From: Jim Riley Newsgroups: news.groups Subject: Re: ANNOUNCEMENT: Status of INET groups Date: Thu, 17 Oct 2002 22:44:37 -0500 Organization: Leningrad Knit Shrimp Lines: 784 Message-ID: References: <1034558191.14057@isc.org> <3dab9c04$0$184$892e7fe2@authen.yellow.readfreenews.net> <3dae3cdc$0$164$892e7fe2@authen.yellow.readfreenews.net> Reply-To: jimrtex@pipeline.com NNTP-Posting-Host: 3f.be.b1.ec Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Server-Date: 18 Oct 2002 03:45:13 GMT X-Newsreader: Forte Agent 1.91/32.564 On 17 Oct 2002 04:30:21 GMT, Joe Bernstein wrote: [Since I've inserted a lot of references to other messages, I've repeated the above introduction when quoting Joe Bernstein] >In article , Jim Riley > wrote: > >> On 15 Oct 2002 04:39:32 GMT, Joe Bernstein wrote: >> >> >In article , Jim Riley >> > wrote: >> > >> >> Since the Inet groups were created without a demonstration of >> >> interest, >> > >> >Point of order. >[I then went on to say that when most of the inet groups were created, >in late May or early June 1987, demonstration of interest was not a >universally recognised criterion for newsgroup creation. I went on >to explain what I meant by this, but that's not relevant to dealing >with what Jim Riley typed in *before* quoting my saying that:] > >> Examples of counter-evidence with respect to demonstration of >> interest. >> >> Message-ID: >> Newsgroups: net.news.group >> From: cbosgd!mark >> Date: Tue Jun 1 00:52:24 1982 >> Subject: official policy on creation of new newsgroups On 17 Oct 2002 04:30:21 GMT, Joe Bernstein wrote: >I'm curious where you see any evidence in this that a demonstration >of interest is called for. The relevant sentence, I think, is "Once >a consensus is reached, the person who originally proposed it should >make a yes/no decision on the newsgroup and a name, announce it to >net.general, and create it." I don't buy this as requiring a >demonstration of interest; it calls for a consensus. I think that the 'consensus' included whether the group was needed. But I'll see if I can find something else. Throughout 1983, there were quite a few new groups being created. In some cases, the "discussion" was mostly of the form, "I vote YES on net.rec.disc", for example. (This group was created, then removed so it is not the direct predecessor of the current rec.sport.disc). There were complaints about groups being created, and then not having any traffic. There was a proposal to remove groups that received less than 2 messages/week. From: steveb@shark.UUCP (Steve Biedermann) Newsgroups: net.news.group Subject: Newsgroup wars Message-ID: <208@shark.UUCP> Date: Fri, 20-Jan-84 01:53:07 EST This argues that "votes" should not be used, but that actual traffic should be present before creating a new group. This occured at the time when proposals were made to split off net.religion.jewish (renamed to net.culture.jewish during the GR), and to split net.music. There were quite a few proposals at this time for restructuring of newsgroups into more of a hierarchy. This anticipated that the top level hierarchy (e.g. 'net.' would be stripped off and become the distribution). From: alb@alice.UUCP (Adam L. Buchsbaum) Newsgroups: net.news.group Subject: On Proposing New Groups Message-ID: <2612@alice.UUCP> Date: Tue, 21-Feb-84 10:34:44 EST This is worth reading the whole thread. Adams Buchsbaum argues for not creating a group unless discussion already exists. Werner Uhrig argues that in some cases this is impossible because the existing groups may not be suitable. Chuq Von Rospach agrees to a certain extent, but argues that if there was really interest in discussing a topic, the participants could take over a quiet group. From: alb@alice.UUCP (Adam L. Buchsbaum) Newsgroups: net.news.group Subject: Abuse of net.news.group Message-ID: <2634@alice.UUCP> Date: Sat, 3-Mar-84 14:38:49 EST This says that votes should not occur in net.news.group but should take place by email. Voting is not defined. From: alb@alice.UUCP (Adam L. Buchsbaum) Newsgroups: net.news.group Subject: Forget About Newsgroup Restructuring Message-ID: <2643@alice.UUCP> Date: Thu, 8-Mar-84 06:51:32 EST And here he argues that groups should be when there is traffic, and for neat ideas, and dead groups should be removed. From: woods@hao.UUCP (Greg Woods) Newsgroups: net.news.group Subject: Re: Newsgroup Births and Deaths: Here we go again (and again ....) Message-ID: <969@hao.UUCP> Date: Mon, 14-May-84 16:07:46 EDT | 5) (more flames) ORGANIZE THE CREATION OF NEWSGROUPS. While this may sound | like USENET, Inc., that isn't what I am suggesting. I'm not sure exactly | what it is I *am* suggesting here, I'd just like to get the creative | juices flowing. I would like to see some small group of people, | maybe just one at a time, be responsible for creating and deleting | newsgroups. Maybe we could elect a Newsgroup monitor with voting by | Email? From: riddle@ut-sally.UUCP Newsgroups: net.news,net.news.group Subject: Posted to net.sources: newsgroup description hack to vnews Message-ID: <247@ut-sally.UUCP> Date: Tue, 22-May-84 00:44:47 EDT Not really anything to do with group creation, but this is about a news hack to get the short description (from Adam Buschbaum's proto-checkgroups) and to display it. From: chuqui@nsc.UUCP (Chuq Von Rospach) Newsgroups: net.news.group,net.news Subject: If I see one more yes vote, I'll whimper! Message-ID: <965@nsc.UUCP> Date: Fri, 25-May-84 10:32:29 EDT This is a proposal for a steering committee - and says "This absurd business of 'voting' on new topics has GOT to stop." Probably worth scanning the whole thread. From: mark@cbosgd.UUCP (Mark Horton) Newsgroups: net.news.group,net.news Subject: Re: If I see one more yes vote, I'll whimper! Message-ID: <1392@cbosgd.UUCP> Date: Tue, 29-May-84 21:41:44 EDT This is in the follow-up thread to the previous message. Horton says that the idea behind voting was that they should go to the person proposing the group, and that it was the only way to determine whether there was interest in the proposed group. From: liz@umcp-cs.UUCP Newsgroups: net.news.group Subject: major restructuring and idea for group coordinators Message-ID: <7460@umcp-cs.UUCP> Date: Mon, 11-Jun-84 16:38:29 EDT Vaguely related, proposes that groups have "coordinators" to watch over them. The coordinator would have had a role that combines the role of Usenet II czars and us.* hosts. net.news.group would be used only for discussion of truly new groups. Message-ID: Newsgroups: net.music,net.news.group From: hao!woods Date: Fri Feb 11 06:10:03 1983 [***] Subject: net.music.gdead Is this the first results posting? In light of groups like net.wobegon, surely the following list demonstrates that there are sufficient DeadHeads on the network to form our own group. [***] This date is correct. Through the serendipity of Google's cross-threading, Greg Wood's posting above is mixed in with a discussion of a new Dead group that resulted in the following From: rcd@opus.UUCP (Dick Dunn) Newsgroups: net.music,net.news.group Subject: Results of poll on Grateful Dead subgroup Message-ID: <1137@opus.UUCP> Date: Wed, 3-Apr-85 03:05:41 EST Announces a 47:10 in favor of net.music.gdead From: sunil@ut-ngp.UUCP (Sunil Trivedi) Newsgroups: net.followup,net.news.group Subject: Re: Abuse of net.general [MCI ad] Message-ID: <1571@ut-ngp.UUCP> Date: Sat, 6-Apr-85 20:54:21 EST Article-I.D.: ut-ngp.1571 A Spam sighting? From: fair@dual.UUCP (Erik E. Fair) Newsgroups: net.news.group Subject: Getting rid of fa.* (and moving it to mod.*) Message-ID: <1018@dual.UUCP> Date: Fri, 17-May-85 02:28:14 EDT A proposal to rename fa.* to mod.* From: usenet@gatech.CSNET Newsgroups: mod.newslists,net.news.group,net.announce.newusers Subject: List of Active Newsgroups (Last changed: 1 July 1985) Message-ID: <400@gatech.CSNET> Date: Mon, 1-Jul-85 00:39:09 EDT |mod.rec Discussions on pasttimes (not currently active) | mod.rec.guns Discussions about firearms Interesting in that the creation of mod.rec.guns also resulted in the creation of mod.rec, at least as a place-holder. A scan through the other hierarchies shows that there was always a parent group when there were sub-groups. A recurring debate was what conditions warranted creation of a sub-group. From: spaf@gatech.CSNET (Gene Spafford) Newsgroups: net.news.group,net.movies.sw,net.women.only,net.micro.zx,net.rec.coins,net.rec.disc,net.mail.msggroup Subject: Removing newsgroups Message-ID: <579@gatech.CSNET> Date: Mon, 15-Jul-85 12:08:59 EDT |Time to clean out some deadwood, folks. I intend to apply the same |criteria we use for creating new groups to these old, unused ones: |there must be a clearly demonstrated need not filled by another group, |AND a sufficient volume of postings to warrant a separate group. A proposal to remove some low-traffic groups - see Newsgroups list. From: gmp@rayssd.UUCP (Gregory M. Paris) Newsgroups: net.news.group Subject: how many does it take? Message-ID: <859@rayssd.UUCP> Date: Wed, 17-Jul-85 17:35:31 EDT This asks what is the standard for removal of a group (net.flame in this instance): If it takes a respectable amount of positive response to initially create a group, how much response does it take to remove one? From: spaf@gatech.CSNET (Gene Spafford) Newsgroups: net.news.group,net.bizarre Subject: Re: The return of net.bizarre. Message-ID: <629@gatech.CSNET> Date: Wed, 24-Jul-85 11:03:18 EDT This was in response to a poll favoring re-creation of net.bizarre. | For the information of our readers, the procedure is: | | 1) Propose a new group in "net.news.group" and related, existing groups | 2) Poll the readership for comments | 3) Establish that there is/would be sufficient traffic | 4) Based on responses, create the group. A follow-up suggests that the preferred order should be: 3 1 2 4. From: lauren@vortex.UUCP (Lauren Weinstein) Newsgroups: net.news,net.news.group Subject: Newsgroup creations and deletions Message-ID: <725@vortex.UUCP> Date: Sat, 27-Jul-85 16:50:18 EDT An interesting thread in response to the removal of net.general. From: ado@elsie.UUCP (Arthur David Olson) Newsgroups: net.news.b,net.news.group Subject: How to deal with net.general Message-ID: <5204@elsie.UUCP> Date: Sun, 11-Aug-85 13:28:31 EDT An amusing suggestion that off-topic postings in net.general be cancelled (retromoderated). Since the articles in net.general were to be of interest to *all*, then if a particular article was not of interest to you, you could feel free to cancel it. From: reid@Glacier.ARPA (Brian Reid) Newsgroups: net.news.group Subject: the trouble with all these rules is... Message-ID: <10609@Glacier.ARPA> Date: Sun, 11-Aug-85 20:20:04 EDT Reid argues that hardly anyone reads net.news.group, and that it is unrepresentative of the net as a whole. And that there ought to be a way for Everyman to create newsgroups that interest them. Chuq von Rospach responded: | Unfortunately, the network seems to feel that deleting groups is fascist, | and any attempt to do so dumps lots of rotten vegetables on the heads of | those who try. Because of this, it is almost as hard to create a group as | it is to delete one, and I think everyone loses because of it. From: spaf@gatech.CSNET (Gene Spafford) Newsgroups: net.news.group Subject: Re: The Great Net Cleanup Part II Message-ID: <813@gatech.CSNET> Date: Mon, 12-Aug-85 23:12:31 EDT This was a summary of mostly negative feedback that Spaf had received on the proposed removal of a baker's dozen groups. Brian Reid had posited that there were 3 types of groups: 1) information sharing: many readers, few posters. 2) social: readers and posters were mostly the same people. 3) soap box: more writers than readers. Spaf mused that maybe there should be three sets of rules for group creation. From: spaf@gatech.CSNET (Gene Spafford) Newsgroups: net.news.group,net.database Subject: net.database created Message-ID: <840@gatech.CSNET> Date: Wed, 14-Aug-85 14:39:55 EDT net.database is created by "popular demand", which also included discussion of alternative names (e.g. .data, .db, .dbms) From: greenber@timeinc.UUCP (Ross M. Greenberg) Newsgroups: net.news.group,net.micro.pc Subject: Re: request for mod.sources.pc Message-ID: <410@timeinc.UUCP> Date: Wed, 14-Aug-85 19:43:45 EDT |Count me in for a 'yes' vote. Looks like the magic number is about |thirty 'yeses', so start those cards and letters, folks! From: lauren@vortex.UUCP (Lauren Weinstein) Newsgroups: net.news.group Subject: magic numbers Message-ID: <745@vortex.UUCP> Date: Fri, 16-Aug-85 14:10:17 EDT Weinstein's response was triggered by the previous message. From: lauren@vortex.UUCP (Lauren Weinstein) Newsgroups: net.news.group Subject: Vote Fraud and Newsgroups Message-ID: <755@vortex.UUCP> Date: Sat, 24-Aug-85 14:25:19 EDT This proposes that votes be conducted by a independent 3rd party. Basically, it indicates the evolution of a process from where the proposer of a group could be expected to fairly determine whether a consensus exists, evaluate opposing arguments, and then decide whether he (the proposer) should create the group - to one where the evaluation and execution of a proposal should be done by the net community. It notes that what support level is needed is another issue entirely. From: msb@lsuc.UUCP (Mark Brader) Newsgroups: net.news.group Subject: Re: Vote Fraud and Newsgroups Message-ID: <777@lsuc.UUCP> Date: Sat, 31-Aug-85 17:02:44 EDT This is a followup to the previous. It proposes a mod.news.group. Simple votes would be filtered by the moderator. Substantive arguments would be posted - with sub-votes possibly conducted. From: ronse@prlb2.UUCP (Ronse) Newsgroups: net.news.group Subject: how to avoid vote fraud? Message-ID: <804@prlb2.UUCP> Date: Thu, 29-Aug-85 14:55:19 EDT This article suggests that the posting of votes would be one way to counter vote fraud. From: kaufman@yale.ARPA (El Magnifico Qux) Newsgroups: net.news.group Subject: A proposal for combat of newsgroup proliferation Message-ID: <116@yale.ARPA> Date: Wed, 4-Sep-85 22:09:06 EDT This suggests a group for controlling which groups should be created. From: lauren@vortex.UUCP (Lauren Weinstein) Newsgroups: net.news.group Subject: Doomsday cometh (VERY LONG) Message-ID: <781@vortex.UUCP> Date: Thu, 5-Sep-85 16:06:53 EDT Expresses concerns about proliferation of new newsgroups and volume will destroy Usenet. From: chuqui@nsc.UUCP (Chuq Von Rospach) Newsgroups: net.news.group,net.news Subject: mailing lists vs. newsgroups: facts (was Vote Fraud and Newsgroups) Message-ID: <3221@nsc.UUCP> Date: Sat, 7-Sep-85 02:01:33 EDT This proposes some criteria by which it could be more efficient to convert a mailing list to a newsgroup. Which brings us to this message: >> From: spaf@gatech.CSNET (Gene Spafford) >> Newsgroups: net.news.group,net.music >> Subject: Yet Another Rogue Group -- net.music.guitar >> Message-ID: <1207@gatech.CSNET> >> Date: Thu, 12-Sep-85 08:58:37 EDT On 17 Oct 2002 04:30:21 GMT, Joe Bernstein wrote: >Oh, hey: that date brings us to this group. > >And here we *do* find a clearly written procedure. It even has >voting. Fascinating. I had tended to assume that since very few >groups were created in net.* from 1985 to the Great Renaming, there >couldn't have been votes; and I noted a little discussion archived >in news.groups's early days that seemed to indicate the voting >procedure being worked out. Either the disdain for unmoderated >groups was *much* wider spread in Usenet's Golden Age than I would >have thought, though, or the backbone admins really were stomping on >voted groups left, right and center. I wonder which. Well, that >stage of my research is a long way off. Reading back over the 1983-85 period in net.news.group, it is clear that the the procedure outlined by Spaf was widely understood. It was only when the procedure wasn't followed (e.g. net.bizaare or in this case, net.music.guitar), that the procedure was more formally enunciated. As the introduction to Spaf's list notes: | | For future reference, the proper procedure for creating a new group is: | A response to Spaf's message suggested that a poll be conducted. Others argued that this would mean a breakdown in the approved procedure. In this time frame, there was also discussion of alternate voting schemes, including those that would be site-based and weighted on the basis of the number of newsfeeds (in effect, biased toward backbone sites) On 17 Oct 2002 04:30:21 GMT, Joe Bernstein wrote: >In any event, I do concede that this procedure does call for a >demonstration of interest, both by doing what we would do today >through writing a Rationale, *and* by holding a vote. Continuing on from September 1985. From: cher@ihlpm.UUCP (cherepov) Newsgroups: net.sport,net.news.group Subject: Tennis newsgroup (attn. Gene Spafford) Message-ID: <493@ihlpm.UUCP> Date: Fri, 20-Sep-85 20:31:26 EDT This argues that a +40/50 vote margin is too rigorous. From: qux@yale.ARPA (El Magnifico Kaufman) Newsgroups: net.news.group Subject: Vote count for/against net.bizarre Message-ID: <232@yale.ARPA> Date: Mon, 23-Sep-85 14:31:22 EDT This is a vote result on the proposed removal of net.bizarre. From: reid@Glacier.ARPA (Brian Reid) Newsgroups: net.news.group Subject: current newsgroup creation rules are silly Message-ID: <13220@Glacier.ARPA> Date: Tue, 15-Oct-85 22:50:53 EDT Reid argues readership is more important than writers. This will eventually lead to the readership statistics. From: mcb@k.cs.cmu.edu (Michael Browne) Newsgroups: net.news.group,net.flame,net.misc Subject: New (?) Newsgroup Request - net.bizarre Message-ID: <615@k.cs.cmu.edu> Date: Mon, 21-Oct-85 23:17:14 EDT This proposes a "new" group called net.bizarre. It is suggested that discussion of bizarre things take place in net.misc to demonstrate existing traffic, and that people send their votes in to indicate interest. From: thau@h-sc1.UUCP (robert thau) Newsgroups: net.news,net.news.group,net.flame Subject: Fear and Loathing on the Clouds Message-ID: <614@h-sc1.UUCP> Date: Tue, 22-Oct-85 20:31:56 EDT Worth reading the whole thread. Thau complains about Spaf's removal of net.bizarre and net.internat. The debate is about the backbone enforcing the rules vs. what the people want. And whether the rules had been consistently applied. From: rick@seismo.CSS.GOV (Rick Adams) Newsgroups: net.news.group Subject: net.flame will no longer be distributed by many sites Message-ID: <738@seismo.CSS.GOV> Date: Fri, 8-Nov-85 20:49:06 EST Announces that net.flame will no longer be distributed by a number of sites. Henry Spencer also announced that he would no longer be propagating a number of troups. From: ln63fkn@sdcc7.UUCP (Paul van de Graaf) Newsgroups: net.news.group Subject: Ban the binaries! Message-ID: <162@sdcc7.UUCP> Date: Fri, 8-Nov-85 11:04:58 EST This is a complaint about the binaries being posted to net.sources.mac. From: henry@utzoo.UUCP (Henry Spencer) Newsgroups: net.arch,net.unix-wizards,net.lan,net.news.group Subject: Re: Re: net.os Message-ID: <6234@utzoo.UUCP> Date: Wed, 18-Dec-85 13:19:05 EST Complains about public voting, and also says that voting does not demonstrate existing traffic. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ This gets us through the end of 1985. >> >Furthermore, and this is important, the rationale for the creation of >> >the inet groups *WAS* demonstration of interest. What Erik Fair did, >> >at that time and to a lesser extent thereafter, was create a newsgroup >> >for each widely-accepted ARPAnet/Internet mailing list, using the logic >> >that it made more sense to have one copy on every server than to mail >> >one copy to every subscriber. >> >> It is not clear that Erik Fair limited himself to widely-accepted >> mailing lists. See for example, and the thread that follows. > >Note please that the thread in question is nearly two years after the >initial creation. > >At some point, it would obviously be worthwhile to check the inet >groups of 1987 against the ARPAnet List of Lists. I guess what I should >have written is "What Erik Fair said he did", and left it at that. >But frankly, my impression is that Fair simply used inet as a tool >to do whatever he wanted to do *after* 1987; he quite obviously did >not continue his initial manifesto of one newsgroup per mailing >list (thank God), but he never stated (that I know of) any other >overall logic to the inet distribution. > >Thanks for the pointer. > >> >Notice, please, that this is *precisely* the same logic that the Big 8 >> >newsgroup creation system put in place later has used to this day. I >> >think it's basically impossible to believe that the Big 8 system got >> >this logic from anyplace else than the rationale Fair gave for inet, >> >although (since that rationale had been stated less clearly in prior >> >years) he may not deserve sole credit. >> >> I don't see the 1988 Guidelines as being original at all, but merely >> an evolution of standards that existed nearly from the beginning of >> Usenet. > >Spafford was posting guidelines starting in late 1987. Which 1988 >guidelines do you mean? Spaf's November 2, 1987 guidelnes. >I think his text in late 1987 is clearly descended from the post you >pointed me to above, from 1985. I'll have to look in a great deal >more detail for such posts. I'm afraid all I did thus far is chain >back through news.announce.newusers, taking advantage of Spafford's >tendency to think that most of the things he auto-posted were of >interest to newbies. I don't want to go find it now, but there was a post in the latter 1985 period suggesting that the procedure for creating a new group be posted to news.announce.newusers. I think the intent of the Spaf's 1985 post is to inform all the people who may not be aware of the current procedures. Originally, one person took the role of proponent, votetaker, and newgroup issuer. Only the role of list-keeper was distinct. In essence, the proposer was trying to determine whether they should issue the newgroup message. Most participants were either the news admin or close to the news admin, and might be expected to have a concern about the overall well-being of the net. There tended to be more of a common understanding of what might be needed for a new group to be successful (not that the common understanding was neccessarily correct, but that it at least existed). Later participants were more interested in the topics being discussed, less interested in Unix. The process had to be explained to them. >I have posted in the past that a thread archived from December 1981 >contains in capsule all of the systems the "official" hierarchies >have used to date: Wm Leler with the early days anarchy, Mark >Horton with the Backbone Cabal, and Steve Bellovin with Big 8-style >voting, more or less. I do think the *standards* have been evolving >steadily. But look: the raw total of unmoderated groups in the >official lists was 173 at the end of 1984, 176 at the end of 1985, >and 184 at the end of 1986. I compared the following lists: 9/31/84 (just a couple of months after Spaf took over from Buschbaum), 3/31/85, 12/31/85, 9/15/86 (start of GR) 6 month period (9/31/84) to (3/31/85) net: 2 removed, 11 created, net.news.map moved to mod.* mod: 1 removed, 11 created. This is the foundation of mod.* 9 month period (3/31/85) to (12/31/85) net: 11 removed, 10 created, 2 renames. mod: 31 created (including 10 conversions from fa.*), 1 removed. Originally more groups had been proposed for removal. Some of the groups that were removed were later re-created (e.g. net.rec.coins, net.rec.disc, net.movies.sw, net.theater). So part of what you see as stagnation was actually removal of some dead wood. Very few computer groups were created (net.micro.{amiga,att,mac}, net.internat, and net.database). The creation of net.internat had not followed the usual procedures, and originally was presented as being the equivalent to net.bizarre in that regard. But since it had more serious backing, it was eventually regularized. A large share of the mod.* groups were computer related. So there was 10% growth, plus another 20% in mod groups. Maybe the energy of those who knew how to create groups was redirected towards mod.*. And you were also getting into a period where there was more expressed concern about the volume of traffic, especially that which wasn't Unix related. 9 month period (12/31/85) to (9/15/86) This ends just at the start of the Great Renaming. The first 8 talk.* groups had been created, but their net.* equivalents still existed. net: 2 announce groups moved to mod.*; 1 2-for-1 split; 1 rename (sort-of); one removal; 5 mew groups. mod: 17 new groups. It is hard to count since it is not always clear whether the parent group in a sub-hierarchy was active. Another thing that hampered growth at this stage was the difficulty in splitting groups. Many people simply didn't want to split groups. They felt that everything would be cross-posted or they would end up reading all the sub-groups in any case. When you had sub-hierarchies it was for collections of similar groups (net.micro.*, net.rec.*) rather than sub-groups of the main topic. > This compares to strong growth in each >year 1981 (as best I can tell) through 1984, weak growth in 1987, >and strong growth in 1988, 1989, and as far as I know 1990-1998. >I think it's reasonable to assume that the change in 1985 and 1986, >as well as part of 1987, was the Backbone Cabal's control, and I >think it's reasonable to ask why, after two and a half years of >basically requiring an rmgroup as the price of a new unmoderated >group, the Backbone Cabal started relaxing its grip. The three things >I see as having contributed to this are, in chronological order: > 1) the taking-stock of mod.* that came about when mod.* was due for > renaming (a number of mod.* groups were unmoderated while they > were renamed, and unmoderations continued for two years thereafter); > 2) the creation of alt.* by 'rogue' backbone admins; > 3) the creation of inet by a non-'rogue' backbone admin. >Basically, I think the backbone admins found themselves backed into >a corner. Moderation could no longer be defended as the perfect cure >for all that ailed Usenet, since moderators were disappearing all over >the place; and more and more of their own membership were demonstrating >disagreement with the no-new-unmoderated-groups policy. > >In this situation, though, the admins could, in my opinion, pretty much >write their own ticket. They *didn't have* to say "We'll create it >if you hold a vote". If that was already quasi-traditional, then I >suppose their saying it made them look better, but if they'd chosen >something else, they might have gotten away with it, for as long as >they got away with what they did choose. > >The short version of what I'm trying to say here is that 1985 and >1986 were an interruption, and I don't think it's ever historically >sound to assume that a process interrupted will just continue where >it left off. So I think we still have to account for the fact that >in 1987 we got voting guidelines, and I have multiple reasons (one >obviously being the mailing-list exception) for thinking that this >development owed a *lot* to the embrace of demonstration of interest, >as a rationale, by one of the most high-handed and arrogant of the >backbone admins. In 1987 we got interest polling by a person who had been one who had stressed the importance of interest polling before. Others had tended to give more importance to demonstration of traffic. If Brian Reid or Brad Templeton were the listkeeper, you would have been more likely to switch mechanisms. >> >I haven't confirmed this by detailed study of the group creations [1], >> >so don't know how widely practised this was, but the original group >> >creation guidelines written by Gene Spafford explicitly said that >> >a very active mailing list was adequate evidence of traffic, and in >> >that case a vote could be skipped. A *lot* of inet groups were >> >promoted in 1987, and I'm fairly sure this was how most of them were. >> >> Identifying a need for a newsgroup has tended to become less obviously >> a part of the process, because it is not a formal part of the process. > >The Rationale is its fossil. Formality has taken a toll on things >like that that aren't really formal. I suppose if we switched to >these Justification things that alt.config is now asking for, we >could formalise it. Only superficially. 1.Determine if there is traffic. 2.Suggest the group (RFD). 3.Reach consensus (discussion in news.groups, there is a place for it). 4.Take an interest poll (CFV). 5.Issue the newgroup (issue the newgroup). I can only prove that 2, 4, and 5 actually happen. Some semblance of 3 can occur. >> >Among the 1988-1989 groups >> >hardly any are promotions from inet; in 1987 I have several confirmed >> >examples - maybe a dozen? - and strongly suspect more. (It's hard to >> >be sure without checking group by group because Fair didn't bother to >> >post a list of inet groups until the end of 1987.) >> >> There is a list from 1989 by Spaf of inet groups when he proposed >> merging them into the regular Big 8. > >1989 is easy; he was posting lists of inet in the "Alternative >Newsgroup Hierarchies" posts starting in I think September 1988. August 1988. Before that time his entry for the Imet groups gave Erik Fair's email address. It is quite possible that Spaf simply edited Fair's list from the previous December. There are no differences. >1987 is hard because *nobody* posted a full list of inet groups that >is archived before Erik Fair's year-end list, which you note: > >> OK, I found a list of gated newsgroups from December 1987, including >> those in the Inet distribution. > >Every time people complained about inet in 1987, they asked for an >Offishul List Of inet Groups so they could include/exclude them >the way they wanted to. Erik Fair, for whatever bizarre reason of >his own, went *on and on* arguing with his attackers, even specifying >*how many* groups he'd created, but never posting such a list. It's >really crazy-making to me, especially since the archives from that >period are high-backbone-compliant Toronto, and aren't going to tell >me directly which groups he created when. > >Finally at year's end he seems to have gotten over his snit, but >even then, it would have been too simple to post a list of inet >groups, so he posted a complete list of gateways at Berkeley. This >had as one amusing effect a real showing-up of Gene Spafford, whose >list of gatewayed Big 7 groups was thoroughly tattered by that point >(it included only one *fifth* of the groups Fair listed! and claimed >that most of *those* gateways were unreliable to boot!). Spafford >did not use it to update his list. > >But Fair did, oh by the way, note within that list which groups were >Big 7 and which were inet. He got three of them wrong, as I gather >you noticed in your breakdown. Anyway, Spafford, not in his 1/1/1988 >but in his 1/15/1988 checkgroups, then started doing separate >checkgroups with or without inet; and in September 1988 (I think) >started listing inet groups in the "Alternative Newsgroup Hierarchies" >list. Which 3 groups? comp.arch.parallel-sym and comp.sys.pc.net were in Fair's 1987 list, and Spaf's August 1988 inet list. They were removed from Spaf's December 1988 inet list (and hence from the Janaury 1989 list). sci.psychology was created as a Big 8 group in 1988. It had existed as an Inet group. This wasn't really characterized as a conversion, but rather as creation of a group covering the same topic on a different distribution. The proponent had originally proposed a different name. comp.lang.scheme.c was a new Inet group around the time of Spaf's initial listing of Inet groups. It didn't exist in 1987. -- Jim Riley