Xref: utzoo news.groups:15514 news.misc:3951 Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uunet!tut.cis.ohio-state.edu!pt.cs.cmu.edu!rochester!udel!burdvax!gvlv2!lock60!mhw From: mhw@lock60.UUCP (Mark H. Weber) Newsgroups: news.groups,news.misc Subject: Results of Newsgroup Guidelines Survey (conclusions) Message-ID: <373@lock60.UUCP> Date: 10 Dec 89 09:04:07 GMT Reply-To: mhw@schuylkill.canal.org (Mark H. Weber) Followup-To: news.groups Organization: Lock #60 on the Schuylkill Canal, Phoenixville, PA Lines: 117 [Last in a series. For reference, the entire series (with slight corrections) has been posted in news.misc. Thanks for bearing with me - Mark] First, a word about possible biases in the survey. Certainly, the way the survey was presented made it seem that some change was necessary, so a majority of the responses were from people who feel that some change is required. Also because Approval Voting (MAUVE) and Transferable Voting (STV) were involved, the majority of responses were from people who felt that these mechanisms were worth trying. Whether the survey results reflect the feelings of the USENET community at large is difficult to say. At a minimum, the results reflect the feelings of group which cares enough about the community to try some new approaches to making USENET run more smoothly. Secondly, a passing mention of my personal biases. If you really need to categorize me, you could throw me in with the dreaded "namespace purists". I believe that a concise, well structured namespace hierarchy is a key to the usefulness of USENET. It allows users to find the appropriate place to post their ideas, and allows administrators to control the overall categories of information which flow through their computers. If users can't find where to exchange information, and administrators feel they have no control, then no one will bother with USENET (yes, the Imminent Death of the Net as We Know it). So, taking this into account, on to my conclusions as to what needs to be changed. I see three ways to proceed 1) a simple, quick fix, 2) a slightly more complex, slightly more satisfactory fix, and 3) a fairly radical change which could reduce the amount of useless discussion in news.groups by a factor of about 10. All three of these possible solutions are supported by the results of the survey, the comments I received during the survey, and the general discussions in news.groups over the past 6 months or so. Note that I am not proposing specific guideline language here, I am just presenting the overall idea of the changes I am proposing. These proposals will require some further refinement before adoption. 1) Quick fix - The quick fix is to add the 2:1 Yes:No ratio check to the existing guidelines, in addition to the existing 100 more Yes's than No's rule. This is the single most popular proposal among all people who responded to the survey, whether by MAUVE or STV. At this point, it appears that Greg has already gone ahead and made this change to the gudelines, and has not been flamed to a crisp yet, so no real further discussion is necessary on this decision. This quick fix addresses the biggest problem we are currently facing: how to prevent fragmentaion of the net when new groups with significant opposition are approved. It does not make it any easier to resolve controversy over popular, but mis-named groups. 2) Slightly better fix - Separate voting for the charter of the group and the name of the group. This has been suggested by many posters, and was mentioned in many of the comments from survey respondents. In it's simplest form, it would allow voters to vote yes/no for the group and select name choices name from a pre-selected list of names (write-in doesn't really work well) all on the same ballot. Either MAUVE or STV could be used for the name choice. Several of the recent newgroup proposals are attempting to use this mechanism; it will be interesting to see how the net reacts to this, and how well this will work in practice. This should resolve the problem of selecting the appropriate name. It still does not address other pending issues - removal of outdated groups, and renaming of groups as the net evolves. No guidelines exist to address these issues. 3) Major overhaul - Newsgroup committee with net-at-large to veto/override comittee decisions. A large percentage of newgroup proposals are non- controversial. Wouldn't it be nice if these groups could be created without having any extensive discussion or voting by the net-at-large? Under this proposal, the newgroup creation process would go something like this: A) Newgroup proponent petitions newsgroup committee to create new group. Included in petition is 1) Proposed charter of group, 2) Proposed name of group, and 3) Supporting documentation showing need for new group. This would take the form of a) existing volume in a group which needs to be split, b) heavily used and/or growing mailing list, or c) alt newsgroup (or other trial newsgroup mechanism). Committee considers request, and either agrees or disagrees with proponent. If the disagreeement is over the name, the committee may propose an alternate name to the proponent, which he may or may not accept. All of this would take place in private email, not on the net. The elapsed time here would probably be about 1 week. B) Committee posts announcement of newgroup proposal including committee decision to news.announce.newgroups. General discussion in news.groups follows. If proponent (or someone else) disagrees with the decision of the committee (whether for or against creation of the group), he or she can call for a vote by the net-at-large, which takes us back to the situation we have currently. If no one makes a formal challenge, the decision of the committee will be considered final, and the newgroup created, assuming that was the decision of the committee. Due to net propogation, this phase will probably need to last about 2 weeks. So, for a non-controversial group, we could have a new group created in about 3 weeks, without a vote of the net-at-large. If someone wanted to mount a formal challenge, it would take as long as the current procedure does. Ideally, only 1 group in 10 would be truly controversial (after being pre-processed by the committee), so the traffic in news.groups could be reduced by a factor of 10. The committee could also be empowered to consider petitions to remove and rename groups. Once again, the decisions of the committee could be challenged and overruled by the net-at-large. Obviously, the makeup of and selection process for the committee would be critical in making this proposal work. I envision a committee of about 7, with staggered expiration of terms, with a term lasting for a year or so. This proposal will need more development before it is ready for implementation. A few last words. I only discovered this net a year ago. When I did, I was amazed that anything so user-oriented and free could survive. Now that I have take the time to try to understand the history and operation of the net, I am still amazed. Despite the amount to bitter feelings netters express toward each other, they still find the time and energy to try to make the net a more and more useful system. I hope I have made a contribution toward this effort. I have not made these proposals out of a personal desire to assert some control over the net. I feel that the net must adopt some additional structure in order to enable it to continue to evolve and grow. -- Mark H. Weber ( mhw@Schuylkill.Canal.Org ) "Schuylkill" (skool' kill) Mont Clare ( ...!uunet!lgnp1!lock60!mhw ) is a Dutch word meaning PA USA ( ...!psuvax1!burdvax!gvlv2!lock60!mhw ) "hidden river"