[HECnet] How long has your 20 been up?

[Thomas DeBellis]

Thanks for the URL, I'll put it into the enhancement list.  I'm still 
not sure what I would put into the monitor and what would go into a 
light weight (or full) NTP client, but I'll have a think about it.

Passing on to the Department of Pointless Programming, I 'enhanced' 
uptime to handle the maximum 70 bit millisecond uptime.  It is:

    VENTI2 up for 37539161 Millennia, 7 Centuries, 2 Decades, 9 Years, 8
    Weeks, 2 Days, 11 Hours, 35 Minutes, 3 Seconds and 423 Milliseconds

So 37 million millennia is ...umm... well it's a long time.

The math was actually unremarkable, I changed an integer divide 
instruction into a routine to handle single (IDIV), double (DIV) and 
quadruple intermediate results (DDIV), picking the fastest divide that 
wouldn't overflow.

The harder part was rewriting the grammatical routines to properly 
inflect the plural noun case.  Once you get past Years, English  
numerological noun inflection is no longer agglutanive (like Spanish) 
but rather fusional, being more akin to Latin and Italian.  So the 
routine did the wrong thing.

A little more than half the systems programming staff at Columbia were 
bi-lingual, some spoke three or more languages.  So it bugged us to see 
Unix cop-outs like "1 item(s)" processed.  I mean, it's a comparison and 
conditional transfer; an if statement--barely more than 5 minutes' work.

> ------------------------------------------------------------------------
> On 1/20/22 5:21 PM, Peter Lothberg wrote:
>
> Time....
>
>> Not quite.  The acronym actually says that it is not solar time, it's 
>> Greenwich MEAN Time.  UTC is merely the new name for GMT, obviously 
>> adopted to keep the French from fulminating. 
>
> GMT was established by the Greenwich observatory observing the sun.
>
> When the atomic second was created the target was the year 1900 GMT 
> second.
>
> GMT is based on earth position, and it's modern equivalent is UT1,  
> when UTC and UT1 differs more than 0.92 sec
>
> we do leap seconds.
>
> https://www.iers.org/IERS/EN/Science/EarthRotation/UT1LOD.html
>
> UTC is based on a constant running timescale based on the frequency of 
> CS133 oscillating between two
>
> hyper fine states.
>
> So the difference between GMT and UTC can be almost a second, and 
> there is no system that
>
> distributes GMT, BTP, GPS etc are all UTC based.
>
> -P
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.sonic.net/pipermail/hecnet-list/attachments/20220121/62506ea1/attachment.htm>