From rsalz@bbn.com Thu Aug 15 11:01 PDT 1991 Return-Path: Received: from PINEAPPLE.BBN.COM by weber.ucsd.edu (15.11/UCSDGENERIC.4) id AA06840 to bjones; Thu, 15 Aug 91 11:01:38 pdt Received: from LITCHI.BBN.COM by pineapple.bbn.com id ; Thu, 15 Aug 91 14:01:31 -0400 From: Rich Salz Received: by litchi.bbn.com id ; Thu, 15 Aug 91 14:01:26 EDT Date: Thu, 15 Aug 91 14:01:26 EDT Message-Id: <9108151801.AA25900@litchi.bbn.com> To: bjones@weber.ucsd.edu Subject: Re: The Usenet History List Status: RO Thanks for your kind words. Anyhow, while I read the log in one window, I'll comment in this one. Albert has a hard time believing there is no central Usenet authority. There really, really isn't one. Usenet flows because of point-to-point agreements. There are distributed authorities whom most sys admins trust -- Dave Lawrence for Usenet general, Robert Elz in Australia, etc. Misc.activism.general would have really caused a large number of sites to re-evaluate their "lazy evaluation decision." The lazy eval is, "if Lawrence posts the message, then the vote is valid enough for me." There are groups which got approved, but most sites don't carry them; comp.protocols.tcp-ip.eniac and soc.sex, e.g. When this example was mentioned to Albert he rejected it out of hand, apparently being ignorant of when it came about. Albert says "Chip is apparently authorized to speak on this because nobody has said he's wrong." Thre are two things wrong with this. First, anybody is authorized to speak -- who could gainsay him? Spaf has no authority. Second, most people don't care. Albert seems to really want an elected leadership with formal structure. I am sure most sites are against this. They will not forcibly relinquish control that they current in-effect give, becuase of the lazy eval. Ed Vilmietti's gives an argument in favor of my lazy eval description. Just got someone to duplicate existing work, any many people will believe it because it is easier. >Note that just about anyone with a copy of perl and a monthly cron job >can come up with a reasonable list; write something once, post it >every month, and people will start to believe it. Mythology is a >powerful tool. The myth of the "official list" is but one of the >controlling myths that organize usenet. Albert shows very little knowledge of the history and development of Usenet. "It would be surprising if this was achieved without any central authority." It was. gnu.* is a recent case in point. One person started doing it (Karl, I think, had the idea while in the shower) and two days later it was spread to a half-dozen INET sites; two weeks later UUNET made it available. All individual decisions. Around line 576: >Britain is a democracy with an uncodified constitution and >it works quite well. So does Usenet. >some >say it is possible to receive any newsgroup whatever through email, >whether one's "sysadmin" approves or not... This is not true. >Again one might imagine that the establishment of "feeds" between >Usenet sites is the result of conscious planning by some organization. Conscious planning? Hahaha! You go into a crowded room and yell "who will give me a feed?" At the first NNTP-Manager's BOF at the June Usenix in 88 (89?) I asked if anyone was interested in organizing NNTP newsfeeds. At that time, a very small group of sites. The answer was a resounding no. I believe the fundamental mistake in Albert's views is that he sees an end result "Lawrence posts a newgroup, Spaf lists it, everyone carries it" and draws the conclusion that there is (an) organization. In some sense, this is true, but only because it is a surface analysis. If you look closely at "what is temperature" you get random brownian motion that just happens to present an average motion, giving a temperature. >Since control over naming could easily be abused to enforce private >agendas, I think it is essential that any revised scheme include >some control by the general participants in Usenet at large. The success of the naming used at places like Compuserve prove this wrong. >At the moment I can't see any need for elected officials. But when >the "founders" end up retiring, Usenet could find itself facing >the same problems that lead many other informal associations to >hold elections, just as it hit on the idea of voting for taking >decisions. False. Chuq,Horton,Adams don't post anything. Lawrence is the third person to moderate that group. How could put a great renaming through now? Most software (B2.11 and C News) allow default group creation. They also say "give me comp.ANYTHING even if I don't have comp.foo in my active file." This leads to illusion of organization carrying out votes. If a site didn't pass on a group it didn't explicitly carry, things would be obviously more anarchic. In <10645@idunno.Princeton.EDU> vnend@phoenix.Princeton.EDU (D. W. James) claims soc.sex resulted in alt and destruction of the cabal. This was repeated by Albert. This is false. alt came about because Reid lost the cabal vote on what to call mod.recipes in the great renaming. The cabal split up over comp.society.women. He claims OSI protocols "fix" all the problems with message overflow, disk quotas, etc. This is false. The existing protocols allow for software solutions to these problems, too. And then, at line 8300, I gave up reading. Feel free to post this to the list, or just put in the archvies as "rebutting.langer"