarify this with Ian and Ralph. In the meantime, I have
copy pasted DanBri's remark (changing it to my name).

Ivan

Jeremy Carroll wrote:

>
> Oh, one other thing, the errata page says near the top:
> "approved fixes"
>
> Given that (to my knowledge) no one is currently chartered to approve
> any fixes, this could lead to the state where no errata are possible.
> Or perhaps, in the absence of a WG, it's part of your job to approve
> fixes (this part of the process is not clear to me), in which case the
> page in its current state is accurate.
>
> I thought DanBri's comment
> This is a draft errata note, for review. ----danbri
> was appropriate and helpful, and a similar comment would more accurately
> reflect the status of the latest erratum.
>
> Jeremy
>
> Ivan Herman wrote:
>> Thanks Jeremy. I have added the error to:
>>
>> http://www.w3.org/2001/sw/RDFCore/errata.html
>>
>> Thanks again.
>>
>> Ivan
>>
>> Jeremy Carroll wrote:
>>> We suggest that the following text is in error:
>>> http://www.w3.org/TR/2004/REC-rdf-mt-20040210/#DTYPEINTERP
>>> [[
>>> If D is a datatype map, a D-interpretation of a vocabulary V is any
>>> rdfs-interpretation I of V union {aaa: < aaa, x > in D for some x }
>>> which satisfies the following extra conditions for every pair < aaa, x >
>>> in D:
>>> ]]
>>>
>>> we suggest that this text is better:
>>> [[
>>> If D is a datatype map, a D-interpretation of a vocabulary V is any
>>> rdfs-interpretation I of V union {aaa: < aaa, x > in D for some x }
>>> union { "sss"^^aaa : < aaa, x > in D for some x and "sss" in the lexical
>>> space of x }
>>> which satisfies the following extra conditions for every pair < aaa, x >
>>> in D:
>>> ]]
>>>
>>> This is motivated so that the D-entailments recorded in RDF Test Cases
>>> do in fact hold.
>>> http://www.w3.org/TR/2004/REC-rdf-testcases-20040210/#tc_cert
>>> e.g.
>>> tests
>>> http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/Manifest.rdf#semantic-equivalence-within-type-1
>>>
>>>
>>> http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/Manifest.rdf#semantic-equivalence-within-type-2
>>>
>>>
>>>
>>> involving files
>>> datatypes/test003a.nt
>>> and
>>> datatypes/test003b.nt
>>>
>>> With the currently worded RDF Semantics these entailments do not hold.
>>>
>>> Given that the spec is no longer in active development, we would be
>>> happy with a note in the erratum document to this effect, for formal
>>> consideration when this specification is next reviewed by a working
>>> group.
>>>
>>> Jeremy Carroll
>>> David Turner
>>>
>>>
>>
>

--

Ivan Herman, W3C Semantic Web Activity Lead
URL: http://www.w3.org/People/Ivan/
PGP Key: http://www.cwi.nl/%7Eivan/AboutMe/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf

[Ivan Herman-2]

Jeremy,

thanks again for point this out. I have talked to Ian, and we agreed to
change the status section of that document because the wording was
indeed inappropriate. Have a look at it to see if it is clearer now...

Thanks again

Ivan

Jeremy Carroll wrote:

>
> Oh, one other thing, the errata page says near the top:
> "approved fixes"
>
> Given that (to my knowledge) no one is currently chartered to approve
> any fixes, this could lead to the state where no errata are possible.
> Or perhaps, in the absence of a WG, it's part of your job to approve
> fixes (this part of the process is not clear to me), in which case the
> page in its current state is accurate.
>
> I thought DanBri's comment
> This is a draft errata note, for review. ----danbri
> was appropriate and helpful, and a similar comment would more accurately
> reflect the status of the latest erratum.
>
> Jeremy
>
> Ivan Herman wrote:
>> Thanks Jeremy. I have added the error to:
>>
>> http://www.w3.org/2001/sw/RDFCore/errata.html
>>
>> Thanks again.
>>
>> Ivan
>>
>> Jeremy Carroll wrote:
>>> We suggest that the following text is in error:
>>> http://www.w3.org/TR/2004/REC-rdf-mt-20040210/#DTYPEINTERP
>>> [[
>>> If D is a datatype map, a D-interpretation of a vocabulary V is any
>>> rdfs-interpretation I of V union {aaa: < aaa, x > in D for some x }
>>> which satisfies the following extra conditions for every pair < aaa, x >
>>> in D:
>>> ]]
>>>
>>> we suggest that this text is better:
>>> [[
>>> If D is a datatype map, a D-interpretation of a vocabulary V is any
>>> rdfs-interpretation I of V union {aaa: < aaa, x > in D for some x }
>>> union { "sss"^^aaa : < aaa, x > in D for some x and "sss" in the lexical
>>> space of x }
>>> which satisfies the following extra conditions for every pair < aaa, x >
>>> in D:
>>> ]]
>>>
>>> This is motivated so that the D-entailments recorded in RDF Test Cases
>>> do in fact hold.
>>> http://www.w3.org/TR/2004/REC-rdf-testcases-20040210/#tc_cert
>>> e.g.
>>> tests
>>> http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/Manifest.rdf#semantic-equivalence-within-type-1
>>>
>>>
>>> http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/Manifest.rdf#semantic-equivalence-within-type-2
>>>
>>>
>>>
>>> involving files
>>> datatypes/test003a.nt
>>> and
>>> datatypes/test003b.nt
>>>
>>> With the currently worded RDF Semantics these entailments do not hold.
>>>
>>> Given that the spec is no longer in active development, we would be
>>> happy with a note in the erratum document to this effect, for formal
>>> consideration when this specification is next reviewed by a working
>>> group.
>>>
>>> Jeremy Carroll
>>> David Turner
>>>
>>>
>>
>

--

Ivan Herman, W3C Semantic Web Activity Lead
URL: http://www.w3.org/People/Ivan/
PGP Key: http://www.cwi.nl/%7Eivan/AboutMe/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf

[Jeremy Carroll]

Far exceeds my expectations. Brilliant.

Jeremy

Ivan Herman wrote:

> Jeremy,
>
> thanks again for point this out. I have talked to Ian, and we agreed to
> change the status section of that document because the wording was
> indeed inappropriate. Have a look at it to see if it is clearer now...
>
> Thanks again
>
> Ivan
>
> Jeremy Carroll wrote:
>> Oh, one other thing, the errata page says near the top:
>> "approved fixes"
>>
>> Given that (to my knowledge) no one is currently chartered to approve
>> any fixes, this could lead to the state where no errata are possible.
>> Or perhaps, in the absence of a WG, it's part of your job to approve
>> fixes (this part of the process is not clear to me), in which case the
>> page in its current state is accurate.
>>
>> I thought DanBri's comment
>> This is a draft errata note, for review. ----danbri
>> was appropriate and helpful, and a similar comment would more accurately
>> reflect the status of the latest erratum.
>>
>> Jeremy
>>
>> Ivan Herman wrote:
>>> Thanks Jeremy. I have added the error to:
>>>
>>> http://www.w3.org/2001/sw/RDFCore/errata.html
>>>
>>> Thanks again.
>>>
>>> Ivan
>>>
>>> Jeremy Carroll wrote:
>>>> We suggest that the following text is in error:
>>>> http://www.w3.org/TR/2004/REC-rdf-mt-20040210/#DTYPEINTERP
>>>> [[
>>>> If D is a datatype map, a D-interpretation of a vocabulary V is any
>>>> rdfs-interpretation I of V union {aaa: < aaa, x > in D for some x }
>>>> which satisfies the following extra conditions for every pair < aaa, x >
>>>> in D:
>>>> ]]
>>>>
>>>> we suggest that this text is better:
>>>> [[
>>>> If D is a datatype map, a D-interpretation of a vocabulary V is any
>>>> rdfs-interpretation I of V union {aaa: < aaa, x > in D for some x }
>>>> union { "sss"^^aaa : < aaa, x > in D for some x and "sss" in the lexical
>>>> space of x }
>>>> which satisfies the following extra conditions for every pair < aaa, x >
>>>> in D:
>>>> ]]
>>>>
>>>> This is motivated so that the D-entailments recorded in RDF Test Cases
>>>> do in fact hold.
>>>> http://www.w3.org/TR/2004/REC-rdf-testcases-20040210/#tc_cert
>>>> e.g.
>>>> tests
>>>> http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/Manifest.rdf#semantic-equivalence-within-type-1
>>>>
>>>>
>>>> http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/Manifest.rdf#semantic-equivalence-within-type-2
>>>>
>>>>
>>>>
>>>> involving files
>>>> datatypes/test003a.nt
>>>> and
>>>> datatypes/test003b.nt
>>>>
>>>> With the currently worded RDF Semantics these entailments do not hold.
>>>>
>>>> Given that the spec is no longer in active development, we would be
>>>> happy with a note in the erratum document to this effect, for formal
>>>> consideration when this specification is next reviewed by a working
>>>> group.
>>>>
>>>> Jeremy Carroll
>>>> David Turner
>>>>
>>>>
>

--
Hewlett-Packard Limited
registered Office: Cain Road, Bracknell, Berks RG12 1HN
Registered No: 690597 England