[CQ-Contest] logging accuracy and master databases

[Pete Smith]

Kudos to K9TM and K8CC for taking on the task of generating master databases.

It's a tricky thing to get accuracy.  Unfortunately, the starting point 
(people's logs) can bring with it a lot of chaff.  For example, I've been 
told that the CQWW SSB log-checking database shows about 97,000 calls, of 
which only ~30,000 are good calls.  In other words, 2/3 of the call-signs 
that could be gleaned if you had access to everyone's logs over a number of 
years would be bad!

I'm sure that the logs submitted to K9TM and K8CC will be a lot cleaner 
than that, and techniques will be applied to screen the unique/probably bad 
calls out of that input.  But even then, a lot of the common busts -- H for 
S on CW, for example -- will undoubtedly sneak through.

Ironically, I find that rather helpful.  Whenever I'm tempted to rely too 
heavily on the database, I need only look at the screen when I'm part-way 
through entering a call and see what look like two or three variations on a 
single call.  Or are they different calls?  Better just copy the station 
and be sure!

73, Pete N4ZR

Check out the World HF
Contest Station Database at
www.pvrc.org