[ntpwg] Further to the timestamping issue

[Rob Seaman]

Hi Warner,

I'm not going to buy your arguments here any more than in LEAPSECS,  
where they belong :-)

Interested members of the peanut gallery are directed to the archives.

>> NTP already handles this case

> The ntp protocol does.  The last ntpd I looked at specifically  
> filtered out all leap second indicators that weren't done on the  
> last day of June or December.


As with your message yesterday, these are things to fix, not work  
around.

NTP implements UTC.  UTC currently requires leap seconds.  Legal leap  
second scheduling policies are broader than currently supported by  
NTP.  This is a good thing since it permits the longer range schedule  
desired by many.  Lengthening the schedule is where our efforts should  
be directed.

Rob