[LMB] OT: Timezones on Barrayar

Paula Lieberman paal at gis.net
Mon Jul 6 04:02:13 BST 2009


----- Original Message ----- 
From: "Greg Hennessy" <greg.hennessy at cox.net>
To: "Discussion of the works of Lois McMaster Bujold." 
<lois-bujold at lists.herald.co.uk>
Sent: Sunday, July 05, 2009 10:47 PM
Subject: Re: [LMB] Timezones on Barrayar


>
>>> Me, I'd rather just get all the watches made with a software adjustment 
>>> so
>>> we could allow adjustments in the day length via extra or shortened
>>> midnight
>>> hours.
>>
>> NONONONONO!
>>
>> "Timing and synchronization are ALWAYS problems."
>
> Even GPS, Glonass, and Galileo (the upcoming european system) have 
> software
> adjustments for leap seconds. In theory you could do things like daylight
> savings adjustments, but those aren't done in GPS today. Although WWV does 
> have
> a DST indicator bit.

DST is a noxious abomination.... it can even cause software to crash in some 
cases.

The adjustments for leap seconds involve the Master Control Station 
uploading stuff to the satellites... it doesn;t' change the high precision 
onboard clocks, it adjusts the downstream timing chain....

The whole idea of a timing system is to try as much as possible to 
have -consistent- UNAMBIGUOUS system time.  The #O(Y#$#@DE! DST had an 
overlapping NON unambiguous hour in the "fall back" situation of the 25 hour 
day... the noxiousness involved is extremely noxious.  The 23 hour day at 
least has no ambiguity to it, there's a discontinuity of an hour with the 
skipped hour, but it'snot repeated hour....

For records keeping purposes, duplicate hours are horrible things... 




More information about the Lois-Bujold mailing list