RE: Feedback on the W3C/OGC WG proposal for a time ontology

Thanks Yolanda –

The likely vote on Wednesday is merely to release the First Public Working Draft, so nothing is finalised.
Public comment is invited, so you can still organize EarthCube feedback.

Meanwhile, a few detailed comments:


1.      I’m generally aware of the EarthCube use-cases, and know some of the people involved (*). Indeed, the key change relative to the 2006 OWL-Time draft (to allow a temporal reference system to be indicated) is exactly to support the application to deep time.

2.      The issue of ‘uncertainty’ – quantified as +/- values, or unquantified such as intervals with unknown boundaries – has not yet been addressed. Specific comments on this topic are welcome. I anticipate this can be accommodated without breaking the proposed model, or the legacy.

3.      However, if your comment about the beginning and end of eras means that they must be defined using intervals (and not just uncertain points or instants) then that would be a significant modification of the model.

To elaborate on the last issue - while Allen’s model certainly allows for the fact that every instant is actually an interval when looked at closely enough, this is essentially a matter of precision. It does not mean that the notion of a temporal topology – with temporal edges of finite length joined at temporal nodes of zero length – is not the underlying idea.

I don’t think that is what the geologic timescale people would be suggesting either – it is probably just that the boundaries of eras are uncertain, not that they have a finite duration in principle. However, even if the latter is the case, then that can still be accommodated by the topological model, it is just that some intervals would be labelled ‘boundary’ and others would be labelled ‘era’. Since the set of interval relations is provably complete, all the necessary relationships between boundaries of finite length and eras can still be expressed. So I doubt if there is a real problem, it is just that the geologic timescale application needs to use the model appropriately.

Simon

(*) I’m in contact with Doug Fils regularly, and Bob Arko from time to time. Steve Richard was my co-author on two key papers that provide part of the basis for the proposed changes relative to the 2006 OWL-Time draft.



From: Yolanda Gil [mailto:gil@isi.edu]
Sent: Saturday, 2 July 2016 2:22 AM
To: Kerry Taylor <kerry.taylor@anu.edu.au>; Ed Parsons <eparsons@google.com>; Cox, Simon (L&W, Clayton) <Simon.Cox@csiro.au>
Cc: Simon Goring <goring@wisc.edu>; Douglas Fils <dfils@oceanleadership.org>; Robert Arko <arko@ldeo.columbia.edu>; Jack Williams <jww@geography.wisc.edu>; Nick McKay <Nicholas.McKay@nau.edu>; Julien Emile-Geay <julieneg@usc.edu>
Subject: Feedback on the W3C/OGC WG proposal for a time ontology

Hi Kerry, Ed, Simon,

Although I have not been able to manage to participate in the WG, I have been tracking the activities and noticed the proposal for an update to the OWL Time ontology (http://w3c.github.io/sdw/time/).  I wanted to help gather feedback  from geoscientists participating in the EarthCube program, but if you have a final vote on Wednesday I am afraid that will not be entirely possible.  We had a workshop last week on paleoclimate standards and ontologies, where chronologies are very important.  We also have an EarthCube PI meeting next week where semantics will be discussed.  I don’t want to suggest that you postpone the vote, but if there is no terrible urgency it would be ideal to check that the current proposal for OWL Time will support extensions that will be appropriate to represent geological deep time.

I want to introduce a group of people, copied in this message, who were at the workshop last week and will be meeting again at the PI meeting next week.  We will need some time to get our thoughts together and write up some feedback, but for starters here are a few rough notes I took from initial observations from Jack and Simon in our quick discussions:

- the beginning and ending of eras are often expressed not as a timepoint but a range of hundreds or maybe thousands of years
- timepoints should be expressed with a reference point, sometimes the reference timpoints vary in Europe and the US
- timepoints may have an associated uncertainty which can be expressed as +/- an interval

They also made some comments about how useful it would be to align the time representation with Darwin Core stadards for biodiversity (http://rs.tdwg.org/dwc/).

Simon also strted to put together a scenario to exercise the Time ontology, but he does not know OWL and the rest of us have had no time to comment. He can send it along if you want it in its current rough form.

Anyway, just wanted to make you aware of our interest to provide feedback, and check with you if we could do this in the next month or two (there are vacations and field trips that will be happening in July).  Perhaps these requirements are too exotic, but I think we could provide a very compelling and different use case as well as a user community because we would want to align our work with this.

Best,

Yolanda

Yolanda Gil
Director of Knowledge Technologies, USC/ISI
Associate Director for Research, Intelligent Systems Division, USC/ISI
Research Professor of Computer Science
Information Sciences Institute
University of Southern California
4676 Admiralty Way, Suite 1001
Marina del Rey, CA 90292 (USA)
+1-310-448-8794
http://www.isi.edu/~gil

@yolandagil





Begin forwarded message:

From: Kerry Taylor <kerry.taylor@anu.edu.au<mailto:kerry.taylor@anu.edu.au>>
Subject: OWL Time for FPWD
Date: June 27, 2016 at 2:56:41 PM GMT+2
To: "public-sdw-wg@w3.org<mailto:public-sdw-wg@w3.org>" <public-sdw-wg@w3.org<mailto:public-sdw-wg@w3.org>>
Resent-From: public-sdw-wg@w3.org<mailto:public-sdw-wg@w3.org>

Dear SDW-ers
We postponed a vote to progress the Time Ontology in OWL   deliverable towards FPWD in the meeting last week (https://www.w3.org/2016/06/22-sdw-minutes) due to low attendance, and suggested bringing the vote to the SSN meeting this week instead, supplemented by an invitation for email votes.

Instead, the Chairs have decided to bring the topic back to the plenary meeting of SDW next week, ie 6th July 2016 GMT.  Please review the Editor’s draft before that meeting to inform your vote: http://w3c.github.io/sdw/time/


--Kerry & Ed