William H. Blair's discussion of some of the proto-history of what came to be
called the Y2K problem is of great interest. In particular,
| There was some question which representation would be
| best to represent dates>1999: yyyydddF or 0cyydddF...
is of interest because, like almost all discussions of this problem, it omits
to distinguish between internal and external data representations of dates.
None of us supposes that an approximation to the internal representation of a
signed binary halfword having the decimal value +3, viz., 0000000000000011b, is
an appropriate external representation for use by most people at any time or
indeed by any people most of the time. Signed decimal integers have a hallowed
external representation that has two not very controversial characteristics:
It is the appropriate one for data entry and display, and
It is radically inappropriate for internal, computational use.
Analogously, 2009 September 27, 20090927, and the like are appropriate to data
entry and display, and radically inappropriate to computation. For computation
internal to a computer a signed fullword Gregorian-Day (GD) value that reflects
the correspondences
0000 December 30 <==> -1
0000 December 31 <==> 0
0001 January 01 <==> +1
is the appropriate one.
It permits all dates 5.87 million years before and after 0001 January 1 to have
a simple four-byte representation, and it trivializes date arithmetic.
Moreover, conversions of the traditional data-entry/display formats to and from
GD values are trivial too.
Astronomers, who use Julian-Day (GD) Numbers, which are much like JDs except
that they have the [Gregorian] epoch origin -4713 November 24, have long
experience with date arithmetic; and their experience should have been, but was
not, controlling.
The Y2K problem was indeed a laughable one in many ways, but it was also a
botched opportunity. Its "solutions" were dominated by the dubious notions of
people who, with apologies to the shade of C. Wright Mills, it is appropriate
to call crackpot realists.
John Gilmore, Ashland, MA 01721 USA
_________________________________________________________________
Bing™ brings you maps, menus, and reviews organized in one place. Try it now.
http://www.bing.com/search?q=restaurants&form=MLOGEN&publ=WLHMTAG&crea=TEXT_MLOGEN_Core_tagline_local_1x1
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
