Re: AW: AW: AW: Time Vocabulary

There are others: MIREOT [1] and Modularity [2], but probably this list 
might not be the best place to discuss these in detail (only for the 
fear of clogging the list). We can take it offline if you would like to 
discuss further.

[1] http://obi-ontology.org/page/MIREOT
[2] http://owl.cs.manchester.ac.uk/research/modularity/

On 26/06/2015 13:30, Wetz Peter wrote:
> yes, 'linked data style'. I wanted to avoid owl:imports since that 
> semantically makes PROV-O a part of our ontology, and I am sure we do 
> not want that. I am curious, what other ways are there?
>
> Best
> ------------------------------------------------------------------------
> *Von:* Monika Solanki [monika.solanki@gmail.com]
> *Gesendet:* Freitag, 26. Juni 2015 14:21
> *An:* Wetz Peter; Daniele Dell'Aglio; public-rsp@w3.org
> *Betreff:* Re: AW: AW: Time Vocabulary
>
> Sure. I can see that you have used the relationships linked-data style 
> in the vocab, this in general should not be a problem but you are 
> probably aware that PROV-O is in OWL-RL. For reasoning related tasks 
> that may occur at some point  while stream processing, If any reasoner 
> did decide to pull in all of PROV-O, it may be a good idea to point 
> this out somewhere in the docs- right not I cannot foresee any major 
> consequences, but just something to be aware of.
>
> Monika
>
> On 26/06/2015 13:05, Wetz Peter wrote:
>> Thanks for the explanations. I can follow and I agree mostly. But 
>> isn't time related provenance exactly what we want to achieve here? 
>> We want to know the time related provenance, e.g. generation time, 
>> from an entity (graph). Also see the description of 
>> prov:generatedAttime, that is, "Generation is the completion of 
>> production of a new entity by an activity. This entity did not exist 
>> before generation and becomes available for usage after this 
>> generation.". IMHO, that fits to our case.
>>
>> Anyway, at the same time I'd prefer to be "semantically safe" by 
>> extending from DUL.
>>
>> After looking at DUL I can offer one proposal: What if we make our 
>> new properties subproperties of DUL:hasEventDate, but keep the rest 
>> as is? Would that be satisfying to you?
>>
>> Best,
>> Peter
>>
>> ------------------------------------------------------------------------
>> *Von:* Monika Solanki [monika.solanki@gmail.com]
>> *Gesendet:* Freitag, 26. Juni 2015 13:48
>> *An:* Wetz Peter; Daniele Dell'Aglio; public-rsp@w3.org
>> *Betreff:* Re: AW: Time Vocabulary
>>
>> Well, it really depends on whether you are after a lexical term that 
>> conveys the semantics  (available in plenty from multiple 
>> vocabularies/ontologies) or if you subscribe to a view point. The 
>> advantage of using foundational ontologies is their neutrality and 
>> therefore when you extend from one of those, you make the term your 
>> own in some sense, while still retaining the real world scope.
>>
>> PROV-O is excellent and I use it profusely, but the view point it 
>> takes is that of provenance, so when you extend from PROV-O, you are 
>> essentially subscribing to its notion of provenance, which IMHO, is 
>> not exactly why we are building the time vocabulary  within the 
>> context of streams (I know I am treading dangerous waters here, but 
>> hey.....). Our motivation is different and we should try and retain 
>> that in our vocabulary too. So, While I would not object to using 
>> PROV-O, I would still wonder if time related provenance, as we claim 
>> when we extend from it has any semantic implications or is even 
>> needed. I know with DUL, we will be "semantically safe" :-) .
>>
>> Cheers,
>>
>> Monika
>>
>> On 26/06/2015 12:33, Wetz Peter wrote:
>>> Yes it'd be good. Please note, that I did not ignore your proposal 
>>> of extending DUL. I was just looking at PROV-O first and it seemd to 
>>> be well providing what we aim for (at least partly). I don't know if 
>>> it's doable to do both, that is, a) 'extending from DUL' and b) 
>>> 'extending from PROV-O' in one vocab.
>>>
>>> What do you think?
>>>
>>> best,
>>> Peter
>>>
>>> ------------------------------------------------------------------------
>>> *Von:* Monika Solanki [monika.solanki@gmail.com]
>>> *Gesendet:* Freitag, 26. Juni 2015 12:25
>>> *An:* Daniele Dell'Aglio; public-rsp@w3.org
>>> *Betreff:* Re: Time Vocabulary
>>>
>>> Hi,
>>>
>>> Indeed, only that it cannot happen today because of a deliverable 
>>> deadline. I will be joining the call though, so I can certainly 
>>> participate in the discussion.
>>>
>>> Monika
>>>
>>> On 26/06/2015 11:20, Daniele Dell'Aglio wrote:
>>>> Hi,
>>>> Peter, thank you for the effort. Monika, could you try to propose 
>>>> an extension (or alternative) version from DUL? It would help the 
>>>> comparison and the discussion.
>>>>
>>>> Daniele
>>>>
>>>> Il giorno ven 26 giu 2015 alle ore 11:08 Monika Solanki 
>>>> <monika.solanki@gmail.com <mailto:monika.solanki@gmail.com>> ha 
>>>> scritto:
>>>>
>>>>     I still think extending from a foundational ontology such as
>>>>     DUL has its purpose and advantages, so besides PROV-O it would
>>>>     also be useful to extend from DUL. However, I will leave it at
>>>>     that.
>>>>
>>>>
>>>>     Monika
>>>>
>>>>
>>>>
>>>>     On 26/06/2015 10:44, Wetz Peter wrote:
>>>>>
>>>>>     Dear all,
>>>>>
>>>>>     as discussed extensively on the list (cf. [1]) there is a need
>>>>>     to create a small time vocabulary (or reuse existing terms)
>>>>>     for capturing relations between streamed graphs and time instants.
>>>>>
>>>>>     I have done a first simple proposal and uploaded it to the
>>>>>     repo at [2]. I reuse properties of PROV-O by specializing from
>>>>>     some PROV-O terms. I hope you can take a look so we can have a
>>>>>     discussion in today’s telco.
>>>>>
>>>>>     Looking forward!
>>>>>
>>>>>     Best,
>>>>>
>>>>>     Peter
>>>>>
>>>>>     [1] https://github.com/streamreasoning/RSP-QL/issues/10
>>>>>
>>>>>     [2]
>>>>>     https://github.com/streamreasoning/RSP-QL/blob/master/TimeVocab.owl
>>>>>
>>>>
>>>
>>
>

Received on Friday, 26 June 2015 12:37:12 UTC