[ntp:questions] Re: Trimble Setup Question

Brad Knowles brad.knowles at skynet.be
Sat Aug 30 20:14:41 UTC 2003

At 7:52 PM -0600 2003/08/29, Vernon Schryver wrote:

>  The problem here is that Dave Mills has been too competent for too long.
>  He keeps plugging away on the protocol and a series of implementations
>  that have been better than what anyone else has been able to or cared
>  to take the time and effort to produce.

	Dave Mills is not the sole source of everything related to NTP, 
or all positive things in the world related to NTP.  There are a 
number of people who have been contributing to NTP for many years. 
Some of the more recent active contributors are shown at 

>  Good, solid work attracts people who would be committee chair and tell
>  whoever has been doing the work the way it should have been done and
>  how to fix it.

	Harlan has been the release engineer for NTP for quite some time 
(as far as I can tell, since at least September of '97).  I don't 
know how long Steve and Danny have been involved, but my 
understanding is that it has been for years.  There are other names 
on the list of contributors that I have seen before, but I have never 
received any e-mail, irc messages, or other forms of communication 
from them -- there's no telling how long they've been involved with 
the project.

	Myself, I am a more recent addition to the team, and I do not yet 
consider my contributions to be significant.  However, I hope that 
one day I will be able to say that I have done so.

>                                     I tend to tell everyone "Here's
>  the source.  Do whatever you want but don't ask me to endorse (what
>  you say are) your ideas by making them work."

	I'd be perfectly happy if that was the kind of relationship we had.

>  For better and worse, I'm proud to wear the many "Non-Team Player"
>  badges I've been awarded by people felt that their winning ways
>  and ambitions required them to appoint themselves my boss.

	No significant changes occur to the ntp.org site or to the NTP 
code without some consensus.  If any of the contributors feels 
strongly about the subject, we let the others know and we discuss the 

	The only thing that I am aware of which has come to light 
recently and which has been seriously questioned was the use of 
reverse DNS checks in deciding whether or not to accept connections 
from mail servers.  That was a result of changes that were made back 
in April (or earlier), before I was a member of the team, changes 
which were in conflict with an unwritten UDel EECIS policy that I 
could not possibly have known about.

	General consensus has been that the installation of mailman was a 
huge improvement.  In one fell swoop, it gave us many of the features 
that had been requested a long time ago, and it brought along other 
capabilities that were heartily welcomed.  There is no way we could 
have easily gotten all the functionality we wanted out of majordomo, 
although we could have gotten closer by the installation of several 
additional packages from other sources.  Steve did most of the work 
on getting mailman working correctly.

	The installation of bugzilla has been a huge improvement for us. 
We now have a system that we can use to track bug reports, request 
improvements, etc... for the NTP code itself, the ntp.org site, or 
the FAQ.  It allows us to have an instant view of all the activity on 
anything in the system, and if certain tasks need to be split, 
joined, or handed over to other personnel, we can easily do that. 
ISTR that Steve and Harlan were primarily involved in this effort.

	While it has not been as widely publicized, I have found TWiki to 
be very useful in documenting the system as it exists today, so that 
we can have a reference going forward of what all bits exist where on 
the system, at least as far as anything related to e-mail is 
concerned.  I believe that this will continue to be a tool that is 
heavily used within the group of contributors to ntp.org, and I 
believe that the community as a whole would be well served by 
becoming more familiar with it, with interested parties being able to 
provide their own direct contribution back to the project.  Steve 
installed TWiki, and I believe that he has agreed to be the TWiki 
Editor, to help ensure consistent content and quality throughout the 

	I know that I could not have survived the recent SoBig.F attack 
without mailman.  If we had tried to do this with majordomo, I would 
have been toast.  Moreover, I believe that many members of this list 
would probably have become infected through the list, thus 
compounding the problem further by many orders of magnitude.

	I don't think any of us have tried to appoint ourselves as the 
boss.  We have tried to do the best jobs we can to support NTP, the 
ntp.org site, and the community as a whole.  Throughout our work, we 
have been applying best industry practices to the best of our ability.

	Nevertheless, we have been told that the things we are doing are 
completely value-less and we shouldn't be wasting our time on them. 
Instead we should be doing something "useful".  Indeed, the 
implication is that we should not be allowed to waste our time on 
such useless endeavors.

	How else would you react?

	It is my feeling that your comparison of the situation to 
Stalinist Russia is done out of ignorance, and is inappropriate.  If 
you wish to become a contributor and get your name listed at 
<http://www.ntp.org/contributors.html>, that is a different matter.

	However, I would encourage you to avoid making a fool out of 
yourself through ignorance.

Brad Knowles, <brad.knowles at skynet.be>

"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety."
     -Benjamin Franklin, Historical Review of Pennsylvania.

GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI++++$ P+>++ L+ !E-(---) W+++(--) N+
!w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++)
tv+(+++) b+(++++) DI+(++++) D+(++) G+(++++) e++>++++ h--- r---(+++)* z(+++)

More information about the questions mailing list