Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!purdue!haven!udel!rochester!uhura.cc.rochester.edu!uhura!saaf From: saaf@joker.optics.rochester.edu (Lenny Saaf) Newsgroups: sci.electronics Subject: Re: detecting TV commercials, is it possible? Message-ID: Date: 16 Sep 89 03:14:57 GMT References: <4482@natinst.natinst.com> <78@uwm.edu> <3553@dgis.daitc.mil> <1669@ns.network.com> Sender: news@uhura.cc.rochester.edu Organization: The Institute of Optics, University of Rochester Lines: 33 >> through the user's guide and the principle that it used was that >> the _only_ time that a total lack of video signal was present was during >> the pause right before a commercial was to start. > It is true that you often will see "black" between program and > commercial, it is certainly not a requirement -- but worse still is > that you might often see "black" between one scene and the next, > especially if they want to highlight a time or location transistion. My idea would be to have the TV on a tape loop similar to the ones used on radio talk shows. [Similar, that is, in purpose: to eliminate vulgarity.] The loop should be longer than the longest commercial, which might be 60 seconds. The "blurb-zapper" would look for short periods of black silence that are spaced by the length of commercials, e.g. 15, 30, and 60 seconds. This would avoid most of the problems with black silence occurring during a program. When you find such a segment, you can replace it with whatever you like when it comes through on the tape loop. Two questions: 1. What would you replace the commercials with? Music? A video tape of wildebeasts thundering across the Kalihari? Public service announcements? Silence? 2. Don't people who watch commercial television like commercials? If they didn't everyone would watch PBS, right? -- ------------------------------------------------------------------------ | Len Saaf, The Institute of Optics, Univ. of Rochester, Rochester, NY | | Internet: saaf@joker.optics.rochester.edu Bitnet: SAAF@UOROPT | ------------------------------------------------------------------------