On Tue, 2004-10-19 at 20:45, Roger Searle wrote: > this has been bugging me all afternoon. it would appear we're both > partly wrong. i didn't take into account the different number of leap > days between 1900 and 1934 (8) [1], and 1970 and 2004 (9) so that brings > us to the same day. your method is altered by being wrapped by nzdt > (made later by 13 hours). therefore after making the appropriate > adjustments to both methods a reconciliation occurs and the correct time > and date is yesterday morning at 6:09. > > look what 60 seconds does... > > SuSEbox:/home/roger # date -d '1970-01-01 UTC 60 seconds' > Thu Jan 1 12:01:00 NZST 1970 > > makes it the afternoon! obviously it should be Jan 1 00:01:00 > > > [1900 wasn't a leap year, years divisible by 100 are not, unless they > are divisible by 400. hence 2000 was. gives 97 leap years per 400 years]
my answer was done as i was about to leave the office for the day and was a bit rushed. see chris's response, and also you are right, the emerge was started at 06.09 yesterday UTC which was 19.09 NZDT which brings about an interesting point - to gentoo users only - genlop regards the download time as part of the emerge time, and with a 225 odd MB source file at 128 kb/s (about 1MB/minute) that inflates the time. I know this because I went out at 19.15 ish, not long after it started downloading. > > Nick Rout wrote: > > >bzzzt incorrect, not sure if timezones are the problem. > > > >Mon Oct 18 19:09:16 NZDT 2004 > > > >it may have been 6.09 am in Greenwich - no thats still out by a day? > > > >1900 was a leap year and 2000 wasn't, or is it the other way round? > > > >I'll stick to date as proposed by mjg and the date info page :) > > > > > >On Tue, 19 Oct 2004 14:34:43 +1300 > >Roger Searle <[EMAIL PROTECTED]> wrote: > > > > > > > >>putting that into a spreadsheet, which starts counting from 1900 I believe, > >>returns 17/10/1934 6:09am. So I believe your answer is 17/10/2004 6:09am - > >>Sunday morning just gone? > >> > >> > > > > > >
