First, the relevant like for date(Time) and duration values in XML Schema Datatypes is:
http://www.w3.org/TR/xmlschema-2/#isoformats
The section Duane cited is about doing arithmetic on date time values.
[OMG Date Time Vocabulary (http://www.omg.org/spec/DTV/1.0) offers an alternative approach to date time arithmetic, which is a bit more general at some cost.
The best solution for fixed time intervals is to convert to absolute durations in integer multiples of your favorite fraction of a second, add them, and convert back to a dateTime representation. But that only works for fixed intervals. For repeating intervals,
the conversion of the start times is a complex function: Consider a group that meets quarterly at 13:00 local time on the first Monday of the quarter, alternately in New York and Paris. (ICalendar is designed to support this kind of thing.)]
Second, XML Schema allows decimal fractions of seconds, not “Real” fractions. As it turns out, there is a problem with this, in that Internet Time is reported
in integer multiples of 2^-31 second (roughly 100 picoseconds). While one can convert those to decimal values, they tend to be very long, and the inverse mapping is only approximate, not 1-to-1. The point of this cavil is that the binary fractions can be
represented exactly in the internal binary forms associated with the REAL and ‘double’ data types of (most) programming and data modeling languages, but the decimal fractions usually cannot be (unless the language has a DECIMAL data type). Conversely, if
your business is Swiss timing for Olympic sports events, you have to carry the decimal value to 100ths of a second (and no more).
We have ten fingers; computers have only two; and sometimes that creates problems.
-Ed
From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx]
On Behalf Of Duane Nickull
Sent: Thursday, March 27, 2014 8:36 PM
To: [ontolog-forum]
Subject: Re: [ontolog-forum] Time expressions
***********************************
Technoracle Advanced Systems Inc.
Consulting and Contracting; Proven Results!
i. Neo4J, PDF, Java, LiveCycle ES, Flex, AIR, CQ5 & Mobile
XSD durations can represent any real number of seconds.
iCalendar durations can represent any integral number of seconds.
On Thu, Mar 27, 2014 at 3:32 PM, Toby Considine <Toby.Considine@xxxxxxxxx> wrote:
Thanks John.
That was sublime.
Moving to the mundane:
ISO 8601 specifies the _expression_ of time, including relative, absolute,
with or w/o time zones.
ISO 8601 specifies the _expression_ of intervals, that is duration of time,
including 1 year, two months, 3 weeks, and 7 minutes.
iCalendar [RFC 5545] (and therefor XCal and WS-Calendar) only allows a
subset of the legal ISO 8601 durations.
XML only allows a subset of the legal ISO 8601 durations.
The subsets used by iCalendar and XML are not the same
tc
-----Original Message-----
From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
[mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of John F Sowa
Sent: Wednesday, March 19, 2014 9:23 AM
To: ontolog-forum@xxxxxxxxxxxxxxxx
Subject: Re: [ontolog-forum] Time expressions
In my previous note in this thread, I included a note from Corpora list that
had some useful references on temporal expressions in NL texts.
I also sent an excerpt from that note to Corpora list with the URL of the
60-page "Catalog of Temporal Theories" by Pat Hayes. My note included Pat's
final disclaimer and my summary:
JFS
> In short, don't expect, imagine, or hope that any particular
> formulation is, will be, or can be the final definitive word on time.
Adam Kilgarriff replied:
> even intermittently :)
Adam runs a business that processes large volumes of "Big Data" in NL form,
he has taught "master classes" in lexicography, and he's the author of the
article "I don't believe in word senses":
http://www.kilgarriff.co.uk/Publications/1997-K-CHum-believe.pdf
Observation: In discussing ontology, we must always distinguish the kinds
of applications. The requirements for Big Data, airplane design, buying
stuff on Amazon, and answering Jeopardy! questions are totally different
from each other and from the theorem provers that computer scientists
publish papers about.
I like theorem provers, and I believe they can be useful. But it's
important to ask when, where, why, how -- and by whom and for whom.
John
_________________________________________________________________
Message Archives:
http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr:
http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/ Community Wiki:
http://ontolog.cim3.net/wiki/ To join:
http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
_________________________________________________________________
Message Archives:
http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr:
http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join:
http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
_________________________________________________________________ Message Archives:
http://ontolog.cim3.net/forum/ontolog-forum/ Config Subscr:
http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/ Unsubscribe:
mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx Shared Files:
http://ontolog.cim3.net/file/ Community Wiki:
http://ontolog.cim3.net/wiki/ To join:
http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
|
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (01)
|