I've got good news and bad news regarding the log checking in the
1997 CW Sweepstakes. The good news is you can get your error list
from Billy Lunt at ARRL. The bad news is that in 13 of 30 QSOs
marked as errors, tapes confirmed I did not make an error (a 43%
error rate on the error rate.) In fact, in 3 of 10 cases where I
was penalized for a wrong PREC, QST listed the PREC as I logged
it. The "master" on the error list had the PREC wrong!
I had a total of 32 QSOs removed from my log. 10 were for wrong
PREC, 18 were for bad CHECK and 4 were for miscopied SEC. I was
pleased I had no busted calls or wrong serial #. I was able to
listen to the first 30 QSOs marked as errors on tape. I use a VCR
to record both radios in sync, 6 hours to a tape. The fifth tape
with the last 2 errors wasn't recorded. Either I forgot to change
the last tape or forgot to push record.
Of the 30, I was able to confirm from tape that I had, without
any doubt, correctly copied the PREC 7 times and the CHECK 6
times. It is interesting to note that of those 13 QSOs improperly
removed from my log, 7 were multi-ops and 5 of those were on
CHECKS. I suspect the operators were sending their own different
checks, because some weren't even close to what the error report
said they should be.
Not every error in my report was the fault of the log checking.
In two cases I asked for a fill and then a confirmation and
received a false confirmation of what I copied on the fill. In
one case I asked for a confirmation of serial #, but still blew
the section. In another case I was sent a dit dit dit dit dah
dah, and recorded it as 4 instead of asking for a fill and
getting a 3.
The point of this missive is to state that as much as I want
tight log checking and the benefit of learning from my errors,
there has to be a feedback loop to reduce error checking errors.
The only thing that comes to my mind as a solution is for
operators who are gunning for the top 10 boxes to register
recordings of contests by a log due date. Error reports (UBN
reports) should be generated and made available as soon as
possible. Each operator then has the opportunity to listen to
their tapes and flag error report errors. The audio of the errors
can be submitted for adjudication with random testing against the
original recordings to avoid digital fraud. Now, the first
reaction to this is "who is going to spend all the time checking
this stuff." Well, each operator who records his contests will,
cause why record them if its not going to help confirm your score
and if you're not going to listen to learn from your mistakes
when they are pointed out to you. If 50 operators submit 15
errors each, that is 750 tightly edited QSOs. At a minute each,
that is 12.5 hours of listening time. Not an enormous effort to
significantly raise the confidence in the results.
In my particular case, the 13 QSOs that were improperly removed
from my log wouldn't have made a difference in the standings.
I am sure that will not be the situation with other logs.
You can bet I'll be asking for more fills on PREC and CHECK next
year.
73. Ken, K6LA - Ken Six Los Angeles, KWIDELITZ@DELPHI.COM
ex-AB6FO
--
CQ-Contest on WWW: http://www.contesting.com/_cq-contest/
Administrative requests: cq-contest-REQUEST@contesting.com
|