DejaNews and Alta Vista Search Tools, and Privacy Implications
(Side note: I seem to have returned from my holiday away from this area and away from computers to a war zone, with multiple flames and threats of legal action. Being a part of the Cypherpunks group, and potentially part of the "criminal conspiracy" to defame, am I included in the lawsuits? I wonder.) Anyway, a point of clarification of a point, lest there be the belief that _all_ Cypherpunks are opposed to comprehensive Web search tools: At 8:46 PM 12/23/95, Steven L. Baur wrote:
Furthermore, no one has mentioned the positive changes made to ^^^^^^^^? Dejanews since it got bashed thoroughly on this list a few weeks ago. They've significantly turned down the amount of old information indexed, and have restricted the groups (and mailing lists) they archive.
I for one don't consider this to be a positive development. Reducing the time horizon for searches has no real effect on the compilation of dossiers (for example), but certainly makes DejaNews less useful. (And I'd be willing to bet that the time horizon was scrunched down for space and time reasons, not for reasons of privacy; the horizon will likely _increase_ as users ask for, and perhaps are willing to pay a bit for, longer time horizons.) And I don't believe the dominant thinking of folks who commented was that DejaNews was worthy of "bashing." In fact, I found it all very interesting, and a confirmation of what many of us expected would soon happen, i.e., fast access to past comments. I think I and several others commented on the major implications for privacy, especially vis-a-vis the way corporations will be able to see compilations of postings to "outrageous" groups. Indeed, I know of some people hiring programmers who are already using such tools to get a better understanding of whom they may be hiring, or not. But my comments were not in the vein of "something has to be done," but of recognition that a Brave New World is fast unfolding. Thinking that one is "safe" because a particular search service is not including all the groups or mailing lists it _could_ include is illusory (one is reminded of ostriches....). The same thinking happened several years ago when a great hue and cry in the media caused Lotus to abandon plans to sell a CD-ROM to individuals with publically available census and other data on it about neighborhoods, phone numbers, etc. Inasmuch as the non-individual entities (corporations, mass mailers, courts, law enforcement, etc.) already have full access to such databases, all the hue and cry really accomplished was to give individuals a false sense of security and privacy. A triumph of feelgood style over substance. Real privacy and security comes from steps taken to make the information private in the first place, not to ex post facto limit access. (I am not claiming that Steven Bauer or anyone else on our list is calling for laws to limit Web search engines, just giving my views about this. As a matter of fact, however, I am hearing rumblings in other places that "there ought to be a law" about these archives, indexes/indices, etc. Same old story. Kind of hard to enforce such laws when the indexes are in Holland, or Byelorussia, or "somewhere in cyberspace." ) Face it, every single word written by any of us to any Usenet newsgroup, going back to the beginning of Usenet, and expanding out to many ostensibly-private mailing lists, will fairly soon be searchable. (Add some digital cash and proxy/remailer features, and someone will be incentivized to put some really big arrays of optical disks up for searching. And if the U.S. tries to "regulate" such searches....well, I'm preaching to the choir here....) --Tim May Views here are not the views of my Internet Service Provider or Government. ---------:---------:---------:---------:---------:---------:---------:---- Timothy C. May | Crypto Anarchy: encryption, digital money, tcmay@got.net 408-728-0152 | anonymous networks, digital pseudonyms, zero Corralitos, CA | knowledge, reputations, information markets, Higher Power: 2^756839 | black markets, collapse of governments. "National borders are just speed bumps on the information superhighway."
I was interested in your comments re DejaNews because of an experience that I had about a month ago. I received mail from Rick Broadhead, a co-author of the Canadian Internet Guide. He wanted to know if I was the Tim Philp who wrote a computer column for the Brantford Expositor and if so, would I be interested in reviewing his new books. It seems that he got my name from a clipping service because I had mentioned his first book in my column. As I did not include my E-mail address in my column, he used DejaNews to search for Tim & Philp. Bingo! my name poped out from an old Cypherpunks list posting. In this case it was good as I was interested in talking to him, but the possibilities for privacy invasion are great. In any case, I got an article out of the experience. I think that it will be interesting when some of the future politicians who are now just kids playing on the Internet are presented with some of their juvenile rantings when they are 40 years old and running for office. The mind boggles! Regards, Tim Philp Brantford, Ont., Canada =================================== For PGP Public Key, Send E-mail to: pgp-public-keys@swissnet.ai.mit.edu In Subject line type: GET PHILP ===================================
[Re: regulating dossiers et. al.] I'm sure that Phil Hallam-Baker has brought this up before, but there is a fair body of law on this subject internationaly. Probably the most restrictive is the UK Data Protection Act which regulates storage of personal information on computers (but not on paper). The DPA has a lot of holes, and seems to be honoured more in the breach than in the observance; other european countries have similar requirements. Of course, the UK has no FOIA, and won't until after the next election, so the most interesting databases aren't available. Simon ----- (defun modexpt (x y n) "computes (x^y) mod n" (cond ((= y 0) 1) ((= y 1) (mod x n)) ((evenp y) (mod (expt (modexpt x (/ y 2) n) 2) n)) (t (mod (* x (modexpt x (1- y) n)) n))))
participants (3)
-
Simon Spero -
tcmay@got.net -
Tim Philp