Re: [LEAPSECS] Precise time over time

From: Rob Seaman <seaman_at_noao.edu>
Date: Thu, 11 Aug 2005 11:17:01 -0700

On Aug 11, 2005, at 9:51 AM, Poul-Henning Kamp wrote:

> The problem is not NTP or the quite clued people using it, the
> problem is the people who don't, can't or won't.

There is not one problem, there are interlocking hordes of problems
spanning the full richness of all human activities. Anecdotal
stories, even ones less awkwardly chosen than the robotical
slaughterhouse, are not sufficient to base such a fundamental policy
decision. The only thing that ties together all these vastly
different problems - use cases - is their common context of being
embedded in the real world. A solution like that proposed by the ITU
that ignores the real world is not acceptable. Justifying any policy
by pandering to the lowest common denominator (in this case, clueless
programmers implementing incomplete designs for incompetent managers)
is doomed to failure.

- If there are risks - and there are - they need to be enumerated for
all options.
- If there is an economic impact - and there is - it needs to be
calculated to include all contingent expenses.
- If there are requirements - many, many requirements - they need to
be evaluated in their real world context.

A hurried solution that ignores pertinent input is extremely unlikely
to prove acceptable.

Rob Seaman
National Optical Astronomy Observatory
Received on Thu Aug 11 2005 - 11:22:46 PDT