RE: Time Ontology in OWL review

I just started to peruse this document and have mostly similar thoughts. There are some interesting design choices that they've made and it'll take more than one read through to really "grok" whether these make sense. I have the sense that the authors of the spec understand the underlying issues, so the discussion may be easier for that.

The time zone material seems problematic and their usage of time zones is a mess. 

One example given ("Abby's birthdate") use an incremental time (POSIX time) when it's really a floating time value.

Another read through is probably indicated. I did create a label in our issue tracker (owl-time).

> -----Original Message-----
> From: ishida@w3.org [mailto:ishida@w3.org]
> Sent: Wednesday, July 13, 2016 5:55 AM
> To: public-i18n-core@w3.org
> Subject: Re: Time Ontology in OWL review
> 
> On 13/07/2016 12:58, ishida@w3.org wrote:
> >
> > i read through
> >
> > Time Ontology in OWL
> > https://www.w3.org/TR/2016/WD-owl-time-20160712/

> >
> >
> > not being an expert in linked data nor time/data topics, much of it is
> > hard for me to comment on, however there do seem to be a number of
> > issues surrounding time zones, as the spec itself mentions in
> > https://www.w3.org/TR/2016/WD-owl-time-20160712/#h-issue11.

> >
> > One thing that's not taken into account, however, afaict, is when DST
> > is introduced or discontinued for a given area, eg. it was just
> > discontinued in Egypt this year.
> >
> > I also find myself wondering why so much effort is going into creating
> > ontological terms for much of this when it seems that expressing
> > things as incremental time or looking up things in the Olson database
> > would reduce the computation needed.
> 
> I also suspect that there's a built-in assumption that DST is only switched on
> and off once in a given year.  That is not always the case.
> Egypt in 2014 changed 4 times, to accomodate Ramadan, and iiuc will change
> at least 4 times this year too, do to several changes of mind about DST
> adoption.
> 
> ri
>

Received on Wednesday, 13 July 2016 15:25:24 UTC