- From: RA Poell <poell@fel.tno.nl>
- Date: Sun, 13 May 2001 08:50:32 +0200
- To: "www-rdf-calendar@w3.org" <www-rdf-calendar@w3.org>
- Message-ID: <3AFE2EB8.CECE3362@fel.tno.nl>
Hello Charles, Sorry for the late reaction, but I'm only handling the mails from this list now. Some of the stuff you are looking for has been developed in a component suite called "DateExpert". I have been working on this till 1996. The basic idea behind this was: - be able to provide, as an end user, a date expression in whatever form we are used to and be able for applications / agents etc to use these dates in time restricted searches. Of course this isn't the only application of it but that was the first issue. The solution I choose was the one of an absolute time line from some 15 billion years in the past to 277 billion years in future. The resolution is the second and there is a possibility to indicate fractions of seconds too. (comp integer type) The choice of the absolute zero came from a point back in time beyond the Big Bang, which should enable us to use besides the current historical dates also astronomic dates. This absolute value is used as a reference value of date exchanges between different calendars. The component suite is build as follows: - calendar manager: interface to the outer world, providing capability descriptions amongst which the list of available calendars. - several calendars: Julian/Gregorian, French Revolutionary, User calendar, (started to work on the Hebrew calendar and Maya calendar but never finished them) Each of the calendars has knowledge about the rules to be applied to this particular calendar: - range of validity - algorithms for cyclic events (Easter etc) - special rules for particular stuff (like the switch from the Julian to Gregorian calendar that occurred in different years depending on the country and where some days never existed) - conversion capabilities from text based dates (for several languages) to absolute values and the other way around. These conversion capabilities might use another component suite "Number converter" that converts text numbers in to digits and digits into text (of course also language dependant; available languages for NC are: English, French, Roman). The user calendar is a bit particular and is in fact more like a dictionary allowing the registration of reusable date expressions like "Second Dynasty of Chang". A typical use case for applications: - storage of the date expression in the form the user gave it (in case we have to rederive the absolute value corresponding to it). - transformation (behind his back) into two values for the upper and lower limit of the expression given (each date expression in fact indicates a period in time even if the seconds are given) - storage of one of the values (or both), depending on the intended use, in the date search field(s). Date related queries use exactly the same mechanism. Second phase (after 1996): One the problems with this solution was the language depency elements (The expression for "Easter", "Pasen", "Paques" etc) had all to be known to the calendar. During the development of "Notion System" [1] I stopped the development of these components is the way they were defined in the beginning in order to use the same principle (absolute date reference values) in combination with nodes in the semantic network providing language defined references to e.g. cyclic events (like Easter) and leave the algorithms only within the calendar components. The "User Calendar" has completely disappeared. The "conversion" of normal dates (in various languages) has not yet found a solution that completely satisfies me. But I didn't spend much time on this topic since. [1] http://www.notionsystem.com Friendly greetings -- Ronald Poell Consultant Knowledge Management Netherlands Organization for Applied Scientific Research (TNO) Oude Waalsdorperweg 63 P.O. Box 96864 2509 JG The Hague The Netherlands T +31 70 374 02 00 F +31 70 374 06 52 http://www.tno.nl email poell@fel.tno.nl Charles McCathieNevile wrote: > > On Thu, 12 Apr 2001, Greg FitzPatrick wrote: > > *Maundy Thursday > > Which reminds me: > > One of the things Iwanted to be able to do is work with different calendars > - e.g. Muslim calendar, Jewish Calendar, comparison of modern Gregorian > Calendar with older variants of such for historical purposes such as > understanding why the "October Revolution" didn't happen in what I always > imagine when I think of October 1917, etc. > > One of the best use cases I came up with was figuring out when Greek > Orthodox churches and Australian Protestant churches celebrate easter at the > same time. > > Thursday before first Sunday after first full moon after (southern > hemisphere) autumn eqiunox is my belief about how to calculate maundy > thursday for australian protestant churches. I don't actually know what part > of the rule differs for greek churches, just that sometimes, like this year, > easter is the same time, and sometimes it is different. > > So the important requirements: > > A way of talking about what calendar is being used, and of relating that to > at least one other calendar. > > A way of specifying cyclic events (obvious requirement) > > A way of specifying one event as occurring based on the time of some other > event > > For example: > full moon occurs every 28.(a bit) days, and one of those days in ISO xxxx > time is yyyy > > or > Southern Hemisphere seasonal calendar > Autumn Equinox occurs on northern hemisphere seasonal calendar spring Equinox > > or > Anglican church of Australia Calendar > Maundy Thursday 2001 occurs on ISO xxxxyyyyzzzz > Good Friday 2001 occurs Greek Orthodox Church calendar Good Friday 2001 (??) > Palm Sunday 2001 occurs z days after MyCalendar Beltane xyzyx > > cheers > > Chaals > > (PS a lot of calendaring is not religious, but a remarkable amount of it > is...) > > -- > Charles McCathieNevile http://www.w3.org/People/Charles phone: +61 409 134 136 > W3C Web Accessibility Initiative http://www.w3.org/WAI fax: +1 617 258 5999 > Location: 21 Mitchell street FOOTSCRAY Vic 3011, Australia > (or W3C INRIA, Route des Lucioles, BP 93, 06902 Sophia Antipolis Cedex, France)
Received on Sunday, 13 May 2001 02:51:31 UTC