W3C home > Mailing lists > Public > semantic-web@w3.org > June 2007

Re: owl:sameAs use/misuse/abuse Re: homonym URIs

From: Pierre-Antoine Champin <swlists-040405@champin.net>
Date: Fri, 15 Jun 2007 10:29:15 +0200
Message-ID: <46724DDB.7030807@champin.net>
To: Hans Teijgeler <hans.teijgeler@quicknet.nl>
Cc: 'Pierre-Antoine Champin' <swlists-040405@champin.net>, 'Dan Brickley' <danbri@danbri.org>, 'Bernard Vatant' <bernard.vatant@mondeca.com>, semantic-web@w3.org

Hi Hans,

fine with this subtle distinction. But there are a lot of URIs around
that are not excplictly denoting either the WholeLifeIndividual or a
Temporal part of it, and which are probably used for both in different

Note also that this is ok as well, as long as the application using
those statements do not require the distinction either. But what about
application that *do* require it? Should they rule out such ambiguous
URIs as "errors", or be provided with means to cope with them?


Hans Teijgeler a écrit :
> Hi Bernard,
> In ISO 15926 we have a solution for that. We have to, because our aim is storing the lifetime (say 50 years) information of a process plant and its zillion of parts and subparts, and its streams, people, costs, etc.
> We work with "temporal parts" that are parts of a WholelifeIndividual.
> So we have the instance of WholeLifeIndividual that is you from your birth to your (hopefully very distant) death.
> That individual has its own URI, but does not have a name yet.
> Then, when you were three days or so old, your parents gave you your offical name. 
> That is a property of a temporal part of you (new URI). Temporal, because you may change your name at some time during your life.
> If you do that, you end that temporal part and create a new one (new URI again) with the new name as a property.
> You start to work for Company A, that is, a temporal part of you works there, and that temporal part begins its existence at the date-time that your employment starts. You start with a salary of 1000€, that is: a temporal part of the temporal part that works for Company A makes that salary. When you get a raise, that temporal part ends, and a new one is created with the higher salary as a property, etc, etc.
> Simultaneously you work with Company B, that is: yet another temporal part of you works there, and a temporal part of that temporal part earns 2000€, etc, etc.
> Resuming, in principle there are trillions of temporal parts of you (e.g. for any time that you blink your eyes), but fortunately there is not a business need to put them all on record.
> Thought it might interest you. Knowing the SW scene a little by now this will, at best, be seen as "interesting", without having something better to offer.
> Regards,
> Hans
> ____________________
> OntoConsult
> Hans Teijgeler
> ISO 15926 specialist
> Netherlands
> +31-72-509 2005
> www.InfowebML.ws 
> hans.teijgeler@quicknet.nl
> -----Original Message-----
> From: semantic-web-request@w3.org [mailto:semantic-web-request@w3.org] On Behalf Of Pierre-Antoine Champin
> Sent: Thursday, June 14, 2007 17:12
> To: Dan Brickley
> Cc: Bernard Vatant; semantic-web@w3.org
> Subject: Re: owl:sameAs use/misuse/abuse Re: homonym URIs
> Dan Brickley wrote:
>> Bernard Vatant wrote:
>>> Just to hit this owl:sameAs (ab)use nail a bit more.
>>> Although I agree with Pat below (see my previous message) suppose I 
>>> (or Richard) disagree(s) and want(s) to stick to the assertion 
>>> http://dbpedia.org/resource/Berlin owl:sameAs 
>>> http://sws.geonames.org/2950159/
>>> Does that mean that what I get from the two resources should be not 
>>> only consistent RDF descriptions, but *identical descriptions* ? I 
>>> guess so. It's clear that it's not the current case.
>> The point is, according to the owl:sameAs claim, there aren't two 
>> resources, just one. One thing - with (at least two names (URIs). 
>> Asking an information system (such as the Web itself, or a library 
>> catalogue) about this thing could reasonably elicit different answers, 
>> depending on which name is used. That doesn't mean there are two things.
>> Similarly, in the real world, different people and info systems known 
>> different things about me; they may even consider me to have different 
>> names/URIs. But there's only one me.
> Consider that I work for two different companies (in the morning and in the afternoon).
> Both have a URI for me. Company 1 would state
> comp1:champin emp:name "Champin" ;
>               emp:salary "1000€" .
> Company 2, on the other hand, would state
> comp2:champin emp:name "Champin" ;
>               emp:salary "2000€" .
> using the same standardized properties, which happen to be functional.
> It would seem legitimate to state that
> comp1:champin owl:sameAs comp2:champin .
> But that would lead to inconsistency (two different values for a functional property).
> Both URIs denote me, but not the same "me", only the "me" I am from the point of view of each company.
> Ambiguity is always lurking around.
>   pa
>> And so, anything true of me, is
>> true of me. Some things might be true of one of my *names* (eg. that 
>> it is mentioned in a particular database). So yup, owl:sameAs is a 
>> pretty strong claim. Anything true of the one should be true of the 
>> other; because there is just the one.(*)
>> Whether an HTTP GET that returns a 200 should always return the same 
>> thing, ... is an interesting question. It's certainly (if we believe 
>> the HTTP responses, and we believe the owl:sameAs claim) supposed to 
>> be considered an interaction with the same thing. But plenty of URIs 
>> return different or random or context-specific responses.
>> http://spypixel.com/2006/spanglish/futurebot.cgi names the self-same 
>> resource as http://spypixel.com/2006/spanglish/futurebot.cgi (not 
>> becaues of owl:sameAs, but because it is the same URI :)  ... yet two 
>> GETs typically get different HTTP answers.
>> Dan
>> (*) tiptoing past philosophers of language here
> No virus found in this incoming message.
> Checked by AVG Free Edition. 
> Version: 7.5.472 / Virus Database: 269.8.15/848 - Release Date: 13-Jun-07 12:50
> No virus found in this outgoing message.
> Checked by AVG Free Edition. 
> Version: 7.5.472 / Virus Database: 269.8.16/849 - Release Date: 14-Jun-07 12:44
Received on Friday, 15 June 2007 08:29:33 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 07:41:57 UTC