Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!wuarchive!gem.mps.ohio-state.edu!apple!voder!dtg.nsc.com!andrew
From: andrew@dtg.nsc.com (Lord Snooty @ The Giant Poisoned Electric Head       )
Newsgroups: comp.ai.neural-nets
Subject: Re: Data Complexity
Summary: the message is not the message
Message-ID: <184@berlioz.nsc.com>
Date: 23 Oct 89 01:58:24 GMT
References: <4522@imagen.UUCP>
Organization: National Semiconductor, Santa Clara
Lines: 16


I guess the entropy calcs use the p.ln(p) expression, as per Shannon.. but
as per Hofstadter, what about the complexity of the context or "frame"?
What tells you which jukebox will play your record? If you could give
da Vinci a binary dump of a set of Postscript graphics files describing
his drawings, could he work out that they were his? - or would you have
to provide him with the appropriate "jukebox" (PC + OS  + etc)?

Of course, the correct answer is the latter one. So my question is:
have you included the complexity of the decoding apparatus in your entropy
estimates?

-- 
...........................................................................
Andrew Palfreyman		and the 'q' is silent,
andrew@dtg.nsc.com		as in banana			 time sucks