On Mon, 29 Nov 2010 05:55:16 +0000, Simon Grant wrote 

> But wait a minute.. If I have millisecond precision I must have, at least,
time zone precision.. So where is the interval 1985 defined? By UTC? Or if a
book was published in 1985 in Berlin shall we have call to define the year as
a range using a span of time in Middle European time? 

The standard allows for times to be specified without zone (implicit is the
zone of the speaker).

> Perhaps there isn't a perfect answer to this. But it would be unwise to give
a date as "1985" when you mean the end of 1985-12-31 at a time when other time
zones had already gone over to 1986. 

We then run into a number of problems and can't talk about 12 Dec or 1 Jan,
last day or 1st of the month or even day since the time difference between two
points on the earth can be as much as 24 hours (Adelaide to Alofi in the South

> I think the better analogy is with estimation, as in statistics, where one
routinely sees error bars.

There is a difference between precision and error. Error is about accuracy
while precision is about readability and repeatability. One can have higher
accuracy than the precision of a device just as one can have highly precise
devices with poor accuracy. Expensive lab instruments are about high precision
and high accuracy--- typically also traceable against an accepted national
reference (such as NIST in the US).

ISO8601:2004 gives use also a number of implicit fractional unit precisions.
The expression of time 12:30.5Z is, I'd suggest, not the same as 12:30:30Z
12:30.5 is either reading to 1/10 of a minute or 1/2 a minute.
12:30:30 is implicitly reading to seconds.
In the Hebrew time system I mentioned--- whose reference is 1 degree of
celestial rotation-- we have units of 1/18 minute--- 1080 halakim per hour---
or 3 1/3 seconds.

> There is something inherently vague about vagueness. It would be very
peculiar to say something like 2010-11-29T05:52Z +/- P1Y4M23DT4H43M12.358S
(which could be expressed as a datetime value somewhere within a
rather-too-well-defined interval) - so one simply wouldn't, but there seems no
virtue in excluding that possibility from our vocabulary. 
> Simon  
> -- 
> Simon Grant 
> +44 7710031657 


 Edward C. Zimmermann, NONMONOTONIC LAB 
 Basis Systeme netzwerk, Munich Ges. des buergerl. Rechts 
 Office Leo (R&D): 
  Leopoldstrasse 53-55, D-80802 Munich, 
  Federal Republic of Germany 
 Umsatz-St-ID: DE130492967