Re: Predictable Internet Time

[Nico Williams <[e-mail]>]

On Wed, Jan 04, 2017 at 12:32:23AM +0100, Philip Homburg wrote:
> Unfortunately, there doesn't seem to be an IETF equivalent for operating
> systems, so it will be completely random whether anything will happen in
> this area or not.

There's no reason that the IETF could not a) define an abstract time API
that includes time "types" ({UTC, TAI, local, smeared, whatever} x
{various possible representations}), conversions between them, and
behaviors, then b) define bindings of this API for various programming
languages.  The Open Group, to give one example, could take our advice
on this matter -- or not; they'd not be obligated, natch.

The IETF has done this for other networking technologies (sockets
extensions for IPv6, GSS-API, ...).

Because our protocols must deal with time, this is very much in our
bailiwick.  It also fits in other SDOs, but as none have published such
a spec, we might lead.

Any RFC on smeared time surely would at least define a subset of this
abstract API for conversions to/from UTC.  We might as well go all the
way.

Nico
--