[Lowfer] DK7FC

JD listread at lwca.org
Fri Jan 17 10:58:25 EST 2014


>>> Interesting how the mind takes partial information, consults the mental 
>>> look up table (morse code elements) and makes an incorrect 'best fit' 
>>> determination. A task like this is clearly better left
to the digital modes ... accuracy wise anyway ;~) .

Well, for sake of accuracy....  At least the human mind sees what it can 
correctly and unequivocally decode without prior knowledge of content (the Q 
and the B) and the subsequent incomplete character, recognizes this 
combination as unlikely and puts it in quotes, then asks for additional 
information.

A machine capable of decoding without prior knowledge would have reached a 
similar partially incorrect solution, but would not have asked about the 
discrepancy.

And a machine dependent on a list of possible messages, having only this 
same sample, would output nothing at all unless the message was already in 
its database.

Maybe it's just me, but I don't think I'm comfortable leaving anything 
entirely to a machine until we build one that can stop and ask, "Huh, can 
this really be right?"  :)




More information about the Lowfer mailing list