wayner@hermod.cs.cornell.edu (Peter Wayner) (11/30/90)
I was re-reading H.F Saint's book _The Adventures of An Invisible Man_ which I heartily recommend to everyone in this newsgroup. Not only is it filled with anti-government sentiment, but the author has a fine wit. It was on the Best Sellers list for a bit. The hero of the story becomes invisible by accident and at one point he is at home when his secretary comes to pick up and drop off papers. He hides by not announcing his presence and is scandalized to discover that she procedes to search his room, look through his drawers, scan his pictures, read his checkbook register and thousands of naughty deeds. He then launches into a solilioquoy (sp) that was most insightful. I will try to bring in the book to post a quote or two tomorrow. He essentially notes that privacy is really a social convention, not something that Knowledge Theorists from Distributed Computing could debate about. It is something in the way that people hide what they know and are considerate, not something that can be modeled with set theory. Most people quickly learn many details about all of the people around them in a small amount of time. People can't hide that much and there is just too much redundant information. Here's a hypothetical, social question. Deaths are a public matter-- not just in the same way that drunk driving convictions are, but in more public sense. Newspapers run obituaries. Even so, you would not mention to the cute girl you're chatting with, "So I hear your Grandfather died back in 1985." This is a private matter. You also wouldn't mention, "So have you dried out since the DWI conviction in 85?" The data's certainly public, but it's not polite. So privacy is not a matter of forbidden information-- it is set of social rules of tact that govern when it is polite to mention a topic. This is the danger of the proposed database laws. They propose to set up a class of forbidden facts about the world (which is a very scary thought) and in most cases these facts can all be infered from public sources. Just my usual two cents... Peter Wayner Department of Computer Science Cornell Univ. Ithaca, NY 14850 EMail:wayner@cs.cornell.edu Office: 607-255-9202 or 255-1008 Home: 116 Oak Ave, Ithaca, NY 14850 Phone: 607-277-6678