[ntp:hackers] Autokey specification report

David L. Mills mills at udel.edu
Fri Jan 20 19:36:55 UTC 2006


Folks,

The Autokey specification has been in the IETF standards pipe for almost 
four years without definitive result. A look at the recent NTPv4 
algorithm specification is ample proof that the Autokey document would 
be completely butchered in Postel ASCII format. So, I am formally 
withdrawing the submission as IETF standards document and republishing 
it in PDF format as UDel Technical Report 06-1-1, now linked from the 
NTP project page and at 
http://www.eecis.udel.edu/~mills/database/reports/stime1/stime.pdf. I 
have no plans to publish it in any other format.

A report on the NTPv4 on-wire protocol is under preparation in the same 
format and will be published as a UDel technical report in PDF format. I 
have no plans to publish it in any other format.

As you can tell from the tone of this message, I believe that the Postel 
ASCII format is completely irrational in today's publishing community. 
Can you imagine some poor soul actually implementing one or another of 
the NTPv4 algorithms using only the text in the algorithms proposal? 
This is not in any way to diminish the content of that proposal, just 
the butchery of equations and diagrams in that format. I tried to verify 
the correctness of the equations and simply could not untangle the notation.

We as a publishing community have moved far beyond the days of runoff 
and fixed-pitch printers. The vast bulk of technical publishing today is 
in PDF, which fits the needs of the technical engineering and computer 
science community for equations, diagrams and photographs. Postel ASCII 
is an obstacle in communication and in fact a hazard to progress.

Dave


More information about the hackers mailing list