This happens because the keeper of that stats page is either not aware
that the dead remailers have officially left the network, or is behind
on maintaining his stats pages. Often stats pages are maintained by
remailer operators and their other remailer duties will take priority
over updating their stats pages.

Unfortunately, a poorly maintained stats site will often lead to
problems for remailer users when they plug those URLs into programs
like Jack B. Nymble, Private Idaho or QuickSilver and receive
configuration error messages. The problem will repeat itself every
time your stats are updated and the site with the stale keys is
polled.

One attempt to provide information on the problem described above is
found in Computer Cryptology's Comparison tables:

<http://www.eskimo.com/~turing/remailer/stats/db/rlist.html>
<http://www.eskimo.com/~turing/remailer/stats/db/mlist.html>

Further information is available at Frog's MetaStats Page
<http://www.privacyresources.org/frogadmin/MetaStats/index.html>.

Two comparisons between stats sources can help in selecting pingers to
include in a remailer client's list.  First, the "Last update" time
indicates if the values are current.  Second, a comparison of the
remailers each stats source list can indicate if the problem described
above is occurring.  For example, imagine a particular stats source
lists many remailers at 0.00% reliability ("uptime"), is the only
pinger to list several remailers, and doesn't list a dozen remailers
that the majority of other pingers include.  That stats source is
probably out of date, even if the "Last update" time is current.

For further consideration, the cells in Computer Cryptology's
Comparison shade to indicate suspected error values or outliers, i.e.,
values far from the average of other stats sources.  The details are
on those pages.

Similarly, Frog's "All Pingers' Index Page"
<http://www.privacyresources.org/frogadmin/Pingers.html> rates pingers
as up to date versus poorly or not maintained.  These ratings express
the opinion of the operator.

------------------------------

End of faq.5 Digest
*******************