W3C home > Mailing lists > Public > public-sdw-wg@w3.org > August 2017

Re: OWL Time Ontology to Proposed Recommendation?

From: François Daoust <fd@w3.org>
Date: Mon, 21 Aug 2017 05:34:37 +0200
To: Simon.Cox@csiro.au, warren@muninn-project.org, chris.little@metoffice.gov.uk
Cc: public-sdw-wg@w3.org
Message-ID: <b43220ff-3d69-d7e7-3a11-e0c5725a551f@w3.org>
Hi Simon,

Le 21/08/2017 à 02:50, Simon.Cox@csiro.au a écrit :
> Francois -
> 1. I've made some more improvements to the implementation report https://www.w3.org/2015/spatial/wiki/OWL_Time_Ontology_adoption
> 2. I've added a little more text and clarifications to the SOTD block at the top of the document  http://w3c.github.io/sdw/time/?specStatus=PR;previousMaturity=CR;previousPublishDate=2017-06-06;publishDate=2017-08-31;prEnd=2017-09-29;crEnd=2017-07-04;implementationReportURI=https://www.w3.org/2015/spatial/wiki/OWL_Time_Ontology_adoption

The rationale looks very good to me.
I note a typo: "ommission" should be "omission", I think.
Also the text "This paragraph to be removed from final version" seems to 
apply to the wrong paragraph (I suspect we will keep the "For OGC" 

> 3. In my opinion, the :hasTime element is the only feature that remains at risk. There is one use of it in implementations, but the justification is not as strong as the others - see the note in draft. I could accept dropping this feature-at-risk only.

I would suggest to keep it as-is, simply in the interest of time because 
the call for consensus I sent proposes to keep all features at risk. But 
if you or someone else feels strongly about it, then now is a good time 
to drop the term!

> 4. Otherwise, do you want me to remove the line
>                 [Editorial note:	Feature at risk - added in 2017 revision, and not yet widely used.]
> from the tables in section 4?

Yes, please drop these notes as well.


> Simon
> -----Original Message-----
> From: Francois Daoust [mailto:fd@w3.org]
> Sent: Thursday, 17 August, 2017 17:57
> To: Cox, Simon (L&W, Clayton) <Simon.Cox@csiro.au>; warren@muninn-project.org; chris.little@metoffice.gov.uk
> Cc: public-sdw-wg@w3.org
> Subject: RE: OWL Time Ontology to Proposed Recommendation?
> Hi Simon,
> The distinction you added yesterday to the usage document between ontologies that consume the Time Ontology and data that use terms from the Time Ontology seems a good approach to cater for the producer/consumer axis.
> I understand you would like to preserve at risk features, so there is no need to make any normative changes to the specification. In the interest of time, I will issue a call for consensus to request publication of the spec as a Proposed Recommendation in a separate email (so that it does not get missed in a nested thread).
> I think we should explain why at risk features that are not yet widely implemented are retained in the "Status of This Document" section, but this can be done in parallel. The introduction text in the usage report could serve as basis for that. Rob, if you can add provide a couple of sentences as well, that would be great!
> Thanks,
> Francois.
>> From: Simon.Cox@csiro.au [mailto:Simon.Cox@csiro.au]
>> Sent: Wednesday, August 16, 2017 11:56 PM
>> Thanks Rob -
>> Could you write a couple of sentences with your thinking on this?
>> Francois, Chris -
>> I've done another scrub of the usage page. I found a few more adoption
>> examples, including some small demonstrations of almost all of the
>> at-risk features. Not yet in the numbers requested by the directorate,
>> but definite movement.
>> Probably the biggest question in my mind is around the
>> producer/consumer axis.
>> Not quite sure how to satisfy this, particularly for a product that is
>> not expected to be exposed 'directly' but only as built deep into applications.
>> Simon
>> -----Original Message-----
>> From: Rob H Warren [mailto:warren@muninn-project.org]
>> Sent: Wednesday, 16 August, 2017 04:31
>> To: Chris Little <chris.little@metoffice.gov.uk>
>> Cc: Cox, Simon (L&W, Clayton) <Simon.Cox@csiro.au>; François Daoust
>> <fd@w3.org>; public-sdw-wg@w3.org
>> Subject: Re: OWL Time Ontology to Proposed Recommendation?
>> I can personally think of several use cases for the at-risk features.
>> I'd like to retain them even it if means revisiting the issue later.
>> -rhw
>>> On Aug 15, 2017, at 2:06 PM, Little, Chris
>>> <chris.little@metoffice.gov.uk>
>> wrote:
>>> Simon, François,
>>> I have no more information to add about implementations. I had hoped
>> that colleagues at the Met Office could use the 'at risk' terms as
>> part of a test suite for WKT for Temporal CRS, this is too uncertain
>> and certainly too far away.
>>> I would argue that the 'at risk' terms be retained for completeness,
>> otherwise we would be open to the criticism of 'why have you left this
>> obvious hole that could be completed by an obvious extension?'.
>>> But I am not an ontologist, and would be happy to follow Simon's
>> judgement on the way forward.
>>> Chris
>>>> -----Original Message-----
>>>> From: François Daoust [mailto:fd@w3.org]
>>>> Sent: Sunday, August 13, 2017 9:18 AM
>>>> To: Simon.Cox@csiro.au; Little, Chris
>>>> Cc: public-sdw-wg@w3.org
>>>> Subject: Re: OWL Time Ontology to Proposed Recommendation?
>>>> Hello Simon, Chris,
>>>> The implementation report for the Time Ontology seems to have been
>>>> updated beginning of August:
>>>> https://www.w3.org/2015/spatial/wiki/OWL_Time_Ontology_adoption
>>>> The usage table seems mostly good to me. The terms for which there
>>>> is no implementation evidence are the ones that had been identified
>>>> as "features at risk". Would you say that the implementation report
>>>> is now good enough? I note the report does not seem to make any
>>>> difference between consumers and producers. The Director will
>>>> likely ask about that, since he insisted on that distinction during
>>>> the Candidate Recommendation transition call.
>>>> What do you recommend for the terms with no implementation evidence?
>>>> As features at risk, they can be dropped from the spec without
>>>> problem. Or you may argue that there will prove useful in the long
>>>> term and that they should be preserved. What's your take on this?
>>>> We need to move forward and publish the specification as a Proposed
>>>> Recommendation as soon as possible, so that it may be published as
>>>> a final Recommendation by end of September. To request transition
>>>> to Proposed Recommendation, I need:
>>>> 1. An updated spec if you propose to drop terms with no
>>>> implementation evidence 2. Some demonstration that all CR exit
>>>> criteria
>> have been met:
>>>>  http://www.w3.org/TR/2017/CR-owl-time-20170606/#exit
>>>> 3. A record of a group's resolution to request transition to
>>>> Proposed Recommendation. Could you issue a call for consensus with
>>>> a one week deadline once the spec has been updated?
>>>> I note that the SSN spec has a normative dependency on the Time
>>>> Ontology spec. This means that the Time Ontology spec currently
>>>> blocks the SSN spec. In other words, the SSN spec cannot be
>>>> published as a Proposed Recommendation as long as the Time Ontology
>>>> has not been published as a Proposed Recommendation. I would
>>>> recommend to
>> try
>>>> to publish both specs at once.
>>>> Thanks,
>>>> Francois.
>>>> Le 06/07/2017 à 17:58, Francois Daoust a écrit :
>>>>> Hello Simon, Chris,
>>>>> Now that the minimal period of time in CR has elapsed, the Time
>>>> Ontology could in theory move forward to Proposed Recommendation.
>>>> This requires CR exit criteria to be met though.
>>>>> I think there is a pending implementation to be added to the OWL
>>>>> Time
>>>> Ontology adoption page:
>>>>> https://www.w3.org/2015/spatial/wiki/OWL_Time_Ontology_adoption
>>>>> Could you add it there?
>>>>> Would you consider this implementation report to be complete
>>>> afterwards?
>>>>> If so, we should run a call for consensus within the Spatial Data
>>>>> on
>>>> the Web WG to request a transition to Proposed Recommendation.
>>>>> If not, when could we get to a point where that call can be issued?
>>>>> Thanks,
>>>>> Francois.
Received on Monday, 21 August 2017 03:34:52 UTC

This archive was generated by hypermail 2.4.0 : Thursday, 24 March 2022 20:31:33 UTC