Ok, I'm persuaded that we do not need to enforce equal start and end
precision for an interbal.
> -----Original Message-----
> From: Discussion of the Developing Date/Time Standards
> [mailto:[log in to unmask]] On Behalf Of Edward C. Zimmermann
> Sent: Tuesday, July 05, 2011 3:22 AM
> To: [log in to unmask]
> Subject: Re: [DATETIME] On the usefulness of x
> On Wed, 29 Jun 2011 18:14:06 +0200, [UTF-8?]SaaÅ¡ha Metsärantala wrote
> > Hello!
> > The x-letter has lead to several misunderstandings. I tentatively
> > suggest something.
> I do not think that 'x' has led to any misunderstandings rather I
> suspect that the concept of precision applied to dates continues to
> meet with some "misunderstanding".
> 1x == P10(1n)
> 1xx == P100(1nn)
> 1xxx == P1000(1nnn)
> where n is any of 0..9 and P10, P100 and P1000 could be read as
> precision of 10, resp. 100 and 1000.
> 196x matches P10(1962), P10(1968).. in fact any of 1960..1969
> While a more general precision qualifier (Pn) might be more generic, in
> practice, I do suspect that-- like the year, month, day, hour or minute
> precision we have implicitly defined within ISO 8601-- decade (10
> years) and "century" (100 years) are (by far) the most common. Both
> decade and century are so deeply embeded in western culture that these
> marks effect events themselves.
> While I am on the topic.. Ray has suggested that in intervals the start
> and end are of the same precision. I do not agree. The most trivial
> counter- example are open intervals but we often have less precise
> start than end points. Example: 1763/1776-07-04
> Looking at more contemporary events.. When did the movement to topple
> Hosni Sayyid Mubarak start? He resigned on 11 Feb 2011 but when did the
> demonstartions start? Our measurement precision is not very good. 25
> Jan? 17 Jan? 23 Jan (when the Muslim Brotherhood announced their intent
> to participae in the 25 Jan demo)? 2 Feb? Or much earlier? In Tunesia
> we can, at best, agree on Dec 2010 as the start there... Mohamed
> Bouazizi setting himself ablaze on 17 Dec 2010 became a symbol that
> gave the movement some media momentum but it was not the start of the
> escalation.. All we really have is 14 Jan 2011 as the date when Ben Ali
> was ousted.. There are loads of similiar examples.. The Russian
> Revolution, for example, is a collective name for a series of
> "revolutions".. We call them collectively the "1917 Revolution".
> Examples where the start is of a higher precision than end, I think,
> are also easy to think of... When, for example, did the Soviet Union
> end? Dec.
> 1991? 25 Dec. with the resignation of Gorbachev or in early 1992 with
> the establishment of the "commonwealth"?
> On Fri, 1 Jul 2011 17:50:26 +0200, [UTF-8?]SaaÅ¡ha Metsärantala wrote
> > Hello!
> > > > the "x"-concept [...] in the context of CQL.
> > > It doesn't make sense to me.
> > I would be pleased to discuss this further within the context of a
> > list about CQL or another query related list. Of course, we could
> > extend EDTF
> Those in the CQL and other query language communities are invited to
> join this discussion. This list has been announced serveral times, for
> instance, in various CQL, SRU/W and Z39.50/ISO23950 lists.
> > Regards!
> > [UTF-8?]SaaÅ¡ha,
> Edward C. Zimmermann, NONMONOTONIC LAB
> Umsatz-St-ID: DE130492967