[ntp:questions] The NTP dictionary
David L. Mills
mills at udel.edu
Fri Dec 28 19:39:29 UTC 2007
Folks,
I've been collecting words for the NTP dictionary that has evolved over
the last two decades. The vocabulary is a subset of what some editors
have come to call millsspeak. If you read my papers, book and messaging,
you know what they mean. The vocabulary is routinely abused; new words
have been fabricated, old ones purposely misused and the rules of
English cheerfully ignored.
Following is an initial list probably blamed on me. I invite further
submissions and interpretations. However, they must be peculiar to the
NTP community.
chime. A synonym for synchronize, as in: "That server gives good chime,"
or "Can I chime your server?"
timewarp. An unstable server, as in "Your server threw a timewarp last
night and all my applications tanked."
truechimer. To a computer scientist a loyal Byzantine general. To a
computer timekeeper a trusted clock.
falseticker. To a computer scientist a traitor Albanian general. To a
computer network timekeeper an outlyer or two-faced clock.
huff-n'-puff filter. An algorithm used to improve chime when either the
outbound leg or inbound leg, but not both, has extreme jitter.
popcorn spike suppressor. An algorithm used to delete infrequent but
very large offset or delay spikes.
kiss-o'-death packet. An NTP packet with leap indicator and stratum 0
resulting from an error condition at the server. The reference ID field
contains the error code, called a kiss code.
Autokey. To a classic cryptographer a particular type of chaining
cipher. To a computer network timekeeper a specific security protocol
Manycast. A derivative of the anycast paradigm where the result is a
plurality of servers, not just a single one.
tally code. The first character in the ntpq pe display for each server.
flasher. The bits of flash field of the ntpq rv display for each server.
provenentic. Formally, this is the relation formed from the transitive
closure of the authentic relation. This comes from Autokey slang.
Dave
More information about the questions
mailing list