RE: The 'valid time' requirement

Frans,

On Monday, October 26, 2015 3:26 PM, Frans Knibbe wrote:
 
> 2015-10-22 20:09 GMT+02:00 Svensson, Lars <L.Svensson@dnb.de>:
> Frans,
> [...]
> >
> > The community behind the DCMI metadata terms could also be a good
> > target. It has the definition of dcterms:valid, which can be used to indicate
> > the temporal validity of data. Unfortunately, it is restricted to dates. What
> we
> > probably want is make use of other expressions of time to indicate the
> > interval in which something is valid, and to be able to use temporal
> functions
> > (Allen's algebra) on validity intervals. So once OWL Time gets updated to
> > allow more freedom in expressions of time, it would be great if the DCMI
> had
> > a 'valid' property with a liberal time range, for which OWL Time based
> > expressions can be used.
> dcterms:valid isn't necessarily restricted to dates. The property has range
> rdf:Literal and refines dcterms:date [1]. The range of dcterms:date is also
> rdf:Literal (surprise...), its definition is "a point or period of time associated
> with an event in the lifecycle of the resource" (period is good!) and the
> comment says that "date may be used to express temporal information at
> any level of granularity. Recommended best practice is to use an encoding
> scheme, such as the W3CDTF profile of ISO 8601 [W3CDTF]", the important
> part being "temporal information at any level of granularity".
> 
> [1] http://purl.org/dc/terms/date

> 
> Best,
> 
> Lars
> 
> Thank you for pointing this out. The repeated use of the word 'date' seems
> to have misguided me. So actually it is allowed to use any representation of
> intervals or instants of time, as long as it is a literal. That gives us more liberty
> to use time representations from OWL Time.
> 
> I wonder now... could the rdf:Literal range be too restrictive? I can imagine
> well known time intervals like 'the Jurassic' to be resources instead of literals.
> Just like http://dbpedia.org/resource/1995, an expression of a time interval
> that is not a literal. Could OWL Time be made to work with temporal
> reference systems that define such resources?

I guess that we could create an appropriate sdwwg:valid that takes a time:Interval as its object. I'm not deep enough into OWL to know if we can make sdwwg:valid a sub-property of dcterms:valid but I guess that Antoine can answer that from the top of his head.

Best,

Lars

Received on Monday, 26 October 2015 23:00:43 UTC