[uf-dev] (Citation Format) ISO Dates

[Tantek Ç elik]

On 12/23/05 9:49 AM, "Mark Pilgrim" <pilgrim at gmail.com> wrote:

> On 12/23/05, brian suda <brian.suda at gmail.com> wrote:
>> <abbr class="dt" title="200510">Fall 2005</abbr>
>> 
>> The tricky thing is WHEN does FALL actually start, it is different
>> between the different hemispheres, Fall in the US, is not Fall in Brazil.
>> 
>> So when a publication date is simply FALL 2005, is there away this is
>> handled during encoding?
> 
> Put that way, no, there is no general solution, because "Fall 2005" is
> not a specific enough date to ever be machine-readable.  In practice,
> presumably they publish at a specific time of year (i.e. do they
> really publish their "Fall 2005" issue once in October and once in
> April, for different hemispheres?  I'm guessing not) so figure out
> which "fall" they're referring to and use that date.

This makes sense.  In fact, there is probably even a precise day on which
the "FALL 2005" issue is published.  Thus it makes sense to use that for the
ISO Date.


>> 2) Date References for works in the BCE. How do you reference the Great
>> Pyramids? 4000BC.
>> 
>> <abbr title="-4000" class="dt">4000 BC</abbr>
> 
> I can't speak for the ISO specification, but the IETF's RFC 3339
> <http://www.ietf.org/rfc/rfc3339.txt> clearly specifies the limits of
> the date specification in the introduction:
> 
> "All dates and times are assumed to be in the "current era", somewhere
> between 0000AD and 9999AD."
> 
> So you're probably screwed.

Agreed.  I would also submit that, at least for the present day and next
almost 8k years, RFC3339 covers the 80/20 case quite well.

Alternatively, RFC2550 contains a proposal for representing BCE dates.

 http://www.ietf.org/rfc/rfc2550.txt

Specifically, section 3.5 B.C.E. (Before Common Era) Years.

Thanks,

Tantek