Y2K retrospective / was Re: Algol vs ...

Right - But still not as bad as the Wang bug back in 1984 that shut down VS machines! --- On Sat, 2/13/10, Andy Holt <andyh at andyh-rayleigh.freeserve.co.uk> wrote: From: Andy Holt <andyh at andyh-rayleigh.freeserve.co.uk> Subject: RE: Y2K retrospective / was Re: Algol vs ... To: "'General Discussion: On-Topic and Off-Topic Posts'" <cctalk at classiccmp.org> Date: Saturday, February 13, 2010, 12:59 AM Now the more sophisticated systems - of which the British ICT (ICL) was one - used an offset to hold dates.? The ICL 1900 range? used a technique of holding the number of days since the Jan 1st 1900 <<<< Except that they made a minor blunder ... Originally it had been defined that Day 0 was Jan 1st 1900 Then - after lots of such dates were "in the field" - someone discovered that 1900 was not a leap year. (probably the first well-informed person to write their own date-conversion routine who wondered why their code produced different results from the system standard) The only reasonable solution was to redefine Jan 1st 1900 as Day 1 so all dates from Mar 1st 1900 onwards remained unchanged. Andy