ISO 8601 recommends that, when the date and time are both stored together (in a single field), they be separated by a T.
However, Markus Kuhn's article shows the date-and-time presentation format (as opposed to the storage format) with a space. I think that's eminently sensible, as leaving out the T does not create any ambiguity. Bill Potts, CMS Roseville, CA http://metric1.org [SI Navigator] -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of kilopascal Sent: Tuesday, February 26, 2002 15:02 To: U.S. Metric Association Subject: [USMA:18434] Re: 200220022002 2002-02-26 I've seen some people write ISO-8601 with a "T" separating the date and time. Such as: 2002-02-26T21:42. Other times, I've just seen the space instead of the T. John ----- Original Message ----- From: "Bill Potts" <[EMAIL PROTECTED]> To: "U.S. Metric Association" <[EMAIL PROTECTED]> Sent: Tuesday, 2002-02-26 09:52 Subject: [USMA:18427] Re: 200220022002 > Bill: > > What you're describing is already laid out in ISO 8601. (See > http://metric1.org/dateandtime.htm. Click on the Markus Kuhn link, under "A > Summary of the International Standard Date and Time Notation.") > > To answer your punctuation question, there is only a space between the date > and the hour -- no comma. > > Bill Potts, CMS > Roseville, CA > http://metric1.org [SI Navigator] > > > > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On > Behalf Of Barbara and/or Bill Hooper > Sent: Tuesday, February 26, 2002 03:51 > To: U.S. Metric Association > Subject: [USMA:18422] Re: 200220022002 > > > on 2/20/2002 6:46 AM, Louis JOURDAN at [EMAIL PROTECTED] wrote: > > > Not a real argument for SI nor ISO, but maybe a reason to adopt the > > "European" style for date and time ? > > Why would anyone want to adopt the "European" style of date when the > international style of yyyy-mm-dd us so much more logical. > > One way to get the Europeans and the Americans to get together on something > would be to get them BOTH to abandon their current illogical practice and > adopt the logical, standard, larger-to-smaller sequence. In it's complete > form, it would be > > year - month - date, hour : min : sec > > or > > yyyy-mm-dd, hh:mm:ss > > that is, > the year 2002, on February 25, at 19:45 and 15 seconds > (15 seconds after 9:45 pm), would represented by > > 2002-02-25, 19:45:15 > > For even finer precision (for eacmple, down to the hundredth of a second) we > can write > > 2002-02-25, 19:45:15.01 > > for example. The one thing I am uncertain about is the preferred punctuation > between the day and the hour. I've used a comma above, but I don't know what > is preferred. Can anybody advise me? > > The advantage of the above format is that it has a logical sequence from > larger to smaller just like our numbering system. Also, for convenience, one > can simply omit sections off the left end if they are understood, or omit > sections off the right end if such precision is not needed. Thus, if we know > we are referring to this year, it would be sufficient just to write > > 02-25, 19:45:15 > > since the current year, 2002, would be understood. > Or, if the time is not needed to the nearest second, write it as > > 2002-02-25, 19:45 > > (If both, then write 02-25, 19:45 ) > > One can also drop off just the century and write the date as > 02-02-25 (meaning 2002-02-25, usually but not always). But I wonder about > the wisdom of prepetuating the archaic system of writing years with only two > digits in light of the problems it caused when the calendar rolled over from > 1999 to 2000 > (the "Year 2K" problem). Why should we perpetuate a system that will cause > further difficulties when we roll over from 2099 to 2100, or worse yet, when > we roll over from 2999 to 3000 > or from 9999 to 10 000 ? Let's think ahead, WAY ahead! I've already adopted > the practice of ALWAYS writing 4 digits for the year. This year is 2002, not > '02 as far as I'm concerned. It ain't hard! > > Regards, > Bill Hooper > > ======================== > Keep It Simple - Make It Metric! > ======================== >
