RE: UCR issue 26

I think this email was trapped in limbo in a temporary outbox for the last fortnight. Hope it hasn’t missed the  boat, but probably has as Sapporo has passed.

Dear Frans and others

One thing that I would like to insert into the discussion here is clarification of terminology that we are going to use in the Best Practice (and these discussion of the issues).

Below is “Temporal Reference System”. Is this the same as a ‘Temporal Coordinate Reference System’ or even ‘Temporal System’? Components of such systems include: Instants, durations, periods, events, counts, calendars, notations, axes, datum/epochs, Units of Measure, …. .

At one level, there is an unwritten but valid requirement to be consistent with the  terminology and structures of the spatial reference systems (which perhaps partially answers Bill’s query as to why are we doing time?).

Unless any one can suggest alternatives, preferably pre-existing, here are my suggestions for what we a talking about, based on my presentation to the SDWWG recently and the OGC Temporal Domain WG:


0.       Temporal System: notation, events, relations/operators

1.       Temporal Reference System: notation, relations/operators, clocks, counts

2.       Temporal Coordinate Reference System: notation, relations/operators, instants , durations, UoM, Axis, Epoch

3.       Calendar: notation, units, periods, origin (epoch?), underpinning TRS or TCRS, algorithms.

I think a lot of confusion has been caused by using entities from one ‘regime’ in another.

HTH but it might not, Chris

From: Frans Knibbe [mailto:frans.knibbe@geodan.nl]
Sent: Monday, October 26, 2015 12:02 PM
To: Alejandro Llaves
Cc: Heaven, Rachel E.; SDW WG Public List; Jon Blower; Simon Cox
Subject: Re: UCR issue 26



2015-10-21 12:07 GMT+02:00 Alejandro Llaves <allaves@fi.upm.es<mailto:allaves@fi.upm.es>>:
Hi Frans,

C&P your proposal:

'It should be possible to make use of possibilities of temporal reference systems to express components of time at various levels of precision.

This requirement expresses the need to be able to handle vague, imprecise or uncertain time. Some examples are "early 1950s", "late Jurassic", "during the reign of Khafra", "the afternoon of July 1st". It should be noted that uncertainty in time does not need to be restricted to the highest precision time component in an expression of time. For instance, a photograph might be known to be taken on Christmas day, but the year in which the photograph was taken could be uncertain.''

The first sentence sounds too complex to me. I don't get the part of "to make use of possibilities of temporal reference systems".

Some Temporal Reference Systems (TRS) have possibilities of expressing time at various levels of precision. ISO-8601 for example allows "2015-10-26" and "2015-10". The requirement is for OWL Time to not restrict the freedom in expressions of time that some TRSs allow. Does this make sense? I guess it would help if we add an example like 2015-10-26" versus "2015-10".

And "it should be possible to make use of possibilities" is a bit redundant.

I don't know... the first possibilities should be in OWL Time and the second are the possibilities in TRSs. I think those are different possibilities, so it will be hard to remove one of them. But could you suggest clearer phrasing?


I like the second part with the examples ;) Maybe, it would make more sense using "to be able to represent/describe" instead of "to be able to handle", but I can live with both.

OK, I changed "handle" to "describe". Perhaps the part with the examples could be extended with explanations:

This requirement expresses the need to be able to describe vague, imprecise or uncertain time. Some examples are:

  *    ISO 8601 expressions "2012-10-23T23:46" versus "2012-10-23" versus "2012-10"
  *   "early 1950s"
  *   "later part of the Jurassic"
  *   "during the reign of Khafra"
  *   "the afternoon of July 1st" (the year is unknown)
It should be noted that uncertainty in time does not need to be restricted to the highest precision time component in an expression of time. For instance, a photograph might be known to be taken on Christmas day, but the year in which the photograph was taken could be uncertain.'

Suddenly I am not sure if the example "during the reign of Khafra" is appropriate. The timing could be uncertain because there is no indication of the definition of the reign of Khafra. Different egyptologists may have different dates for that period, and I think those different definitions would be different TRSs. But if a time is unclear because the TRS is absent, that would be another matter than the one this requirement is concerned with.

Perhaps we should remove this example. Or change it to "around the the reign of Khafra"?

Greetings,
Frans


Thanks!
Alejandro

On 21 October 2015 at 00:13, <Simon.Cox@csiro.au<mailto:Simon.Cox@csiro.au>> wrote:

>  Perhaps a new version of OWL time will be based on the idea that instants are actually intervals too?

OWL-Time does take this position already. It follows Allen’s theory, which make intervals the primary structure, and instants a special case where we can’t distinguish the beginning and end, at the current level of precision.

From: Frans Knibbe [mailto:frans.knibbe@geodan.nl<mailto:frans.knibbe@geodan.nl>]
Sent: Tuesday, 20 October 2015 10:28 PM
To: Heaven, Rachel E. <reh@bgs.ac.uk<mailto:reh@bgs.ac.uk>>; SDW WG Public List <public-sdw-wg@w3.org<mailto:public-sdw-wg@w3.org>>
Cc: Jon Blower <j.d.blower@reading.ac.uk<mailto:j.d.blower@reading.ac.uk>>

Subject: Re: UCR issue 26

Hello all,

The photo from Christmas day is a nice example. I think we should add a few examples to this requirement and the Christmas day photo should be one of those.

So here is a new proposal:

'It should be possible to make use of possiblities of temporal reference systems to express components of time at various levels of precision.

This requirement expresses the need to be able to handle vague, imprecise or uncertain time. Some examples are "early 1950s", "late Jurassic", "during the reign of Khafra", "the afternoon of July 1st". It should be noted that uncertainty in time does not need to be restricted to the highest precision time component in an expression of time. For instance, a photograph might be known to be taken on Christmas day, but the year in which the photograph was taken could be uncertain.'

I did change ''..express time" to "...express components of time", but here the distinction between intervals and instants from Rachel's proposal is not made. I am not sure such a distinction is necessary. Perhaps a new version of OWL time will be based on the idea that instants are actually intervals too?

Greetings,
Frans



2015-10-09 17:52 GMT+02:00 Heaven, Rachel E. <reh@bgs.ac.uk<mailto:reh@bgs.ac.uk>>:
The vagueness (e.g. “before 1972” or “early 1950s”, or even “the end of the Jurassic”) can usually be expressed by an interval with a different precision on each end, or an undefined start or end.  “Afternoon of June 1st” is an interval with a precise start time and a less precise end, depending on culture and season...

Then there are the other examples where one component of the date might be known very precisely (a photo from Christmas day), but the year is known with less certainty.

So perhaps:
'It should be possible to make use of possibilities of temporal reference systems to express components of time instants and components of time intervals at various levels of precision'.

Regards,
Rachel

From: Frans Knibbe [mailto:frans.knibbe@geodan.nl<mailto:frans.knibbe@geodan.nl>]
Sent: 09 October 2015 14:25
To: Jon Blower
Cc: SDW WG Public List
Subject: Re: UCR issue 26

Hi Jon,

Yes, I think this is about temporal precision. For Gregorian time it is possible to have different precisions in ISO 8601: 2003-04-27T23:45 is more precise than 2003-04-27, which is more precise than 2003. I don't think playing with precision like this is possible with XSD datatypes, especially when one is limited to xsd:dateTime.

Other temporal reference systems have precision too. For example, in geological time 'Paleogene' is more precize than 'Cenozoic'.

That would bring me to a requirement like 'It should be possible to make use of possiblities of temporal reference systems to express time at various levels of precision'.

Regards,
Frans

2015-10-08 17:38 GMT+02:00 Jon Blower <j.d.blower@reading.ac.uk<mailto:j.d.blower@reading.ac.uk>>:
Hi Frans,

I see your point (both examples could be seen as extremely precise, depending on our expectations and application).

Maybe instead of calling the requirement “temporal vagueness” it should be “temporal precision”, the requirement being to be able to express the precision of a time value.

Cheers,
Jon

On 8 Oct 2015, at 15:59, Frans Knibbe <frans.knibbe@geodan.nl<mailto:frans.knibbe@geodan.nl>> wrote:

Hello,

This is a thread for trying to resolve UCR issue 26<https://www.w3.org/2015/spatial/track/issues/26>. Again, the issue deals with clarification of a requirement. In this case it is about the OWL Time requirement Temporal vagueness<http://w3c.github.io/sdw/UseCases/SDWUseCasesAndRequirements.html#TemporalVagueness>.

Current phrasing is: "It should be possible to describe time points and intervals in a vague, imprecise manner. For instance, to represent an event happened on the afternoon of June 1st or at the second quarter of the 9th century."

The examples seem to be neither vague nor imprecise. Could other examples be supplied, or could be explained why the examples are vague and/or imprecise?

Especially the time specialists among us: please help in getting this requirement in shape.

Greetings,
Frans




________________________________
This message (and any attachments) is for the recipient only. NERC is subject to the Freedom of Information Act 2000 and the contents of this email and any reply you make may be disclosed by NERC unless it is exempt from release under the Act. Any material supplied to NERC may be stored in an electronic records management system.
________________________________




--
Alejandro Llaves

Ontology Engineering Group (OEG)

Artificial Intelligence Department

Universidad Politécnica de Madrid

Avda. Montepríncipe s/n

Boadilla del Monte, 28660 Madrid, Spain



http://www.oeg-upm.net/index.php/phd/325-allaves




allaves@fi.upm.es<mailto:allaves@fi.upm.es>