dww@stl.stc.co.uk (David Wright) (11/19/89)
Following the discussion period, the proposal shall be put to the vote, and shall include the names suggested during discussion, plus 'no group'. The news.announce.newgroups moderator may add any that have been mailed to him. Voters will mail their votes by putting the names in preferred order; 1 first.preference 2 second.preference ... (including 'no' if that is one of their preferences) and should (not must) state their first preference in the subject line. Voters' first preferences are counted. If no choice gets over 50% of the votes, the choice having the least votes is eliminated, and those votes are transferred to the voters' next preference. Repeat until either => the total of all other choices less 'no's is under 100 ===> vote fails => one name obtains over 50% of the vote (the quota) ===> that name passes After the count, the results shall be posted to news.groups, showing the count at each stage, and listing voters by their first preferences. The list may also show their other preferences in any convenient tabular form. After 5 days the group will be created, unless irregularities have been found. For group proposals where discussion leads to agreement on a single name, the above procedure will work exactly as the present system - name vs. 'no'. Regards, David Wright STL, London Road, Harlow, Essex CM17 9NA, UK dww@stl.stc.co.uk <or> ...uunet!mcvax!ukc!stl!dww <or> PSI%234237100122::DWW "The principle of proportional representation is very simple; election requires a quota of votes, not a majority, and votes that would otherwise be wasted on a candidate who does not need them are utilised by transfer, on those voters' instructions, to other candidates" (Enid Lakeman, ERS, 1971)
stodol@freja.diku.dk (David Stodolsky) (12/05/89)
In <2438@stl.stc.co.uk> "David Wright" <dww@stl.stc.co.uk> writes: >Following the discussion period, the proposal shall be put to the vote, and >shall include the names suggested during discussion, plus 'no group'. The >news.announce.newgroups moderator may add any that have been mailed to >him. This places a mail load on the moderator and hides part of the process. How about (optional wording indicated in parenthesis): Following the discussion period, all names that have been posted (and that have received a second) will be included in the Interest Group Survey announcement appearing in news.announce.newgroups. A person may suggest (or second) only one name. If any names are omitted, a corrected survey announcement may be sent to the moderator and will be posted if it is a valid correction (and the moderator will delete the faulty announcement.) > the total of all other choices less 'no's is under 100 ===> vote fails This counts "no"s twice. It also discards "no"s if they appear at different stages in the count and are in the minority at that stage. The current rationale for "no"s is to avoid the creation of groups with inappropriate names. Single Transferrable Voting is a better way to do this. "No"s in Single Transferrable Voting allow people to express preferences among the names independent of expressing support for creation of a group. Thus, to answer: "How many people want a group with the most preferred name?": the total vote less 'no's removed is under 100 ===> vote fails (see my post, "Single Transferrable Vote Counting", for details) This avoids the problem pointed out by dave@cogsci.indiana.edu (David Chalmers): (in article <29992@iuvax.cs.indiana.edu>) |Assume I don't care whether the group exists or not, but I feel very |strongly against the name "sci.aquaria". How do I vote? There's no |way I can express this group preference without effectively voting |eith "YES" or "NO" to the group itself. Either I can vote | |(1) rec.aquaria (2) NO (3) sci.aquaria |or |(1) NO (2) rec.aquaria (3) sci.aquaria. | |But neither of these is right. The first will count as an effective YES |vote for the group, demonstrating an enthusiasm for the group which is not |present. Ditto for the second and "NO". With the counting method suggested above, the second option indicates preference without affecting support for group creation. >After the count, the results shall be posted to news.groups, showing the >count at each stage, and listing voters by their first preferences. The >list may also show their other preferences in any convenient tabular form. If users can check the scoring, then posting the intermediate counts is unnecessary. Thus: Results, including number of votes for the most preferred name, number of votes supporting creation, and the how each person voted, shall be posted to news.announce.newgroups and other groups that received the earlier announcements. -- David S. Stodolsky, PhD Routing: <@uunet.uu.net:stodol@diku.dk> Department of Psychology Internet: <stodol@diku.dk> Copenhagen Univ., Njalsg. 88 Voice + 45 31 58 48 86 DK-2300 Copenhagen S, Denmark Fax. + 45 31 54 32 11
peter@ficc.uu.net (Peter da Silva) (12/13/89)
In article <5015@freja.diku.dk> stodol@freja.diku.dk (David Stodolsky) writes: > In <2438@stl.stc.co.uk> "David Wright" <dww@stl.stc.co.uk> writes: [lots of stuff about how to count NO votes] I think the guy who said the vote should include seperate sections for the name and charter was right. Having run STV votes both ways, it's no more trouble to count YES/NO/NO.OPINION plus a choice of names than just a name vote. And it's clearly more flexible. How about this: Votes should be run as now (100 more YES than NO, with 2/3 YES required or without), but if there is controversy about the name then a simultaneous STV vote should be run on the name. This would make the ballots look like: YES alt.drugs.lithuanian.sheepdogs rec.models.barbie comp.sys.eniac.tcp-ip sci.weemba.sheep talk.sex.in.the.morning <frodo@shire.middle.earth> This could also be extended to other aspects of the vote, such as the charter. -- `-_-' Peter da Silva. +1 713 274 5180. <peter@ficc.uu.net>. 'U` Also <peter@ficc.lonestar.org> or <peter@sugar.lonestar.org>. "It was just dumb luck that Unix managed to break through the Stupidity Barrier and become popular in spite of its inherent elegance." -- gavin@krypton.sgi.com
brad@looking.on.ca (Brad Templeton) (12/16/89)
At least from the postings, it seems that many people think that voting on group names is good solution, and that this does not need to be explained. I want it explained. What is the purpose in naming groups, and given that purpose, how does that dictate how a namespace should be structured? And in particular, why would voting on names attain these goals? It seems to me that allocating names in a namespace is the sort of thing which is almost never voted on by the public-at-large in real life. Why do people take it as a given here? Should internet numbers and top level domains be decided by a vote? Filenames for the location of source code on a Unix system? Street numbers? Should the location of a book in the library be voted upon by the borrowers? If not these things, then why group names? -- Brad Templeton, ClariNet Communications Corp. -- Waterloo, Ontario 519/884-7473
peter@ficc.uu.net (Peter da Silva) (12/17/89)
In article <62274@looking.on.ca> brad@looking.on.ca (Brad Templeton) writes: > It seems to me that allocating names in a namespace is the sort of > thing which is almost never voted on by the public-at-large in real life. Sure it is. Maybe not in the U.S., where street names are generated by developers running a buzz-phrase generator, but in Australia street names are frequently voted on by council, based on suggestions from the population. Sometimes even in general elections. > Street numbers? There goes my irony detector. -- `-_-' Peter da Silva. +1 713 274 5180. <peter@ficc.uu.net>. 'U` Also <peter@ficc.lonestar.org> or <peter@sugar.lonestar.org>. "It was just dumb luck that Unix managed to break through the Stupidity Barrier and become popular in spite of its inherent elegance." -- gavin@krypton.sgi.com
stodol@freja.diku.dk (David Stodolsky) (12/18/89)
brad@looking.on.ca (Brad Templeton) writes in Message-ID: <62274@looking.on.ca>: >At least from the postings, it seems that many people think that voting >on group names is good solution, and that this does not need to be >explained. Voting is the worst solution, except for having no way to reach a fair decision. Voting rules should encourage people to come to a consensus decision, in which all important facts and points of view are satisfied. However, reaching a consensus decision on Usenet is nearly impossible for two reasons. With a very large number of participants, some of who act in an irresponsible manner, a total consensus is never reached. Second, if people agree with a point they typically don't post, "I agree". Let's face it, after the first half dozen, this gets to be boring reading. So, given the current communication structure on Usenet, that does little to discourage irresponsible behavior and has no inherent "back channel" or review message capabilities, a vote is normally necessary to reach a decision, even if a consensus exits. This second reason is why I ran the groupware vote STV. A number of people posted or mailed the opinion that the name should be comp.groupware not sci.groupware. Then, I presented my arguments as to why it should be sci.groupware and there was no rejoinder, actually hardly any comment at all. So, either the critics were convinced or they did not read my arguments, couldn't be bothered to respond, their mail/posts were lost, etc. Well, the vote showed that the critics were not convinced (at least not most of them). An ideal voting system should work like an ideal court system. If you know what the decision is going to be, why waste the time to go through a complex and resource consuming procedure. The key point is that the system is fair, so you can not accomplish anything by using it that can not be accomplished by a clear presentation and understanding of the facts. Greg Wood believes that the current patch to the Guidelines (i. e., the 2/3 rule) will by its threat potential, force group champions to accept a proper name. This will not work for two reasons. First, due to the lack of appropriate feedback, which is typically combined with inexperience, group champions often don't know what a proper name is. And they often don't know various persons on the net well enough to distinguish between people trying to help them, and those who just are opposed to the group or get their kicks by taunting any "new kid on the block". Second, a contest, which is even clearer when there is a 2/3 rule, is just what will encourage certain people to try and "beat the system". That is, show that they can WIN, regardless of what the contest is about. This kind of environment also generates vote fraud and forgeries. Finally, people leave Usenet, start their own hierarchies, or take other measures avoid stupid stuff. There is enough recent experience on the net to illustrate this. If single transferrable voting procedures are available, it will be clearer that forcing a decision to a vote is just a waste of time. This, hopefully, will encourage less people to try to delay or defeat an obvious conclusion. A higher level of consensus is certainly needed on Usenet at this time. -- David S. Stodolsky, PhD Routing: <@uunet.uu.net:stodol@diku.dk> Department of Psychology Internet: <stodol@diku.dk> Copenhagen Univ., Njalsg. 88 Voice + 45 31 58 48 86 DK-2300 Copenhagen S, Denmark Fax. + 45 31 54 32 11