- From: Miles, Simon <simon.miles@kcl.ac.uk>
- Date: Tue, 9 Oct 2012 16:40:45 +0100
- To: "public-prov-wg@w3.org" <public-prov-wg@w3.org>
Ivan, Sorry for the delay in responding to this. To address the specific point of whether the primer should follow the PROV-DM or PROV-O specifications, I have tried where possible to follow PROV-DM because it should be the more technology-agnostic source and so make most sense whichever format you choose to view the examples in. Examples are currently in PROV-N or PROV-O, but might later include PROV-XML also (which I note uses "WasDerivedFrom" rather than "wasDerivedFrom"). With regards to PROV-O's figure including time relations that the primer doesn't, it is true that these are important relations, but so are many of the others that are in the primer but not in these overview figures in either PROV-O or PROV-DM/Primer. The overview figure, for the primer, is just intended to show the key classes that the rest of the model fits around and the primary ways they are connected, as this is an intuitive starting point. Ideally, I would prefer to make the figure even simpler, by removing WasDerivedFrom, WasInformedBy and ActedOnBehalfOf, but this has to be balanced with keeping some consistency with the PROV-DM, so the figure has been copied directly. thanks, Simon Dr Simon Miles Senior Lecturer, Department of Informatics Kings College London, WC2R 2LS, UK +44 (0)20 7848 1166 Transparent Provenance Derivation for User Decisions: http://eprints.dcs.kcl.ac.uk/1400/ ________________________________________ From: Ivan Herman [ivan@w3.org] Sent: 01 October 2012 18:30 To: Luc Moreau Cc: public-prov-wg@w3.org Subject: Re: Editorial error in the primer Luc, I am not sure I fully understand what you are saying. Putting the UML formalism aside, http://www.w3.org/TR/prov-o/#starting-points-figure and the figure in section 2 of the primer look pretty much identical to me except the secondary issue whether the time related predicates are there or not. And they are obviously extremely useful. But I now realize that there is in inconsistency in the capitalization indeed between prov-dm and prov-o. I find this unfortunate, to be honest, but maybe it is too late to change that. The figure in the primer has chosen to be consistent with prov-dm. I must say that, in view of the expected audience of that particular document my choice would be for that figure to be consistent with prov-o. That is what people will use in practice... Ivan On Oct 1, 2012, at 11:25 , Luc Moreau wrote: > Hi Ivan, > > My experience with using this diagram and examples of provenance graphs is that it is crucial to > distinguish what is instance from what is schema. People who come across provenance graphs for > the first time may easily regard them as schemas as opposed to instances of data structures. > > So I am very confortable with a different notation being used for instance and schema. > > If you think that this is problematic for the primer, then a figure such as > http://www.w3.org/TR/prov-o/#starting-points-figure > may be more suitable. > > Of course, then another inconsistency potentially occurs, because Entity/Activity/Agent > will potentially have the wrong capitalization, depending on the serialization you look at. > > ex:dataset1 a prov:Entity . > > entity(ex:dataSet1) > > Luc > > > > On 01/10/2012 15:43, Ivan Herman wrote: >> Many (most?) people will start with the Primer, moving over to prov-o and, maybe, at some point to prov-dm. This is even the 'advice' the group gives to end users. The inconsistency is the first issue they hit... >> >> I know it is a pain because, I presume the figures are reused, but the primer should be consistent... >> >> I. >> >> >> >> On Oct 1, 2012, at 10:34 , Luc Moreau wrote: >> >> >> >>> Hi Ivan, >>> Speaking of prov-dm only, which inconsistency? I don't see one: types are capitalized, instances are not. >>> Luc >>> >>> On 01/10/2012 15:02, Ivan Herman wrote: >>> >>> >>>> Luc, >>>> >>>> this is an editorial and non-normative issue I guess, so I will not make a problem of it. That being said, I do not really agree with the argumentation. If the UML rules lead to an inconsistency in the document, then my choice is to bend the UML rules. UML is in our documents for illustration only. >>>> >>>> Ivan >>>> >>>> On Sep 26, 2012, at 18:15 , Luc Moreau wrote: >>>> >>>> >>>> >>>> >>>>> Hi Ivan, >>>>> >>>>> It's a figure from prov-dm. >>>>> >>>>> It's a compromise between various UML conventions and design decisions: >>>>> - we wanted to label binary associations with verbal form (wasGeneratedBy, wasAssociatedWith, etc) >>>>> - In UML, the expanded version with extra attribute is encoded as an association class >>>>> - Classes in UML are capitalized >>>>> - It is recommended to use the same name for the binary association and the corresponding association class. >>>>> - With all the above, we ended up with capitalized forms, such as WasGeneratedBy >>>>> >>>>> We know that that in prov-n we write this as wasGeneratedBy and in prov-o as prov:wasGeneratedBy. >>>>> >>>>> Luc >>>>> >>>>> On 26/09/12 21:46, Ivan Herman wrote: >>>>> >>>>> >>>>> >>>>>> The capitalization of the property names on the figure right before 2.1 is wrong. It should be wasGeneratedBy, wasAssociatedWith, etc, instead of WasGeneratedBy and WasAssociatedWith. >>>>>> >>>>>> Ivan >>>>>> >>>>>> ---- >>>>>> Ivan Herman, W3C Semantic Web Activity Lead >>>>>> Home: >>>>>> http://www.w3.org/People/Ivan/ >>>>>> >>>>>> mobile: +31-641044153 >>>>>> FOAF: >>>>>> http://www.ivan-herman.net/foaf.rdf >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>> -- >>>>> Professor Luc Moreau >>>>> Electronics and Computer Science tel: +44 23 8059 4487 >>>>> University of Southampton fax: +44 23 8059 2865 >>>>> Southampton SO17 1BJ email: >>>>> l.moreau@ecs.soton.ac.uk >>>>> >>>>> United Kingdom >>>>> http://www.ecs.soton.ac.uk/~lavm >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> >>>> ---- >>>> Ivan Herman, W3C Semantic Web Activity Lead >>>> Home: >>>> http://www.w3.org/People/Ivan/ >>>> >>>> mobile: +31-641044153 >>>> FOAF: >>>> http://www.ivan-herman.net/foaf.rdf >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>> -- >>> Professor Luc Moreau >>> Electronics and Computer Science tel: +44 23 8059 4487 >>> University of Southampton fax: +44 23 8059 2865 >>> Southampton SO17 1BJ email: >>> l.moreau@ecs.soton.ac.uk >>> >>> United Kingdom >>> http://www.ecs.soton.ac.uk/~lavm >>> >>> >>> >>> >>> >> >> ---- >> Ivan Herman, W3C Semantic Web Activity Lead >> Home: >> http://www.w3.org/People/Ivan/ >> >> mobile: +31-641044153 >> FOAF: >> http://www.ivan-herman.net/foaf.rdf >> >> >> >> >> >> >> >> > > -- > Professor Luc Moreau > Electronics and Computer Science tel: +44 23 8059 4487 > University of Southampton fax: +44 23 8059 2865 > Southampton SO17 1BJ email: > l.moreau@ecs.soton.ac.uk > > United Kingdom > http://www.ecs.soton.ac.uk/~lavm > > > ---- Ivan Herman, W3C Semantic Web Activity Lead Home: http://www.w3.org/People/Ivan/ mobile: +31-641044153 FOAF: http://www.ivan-herman.net/foaf.rdf
Received on Tuesday, 9 October 2012 15:41:56 UTC