C.E. SHANNON - A Mathematical Theory of Communication

C.E. SHANNON

"A Mathematical Theory of Communication." This paper is poorly motivated and excessively abstract. It is unclear for what practical problem it might be relevant. The author claims that "semantic aspects of communication are irrelevant to the engineering problems," which seems to indicate that his theory is suitable mostly for transmitting gibberish. Alas, people will not pay to have gibberish transmitted anywhere.
I don't understand the relevance of discrete sources: No matter what one does, in the end, the signal will have to be modulated using good old-fashioned vacuum tubes, so the signal on the "'channel"' will always be analogical.
A running example would have helped make the presentation clearer and less theoretical, but none is provided. Also, the author presents no implementation details or experiments taken from a practical application.
Confidential comments to the editor: The only thing absolutely wrong with this paper is that it doesn't quite "resonate" with what the research community finds exciting. At any point, there are sexy topics and unsexy ones: these days, television is sexy and color television is even sexier. Discrete channels with a finite number of symbols are good for telegraphy, but telegraphy is 100 years old, hardly a good research topic.
The author mentions computing machines, such as the recent ENIAC. Well, I guess one could connect such machines, but a recent IBM memo stated that a dozen or so such machines will be sufficient for all the computing that we'll ever need in the foreseeable future, so there won't be a whole lot of connecting going on with only a dozen ENIACs!
IBM has decided to stay out of the electronic computing business, and this journal should probably do the same!