- From: Antoine Isaac <aisaac@few.vu.nl>
- Date: Tue, 5 Feb 2019 23:52:11 +0100
- To: Graham Vowles <graham.vowles@bloxstore.net>, Rob Atkinson <rob@metalinkage.com.au>
- CC: "Michael Steidl (NIT)" <mwsteidl@newsit.biz>, <public-odrl@w3.org>
I guess we're all on the same line... @Rob, I'm rather optimistic that we can keep a rather open approach without piling up more cases. The Profiles work is already committed to refer to ODRL and I'm not expecting it will say that ODRL Profiles are not profiles. And in fact the use case for the Europeana Data Model is one very similar case already, as Europeana's model does define new terms too, when there's nothing very fit for re-use. Antoine On 05/02/2019 12:04, Graham Vowles wrote: > I second what Rob is saying - the nature of profiles has been a major drag on the standardisation process for OGC and ISO. If this group can find a way to keep things really open and simple then you have a chance of achieving interoperability in the real world. Adding profiles just makes things harder - to be avoided if you can. > >> On 4 Feb 2019, at 22:23, Rob Atkinson <rob@metalinkage.com.au <mailto:rob@metalinkage.com.au>> wrote: >> >> >> This discussion over tyhe nature of profiles has vexed OGC and ISO too somewhat. >> >> I have come to believe its actually an Open world vs Closed world perspective, and the evidence i cite here are things like ISO 19156 (Observations and Measurement, aka O&M ) which has a placeholder of type xsd:any for Result (of an observation). Profiles that extend O&M by defining a datatype of the Result are actually defining a constraint on its type.... >> >> so in any OWA case, defining an extension and mandating aspects of its use is actually putting a constraint on the data shape - the CWA view of the object. >> >> if you define a new property or type without any associated constraints - such as its domain or cardinality, then what we may have is one artefact fulfilling two different roles - and the underlying profile (conformance requirement) can be logically separated from the extension vocabulary. I think the Profiles ontology allows this, but we havent really illustrated an example for the case in point. >> >> suggested resolution: provide an example and some explanatory wording around artefacts that define both conformance requirements and definitions (or vocabularies). We could also add this as another Use Case. In both cases it would be helpful to have canonical URIs and labels for the artefacts involved and any context that ought to be included as metadata - such as the publisher. >> >> >> On Tue, 5 Feb 2019 at 03:27, Michael Steidl (NIT) <mwsteidl@newsit.biz <mailto:mwsteidl@newsit.biz>> wrote: >> >> Hi Antoine, >> I enjoy, just came back from my first cross country skiing tour - everything is white, except the steep black parts of the mountains, and the sun is shining from a bright blue sky :-) >> >> I agree to you considerations, in particular as you have a more unbiased view on this topics: while going over the W3C Profile Ontology document I noticed that its understanding of a profile is: a well defined set of constraints applied to a wider specification. While the traditional ODRL 2x. view is: constraints (this was possible only in 2.0 and 2.1) and/or extensions. >> As ODRL 2.2 is a W3C Recommendation I think it has to stick to the term Profile - but I agree it should be made clear that this is a type of profile which allows to extend the specification. >> In general I like and support the ODRL approach to require that additions have to be applied in a well organized way. Only new terms which are defined in an ODRL Profile can be used with an ODRL policy - nowhere defined ad-hoc terms are invalid. >> A wider consideration is: would it make sense to define a terminology for profiles using only constraints, profiles using only extensions and profiles using a mix of both? >> >> Best, >> Michael >> >> -----Original Message----- >> From: Antoine Isaac <aisaac@few.vu.nl <mailto:aisaac@few.vu.nl>> >> Sent: Monday, February 4, 2019 12:01 PM >> To: Michael Steidl (NIT) <mwsteidl@newsit.biz <mailto:mwsteidl@newsit.biz>>; public-odrl@w3.org <mailto:public-odrl@w3.org> >> Cc: 'Rob Atkinson' <rob@metalinkage.com.au <mailto:rob@metalinkage.com.au>> >> Subject: Re: Profiles Ontology >> >> Hi Michael, >> >> I hope you're enjoying skiing. >> I realize that may mail is maybe too late and too early, but I feel I must say something... In fact I hope that there is no contradiction between the two notions of profiles. The data model I'm involved in at Europeana is indeed a bit like the ODRL profile. Sometimes we may mint our own terms when one is not available in an existing spec, before we 'narrow down' the possible worlds by expressing constraints. >> The Europeana Data Model is part of the Profile Use Case so I hope we can find a common ground. >> It's probably going to happen around the notion that some things that are called profile also happen to provide extensions at the same time. Maybe it requires a distinction between an 'ODRL-profile' and the 'ODRL-extension-used-by-this-ODRL-profile' but it's probably workable. >> >> Cheers, >> >> Antoine >> >> On 21/01/2019 09:39, Michael Steidl (NIT) wrote: >> > Hi Antoine, >> > don't hurry, I'm watching this email list and Issue 7 [1] :-) >> > >> > Hi all, >> > today I had a deeper look into the Profiles Ontology [2] and matched its features against the profile as defined by ODRL [3]. >> > >> > The Profile Ontology describes a profile as "A named set of constraints on one or more identified base specifications or other profiles, including the identification of any implementing subclasses of datatypes, semantic interpretations, vocabularies, options and parameters of those base specifications necessary to accomplish a particular function." >> > By my understanding the key feature of such a profile is to apply constraints on the wider specifications of an information exchange standard. ODRL-related example: ODRL defines multiple sub-classes of the Rule Class (Permission, Prohibition, Duty, Obligation) and such a profile may define that only the Permission sub-class may be used. >> > >> > The ODRL Information Model describes a profile as "An ODRL Profile MUST be defined to provide vocabulary terms (....) that can be used in ODRL policies requiring additional semantics." Details regarding how to define a profile are defined in the ODRL Profile Mechanism. >> > My conclusion: reading the description and having a look at the items of the Profile Mechanism I see "additional" as the key term used there. And the ODRL profile specification does not even allow to apply constraints on what is defined as the ODRL Core Vocabulary/Profile. >> > >> > By my view these two profiles act quite differently: the Profiles Ontology narrows down, a ODRL Profiles defines additional terms, beyond the specification. >> > I'm interested to read if my conclusions are wrong - also as I have taken the role to be behind the use of ODRL profiles. >> > >> > Btw: unfortunately I will not be able to join the call on 4 February (I'll be skiing and throwing a snowball to Australia :-) ) >> > >> > Best, >> > Michael >> > >> > [1] https://github.com/w3c/odrl/issues/7 >> > [2] https://w3c.github.io/dxwg/profilesont/ >> > [3] https://www.w3.org/TR/odrl-model/#profile >> > >> > ============================================================ >> > Gesendet von / sent by: >> > Michael W. Steidl >> > Email: mwsteidl@newsit.biz <mailto:mwsteidl@newsit.biz> >> > 1180 Wien/Vienna – Österreich/Austria >> > >> > -----Original Message----- >> > From: Antoine Isaac <aisaac@few.vu.nl <mailto:aisaac@few.vu.nl>> >> > Sent: Tuesday, January 8, 2019 10:54 AM >> > To: public-odrl@w3.org <mailto:public-odrl@w3.org> >> > Subject: Re: Profiles Ontology >> > >> > Hi everyone, >> > >> > I have probably raised the profile work of DXWG in some discussions, especially with Michael. But not in the recent week. Btw I'm not forgetting I need to come back and read what Michael has written on ODRL profile recently! >> > >> > Cheers, >> > >> > Antoine >> > >> > On 04/01/2019 04:59, Car, Nicholas (L&W, Dutton Park) wrote: >> >> Hi Renato, >> >> >> >> OK, I’ll try and attend the next ODRL teleconf. I guess that’s a casual 10pm Feb 4th, Brisbane time! >> >> >> >> Cheers and talk then, >> >> >> >> Nick >> >> >> >> *From:*Renato Iannella <r@iannel.la <mailto:r@iannel.la>> >> >> *Sent:* Friday, 4 January 2019 1:44 PM >> >> *To:* Car, Nicholas (L&W, Dutton Park) <Nicholas.Car@csiro.au <mailto:Nicholas.Car@csiro.au>> >> >> *Cc:* public-odrl@w3.org <mailto:public-odrl@w3.org> Group <public-odrl@w3.org <mailto:public-odrl@w3.org>>; Cox, Simon (L&W, Clayton) <Simon.Cox@csiro.au <mailto:Simon.Cox@csiro.au>>; Rob Atkinson <rob@metalinkage.com.au <mailto:rob@metalinkage.com.au>> >> >> *Subject:* Re: Profiles Ontology >> >> >> >> Hi Nick - that is great to see ODRL already being used as an example. >> >> >> >> Our next teleconf is 4 Feb 12:00UTC and we will discuss then (if you can make it, that would be great!) >> >> >> >> I saw the Profile Ontology as it was in the weekly W3C news bulletin :-) >> >> >> >> Cheers - Renato >> >> >> >> >> >> >> >> On 2 Jan 2019, at 08:01, Car, Nicholas (L&W, Dutton Park) <Nicholas.Car@csiro.au <mailto:Nicholas.Car@csiro.au> <mailto:Nicholas.Car@csiro.au <mailto:Nicholas.Car@csiro.au>>> wrote: >> >> >> >> Hi Renato, >> >> >> >> In our paper introducing the ontology for the Extended Semantic Web Conference in Slovenia in June, we have included an examples section addressing modelling of ODRL2 already! That bit’s not in the FPWD but will be in the 2nd PWD. When we have it written in to the space doc, I’ll make sure to let you know. >> >> >> >> In addition to the ontology doc, there is the emerging Guidance doc that’s now quite up to FPWD [1]. That doc, which we are Woking on for the FOWD push now, should also be informed by ODRL so I’ll ask you for thought there too. >> >> >> >> How did you come by the Profiles Ont by the way? We haven’t deliberately reached out yet for comment but I’d like to know how word gets around. >> >> >> >> Cheers, >> >> >> >> Nick >> >> >> >> [1] https://w3c.github.io/dxwg/profiles/ >> >> >> >> Nicholas Car >> >> >> >> Senior Experimental Scientist >> >> >> >> CSIRO >> >> >> >> nicholas.car@csiro.au <mailto:nicholas.car@csiro.au> <mailto:nicholas.car@csiro.au <mailto:nicholas.car@csiro.au>> | 0477 560 177 >> >> >> >> >> >> On 1 Jan 2019, at 9:43 pm, Renato Iannella <r@iannel.la <mailto:r@iannel.la> <mailto:r@iannel.la <mailto:r@iannel.la>>> wrote: >> >> >> >> Dear CG, the W3c has released an editors draft of the “Profiles Ontology” [1] described as: >> >> >> >> " The Profiles Ontology is an RDF vocabulary to describe profiles of (one or more) standards for information resources. It describes the general pattern of narrowing the scope of a specification with additional, but consistent, constraints, and is particularly relevant to data exchange situations where conformance to such profiles is expected and carries additional context. The Profiles Ontology enables profile descriptions to specify the role of resources related to data exchange such as schemas, ontologies, rules about use of controlled vocabularies, validation tools, and guidelines. The ontology may however be used to describe the role of artifacts in any situation where constraints are made on a the usage of more general specifications.” >> >> >> >> This maybe useful in our work in defining ODRL Profiles [2]. >> >> >> >> It may also be a good use case for them to use to "test” its applicability to an existing standard (hint, hint Nic and Simon ;-) >> >> >> >> Happy Hew Year! >> >> >> >> Renato >> >> >> >> [1] https://w3c.github.io/dxwg/profilesont/ >> >> >> >> [2] https://github.com/w3c/odrl/issues/7 >> >> >> > >> >> > > Kind regards > > > > > Graham Vowles ● graham.vowles@bloxstore.net <mailto:graham.vowles@bloxstore.net> ● +44 7748 321774 ● www.bloxstore.net <http://www.bloxstore.net/> ● @bloxstore > >
Received on Tuesday, 5 February 2019 22:52:38 UTC