- From: エリクソン トーレ <t-eriksson@so.taisho.co.jp>
- Date: Thu, 28 Mar 2013 23:46:26 +0000
- To: "'Jim McCusker'" <mccusj@rpi.edu>, Pat Hayes <phayes@ihmc.us>
- CC: Rafael Richards <rafaelrichards@jhu.edu>, Oliver Ruebenacker <curoli@gmail.com>, David Booth <david@dbooth.org>, "<public-semweb-lifesci@w3.org>" <public-semweb-lifesci@w3.org>
On Wed, Mar 27, 2013 at 8:30 PM, Jim McCusker <mccusj@rpi.edu> wrote:
>
> On Wed, Mar 27, 2013 at 8:26 PM, Pat Hayes <phayes@ihmc.us> wrote:
>>
>> That "option" is already available, if you use owl:sameAs
>> correctly (and do not confuse information about some thing with
>> meta-information about that information. The meta-information is not about
>> the thing.
>
> If only owl:sameAs were used correctly...
>
> One thing to note is that provenance is for more than data. The issues I
> discuss in that presentation (and the following papers) are more about
> entity provenance than data provenance.
I think focusing on owl:sameAs misses the point. Shouldn't it be
"OWL reasoning (on dirty data) considered harmful". Similar problem
will arise from incorrect use of owl:InverseFunctionalProperty and owl:hasKey.
Perhaps we should add a new property prov:ApproximatelyInverseFunctionalProperty
in order to avoid problems here as well?
Tore
______________________________________________________________
<> dc:creator [
foaf:name "Tore Eriksson",
"トーレ エリクソン"@jp;
foaf:mbox_sha1sum "08445a31a78661b5c746feff39a9db6e4e2cc5cf";
foaf:workplaceHomepage <http://www.taisho.co.jp/> ].
Received on Friday, 29 March 2013 13:16:08 UTC