On 3 Feb 2012, at 2:18pm, Richard Hipp wrote: > On Fri, Feb 3, 2012 at 9:11 AM, Don V Nielsen <donvniel...@gmail.com> wrote: > >> But if we don't start worrying about the year 10,000 now, then we'll end of >> having another disaster like Y2k. > > The date and time functions in SQLite break down on 5352-11-01 10:52:47, > which is thousands of years prior to the year 10,000 issue.
Richard's being funny. Software processing unix-epoch dates as 32-bit integers are going to have a big problem on January the 19th 2038, since they run out of bits then. That's the soonest date we have to worry about. There is currently a lot of such software around, and if we had to deal with that right now we'd be in serious difficulties. But at the speed we're moving to 64-bit architecture it will be less of a problem by the time we actually get to 2038. I assume there'll be the sort of y2k-compliance effort there was for y2k, and that it will be similarly successful. I earned quite a bit of money from y2k-compliance but by 2038 I hope to be retired, so I'll be able to just sit back and laugh at it all. Simon. _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
