From kragen@dnaco.net Mon Aug 10 14:22:54 1998 Date: Mon, 10 Aug 1998 14:22:53 -0400 (EDT) From: Kragen To: Chet Farmer cc: rebecalist@bossanova.com Subject: Re: [rebecalist] the truth behind the amazon buying spree In-Reply-To: <3.0.5.32.19980810111013.00864100@pop.pdq.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Keywords: X-UID: 1168 Status: O X-Status: On Mon, 10 Aug 1998, Chet Farmer wrote: > > At 11:25 AM 8/10/98 -0400, you wrote: > > [with regard to Internet-caused political revolution] > >It already happened in Indonesia. > > > >(I'm not suggesting that we'll have a political revolution in the US. > >But Singapore, PRC, Myanmar, etc. -- definitely.) > > How long, though, before this happens? Doesn't Singapore still > use a bank of proxies to prevent "unsavory" content from penetrating > their borders? I'm sure they do. Nevertheless, there are people in Singapore who use IRC. > In a revolutionary situation, faxes are probably more useful because > they're not "filterable" data -- as images, they can fly around pretty > much willy-nilly, as they did during Tiennamen (sp) Sq. Email can > still be filtered, and as the PRC gets more serious about controlling > the Net within its borders I think we'll see the supposed power of the > Net attenuated tremendously. Email can contain images as easily as text, and emailed images are just as difficult to control as faxes -- if a little less common at the moment. Email also has the major advantage of being much more difficult to trace than faxes are, and much easier to send to groups of a hundred people or so. The Indonesian revolutionaries communicated mostly through Hotmail and such places. > Widespread change on other fronts will > occur before Internet deregulation helps to accelerate it -- it won't be > the catalyst. They know what it is, and they don't like it, and they're > in a position to control it, at least within their own borders. Myanmar is taking exactly this tack. But the countries that forbid or restrict Internet access -- we will bury them economically! *bangs shoe on table* There are only two options: allow unrestricted Internet access, or be outcompeted by your freer neighbors. This is also the choice the US faces with regard to copyright law and cryptographic law. (The US may be able to force its neighbors not to outcompete it, though, by trade sanctions and violence. I hope not!) > Checking references involves some degree of deconstruction and > lead-chasing in and of itself, though. Sure, you've got a thread to > pull, but it may well be a dead end -- and there will be _so many_ > of these "bullshit scenarios" floating around that finding the isolated > grain of wheat in megabytes of chaff may prove barrier enough to > using any of the presumably freely available classified data, even if > the data point you want is even out there. Hide in plain sight, as it were. The scenario you describe is already a fairly good description of the Web, of AOL, and indeed, of person-to-person conversation. Most of the available information is bullshit. It is not an intractable problem; social networks handle this problem already. Not perfectly, but well enough. Here's the way it works. I have a credibility in my mind associated with each person I hear from. If someone with good credibility (Peter Gutmann, Linus Torvalds) says that X is true, I am likely to believe X without checking it. If someone with poor credibility says that X is true, I am likely to not even bother to consider the possibility. (Of course, other factors enter into this. If Linus Torvalds told me the Martians had landed, I would probably dismiss it out of hand -- although I might check.) There's a band of "intermediate credibility". If someone whose credibility is neither so poor that I ignore them outright nor so good that I trust them without checking says that X is true, and X is important to me, I will investigate X. If I do a lousy job of investigating X, I might pass on false information, or fail to pass on correct information. If I do a good job of investigating X, I will be less likely to make these mistakes. I bias my responses so that I'm more likely to fail to pass on correct information than to pass on false information. If I decide that someone passed me false information, they lose credibility in my book. If I decide that someone passed me correct information, they gain (a much smaller amount of) credibility in my book. If I'm not sure, their credibility is unchanged. My own credibility in the eyes of others will depend on how well I investigate things and how well I make my trust decisions -- as well as how well people who listen to me investigate things. When I find people who are very credible, I'm likely to devote more of my time to listening to them. People who develop a lot of credibility with other people become well-listened to. It's possible for many groups to form and maintain beliefs that contradict other groups' beliefs, and give little credibility to members of other groups (since they say things that appear to be false), if people in general either do not investigate very accurately or trust too much. However, with a sufficient (fairly low) level of investigation and reasonable levels of trust, the truth will emerge, even out of huge piles of nonsense. Better yet, since you can often find out who other people trust, there will be some people who have very high credibility with a huge number of other people. I trust the Wall Street Journal partly because my grandfather does. > Imagine I give you a sheet of paper with a ten thousand phone numbers > on it, each one supposedly belonging to Uma Thurman -- with an > attribution for each. This would be a great burden for me, but it would be an easy job for a hundred organized Uma fans, or a thousand disorganized Uma fans. With this logic, I hypothesize that Uma Thurman's home phone number is available on the Internet today. (Although what would you do with it? Call and harass her answering service?) Kragen