[time-nuts] Fwd: IFCS-EFTF 2019: Call for Papers

[Julien Goodwin]

On 11/10/18 09:55, Magnus Danielson wrote:
> Hi Chris,
> 
> On 10/10/18 4:36 PM, Attila Kinali wrote:
>> On Sat, 22 Sep 2018 06:42:59 -0500
>> Chris Howard <chris at elfpen.com> wrote:
>>
>>> I read these "calls for papers" things and try to think of
>>> something "meta" since I am not anywhere near the 'bleeding edge'
>>> of any such topic.  Maybe a survey of the number of NTP-using
>>> devices in the world  (people love charts and statistics),
>>> or something about the history of something.
>>
>> You don't need to be at near the bleeding edge for a successful
>> paper. There are still a lot of questions that nobody has asked,
>> much less answered. A lot of these questions arise only when
>> you work in the trenches and try to properly understand what
>> is going on. So, if you think you have something that nobody
>> has writen about before, please do so.
> 
> I agree fully.
> 
> As you come in and view things with a fresh pair of eyes, seeing
> problems, trying to make sense of things you bring up topics that others
> overlooked. Very often you get "We never spent much time on that" or
> similar comments from the folks you thought had checked *EVERYTHING*.

In my own day job area of data networking "doing X in practice" where
all previous publications have been theory, or at least simplified below
most practical uses is often a huge area for submissions.

Relevant to time-nuttery every year or so I get in discussions with
people doing wide-area time synchronisation and end up having to explain
some of the many ways in which networks don't actually work in the ways
people think they do, one classic that I'm fairly sure has been
mentioned on this list is that if you assume the length of a long-haul
(kilometers+) fiber span is constant (even on the order of seconds)
you're going to have a bad time.

There's a good paper topic just characterising a few of the high level
reasons for that.