[Info-vax] Why is starting epoch 17 Nov 1858?

[Dave Froble]

On 3/10/2021 3:30 PM, VAXman- at SendSpamHere.ORG wrote:
> In article <s2aqvh$o7s$1 at dont-email.me>, Stephen Hoffman <seaohveh at hoffmanlabs.invalid> writes:
>> On 2021-03-10 14:21:17 +0000, David Jones said:
>>
>>> On Tuesday, March 9, 2021 at 3:02:13 PM UTC-5, Jim DeCamp wrote:
>>>> On Monday, November 27, 1995 at 3:00:00 AM UTC-5,
>>>> zrep... at cuppa.curtin.edu.au wrote:
>>>> It's called Smithsonian time, and yes it is the epoch for the Modified
>>>> Julian Day, which is the Julian Day minus 2,400,000.5 days.
>>>
>>> Is that 2400000.5 magic number ensconced in a header file symbol
>>> definition anywhere?
>>
>> It's been a while since I've looked at that code, but I wouldn't expect
>> to see that value around other than in comments, as it is immaterial to
>> the process of converting to and from the base date.
>>
>> Converting from 0 is 17-Nov-1858, and from 17-Nov-1858 produces 0. A
>> day's accumulated centiseconds gets you 18-Nov-1858, obviously.
>>
>> That there's a second and earlier date with a nice round offset from
>> the any base date isn't any more relevant to the conversion process
>> than would be a nice round offset date after the base date.
>>
>> Neither the Smithsonian base date nor the Gregorian base date is
>> anything other than a convenience, particularly given we don't know the
>> offset from the Big Bang (yet?), and usually wouldn't want to store it
>> everywhere anyway.
>
> That might require implementation of  planck-time units and many many
> many more bits than current implementations possess. ;)
>

And next we'll be throwing relativistic effects on time into the mix?

Maybe better to keep it simple.  If some user needs more, let them 
implement it in their apps.

-- 
David Froble                       Tel: 724-529-0450
Dave Froble Enterprises, Inc.      E-Mail: davef at tsoft-inc.com
DFE Ultralights, Inc.
170 Grimplin Road
Vanderbilt, PA  15486