Re: call for feedback! Re: [owl changed] ISSUE-83: Express inverse relationships in Provenance Model as well as ontology

Graham,

Thanks for looking it over.

On Jun 2, 2012, at 4:09 AM, Graham Klyne wrote:

> I have mixed feelings,

Could you scratch at them? I'm just curious.

> but if you're going to do this, I think the placement (appendix) and tone are about right.

Thanks!

> 
> Is it worth noting that the preferred direction for provenance properties had been from the entity described towards it's historical antecedents?

I think it's worth noting that there is a method to the madness of choosing which direction, but we've never really found a Good Way to describe it.
"Pointing back in the past" seems to be the quickest and easiest way to describe it among the group, but I'm worried about cluttering the message. In the end, it's a Recommendation for that reason - the working group "decided" and the public (eventually?) agreed, perhaps after some contest and fixing up.

We're open to suggestions on how to phrase this usefully and without distractions.

Regards,
Tim



> 
> #g.
> 
> -- 
> Sent from my Android phone with K-9 Mail. Please excuse my brevity.
> 
> Timothy Lebo <lebot@rpi.edu> wrote:
> prov-wg,
> 
> The prov-o team is looking for feedback on ONLY the following section:
> 
> https://dvcs.w3.org/hg/prov/raw-file/318372af5976/ontology/Overview.html#names-of-inverse-properties
> 
> It's a screenful of text.
> 
> Suggestions on inverse names are welcome, but don't spend too much time on the current names. Our interest is whether or not we got the discussion correct.
> 
> Thanks!
> Tim
> 
> 
> On Apr 30, 2012, at 3:22 PM, Timothy Lebo wrote:
> 
>> This ISSUE has moved forward a few steps.
>> 
>> Stian has added prov:inverse annotations to the ontology, and 
>> the appendix and table are available at:
>> 
>> http://aquarius.tw.rpi.edu/prov-wg/prov-o#names-of-inverse-properties
>> 
>> Today, the prov-o team agreed to continue on our path of including these inverse names in an appendix and defining the inverses in a separate file.
>> 
>> Next steps:
>> 
>> * prov-o team (and prov-wg) to review the inverse names at http://aquarius.tw.rpi.edu/prov-wg/prov-o#names-of-inverse-properties
>> * Tim to write appropriate narrative for justifying our decisions to define the inverse names separately.
>> 
>> Regards,
>> Tim
>> 
>> On Apr 16, 2012, at 3:18 PM, Timothy Lebo wrote:
>> 
>>> The prov-o team agreed [1] to maintain a separate file that names the inverses of the ObjectProperties in PROV-O, which will be listed in an appendix but not defined in the PROV-O itself.
>>> 
>>> To get started with this, I have added a new annotation property "prov:inverse" that provides the local name of the inverse.
>>> I've also defined the prov:inverse for prov:used and prov:wasGeneratedBy.
>>> 
>>> I will shortly generate the table for the appendix, which will be derived from these annotations.
>>> 
>>> http://dvcs.w3.org/hg/prov/rev/7cfaab5c925d
>>> 
>>> -Tim
>>> 
>>> 
>>> [1] http://www.w3.org/2011/prov/wiki/PIL_OWL_Ontology_Meeting_2012-04-16
>>> 
>>> 
>>> On Apr 8, 2012, at 3:30 PM, Daniel Garijo wrote:
>>> 
>>>> Hi all,
>>>> there are some inverse properties now in the ontology (like generated), but we have avoided to 
>>>> include them as much as we can to keep the model simple. If the creator of this issue is not happy
>>>> with this resolution, I would like to ask him/her to provide some examples showing the need for the inverse
>>>> properties, and why the current model is not enough for capturing them.
>>>> 
>>>> This issue is now pending review, and I propose to close it.
>>>> 
>>>> Thanks,
>>>> Daniel
>>>> 
>>>> 2011/8/25 Provenance Working Group Issue Tracker <sysbot+tracker@w3.org>
>>>> 
>>>> ISSUE-83: Express inverse relationships in Provenance Model as well as ontology
>>>> 
>>>> http://www.w3.org/2011/prov/track/issues/83
>>>> 
>>>> Raised by:
>>>> On product:
>>>> 
>>>> 
>>>> 
>>>> 
>>>> 
>>>> 
>>> 
>> 
>