alien@cpoint.UUCP (Alien Wells) (11/10/89)
In article <1989Nov8.014837.21854@alembic.acs.com> csu@alembic.acs.com (Dave Mack) writes: >There are two major problems here. The first is that it should be possible >for someone to vote for the group without voting for any of the names. The >current consensus seems to be that most people don't really care about >what name the group winds up with as long as it gets created. This is a good point. Perhaps there should be an entry for simply YES or NO to the entire thing (although that could easily be accomplished by putting YES or NO in each proposal - I doubt there will ever be so many that this would be a major chore). >This permits one of the obvious STV pathologies unless the changes from >the previous paragraph are incorporated: it is possible to have a vote >in which several hundred people vote YES, no one casts a NO vote, >yet no newsgroup is created because no name received more than 100 >YES votes. I think that this is actually a place where this proposal is significantly better than STV. If people are actually voting properly (ie: everything that they would find acceptable is voted YES), the only way that you could have all proposals fail is if each vote taken independantly would fail. If that is the case, then the group should not be created. Otherwise, MAUVE is simply choosing between the options that would have passed, picking the one that would have had the 'best' vote. -- --------| Fall not in love, therefore. It will stick to your face. Alien | - Deteriorata --------| decvax!frog!cpoint!alien bu-cs!mirror!frog!cpoint!alien
alien@cpoint.UUCP (Alien Wells) (11/10/89)
In article <4470@solo10.cs.vu.nl> maart@cs.vu.nl (Maarten Litmaath) writes: >I like the proposal, but there's another point to consider: what if there's >a close finish of 2 (or more) groups? >Example: > YES NO > sci.aquaria 666 500 > news.aquaria 665 500 > >Shouldn't both groups be created? You should only create two different groups if they are to have different charters. If they are in the same vote, they are probably a disagreement about the namespace to put a single group in. If it is really two charters for different groups, it should be handled in different votes. In this case, if sci.aquaria was created and enough people STILL wanted an additional news.aquaria, they could have a second vote over the new charter. -- --------| Fall not in love, therefore. It will stick to your face. Alien | - Deteriorata --------| decvax!frog!cpoint!alien bu-cs!mirror!frog!cpoint!alien
peter@ficc.uu.net (Peter da Silva) (11/11/89)
In article <1989Nov11.004856.6078@NCoast.ORG> allbery@ncoast.ORG (Brandon S. Allbery) writes: > As I noted the last time this voting scheme was suggested, it lends itself to > intentional or unintentional divide-and-conquer proposal failure if there are > more YES than NO votes by a reasonable margin (whether the "reasonable margin" > in question is 0, 100, or a percentage of votes is immaterial to this > discussion) but the YES votes are spread out over sufficiently many newsgroup > names that the group fails. This does not properly reflect the desires of the > voters. Since the voters can vote for *all* the names they're interested in, I hardly think this is likely. And as I noted before, any scheme is subject to pathological cases: "Voting Paradoxes". Also, ties for names are unlikely, and can be reasonably solved by net.god fiat. Give someone (spaf?) the tiebreaking vote. > An alternative suggestion is that if there are more YES than NO votes by > whatever margin is decided upon, the name with the most YES votes will be > created (or used, if it is an existing group). This is a reasonable alternative, but to take the sci.aquaria farce as an example, it's not really meaningful to combine votes for sci.aquaria and for rec.aquaria as mutual "yes" votes. A mutual hate society, maybe. -- `-_-' Peter da Silva <peter@ficc.uu.net> <peter@sugar.hackercorp.com>. 'U` -------------- +1 713 274 5180. "*Real* wizards don't whine about how they paid their dues" -- Quentin Johnson quent@atanasoff.cs.iastate.edu
allbery@NCoast.ORG (Brandon S. Allbery) (11/13/89)
As quoted from <6931@ficc.uu.net> by peter@ficc.uu.net (Peter da Silva): +--------------- | > An alternative suggestion is that if there are more YES than NO votes by | > whatever margin is decided upon, the name with the most YES votes will be | > created (or used, if it is an existing group). | | This is a reasonable alternative, but to take the sci.aquaria farce as an | example, it's not really meaningful to combine votes for sci.aquaria and for | rec.aquaria as mutual "yes" votes. A mutual hate society, maybe. +--------------- Richard Sexton and Oleg Kiselev aside, most people voted FOR sci.aquaria on the merits of the "aquaria", not the "sci". I submit that in fact, the votes could be combined to reflect interest in a newsgroup on the subject under *any* name -- even in the case of *.aquaria. ++Brandon -- Brandon S. Allbery allbery@NCoast.ORG, BALLBERY (MCI Mail), ALLBERY (Delphi) uunet!hal.cwru.edu!ncoast!allbery ncoast!allbery@hal.cwru.edu bsa@telotech.uucp *(comp.sources.misc mail to comp-sources-misc[-request]@backbone.site, please)* *Third party vote-collection service: send mail to allbery@uunet.uu.net (ONLY)* expnet.all: Experiments in *net management and organization. Mail me for info.
edhew@xenitec.on.ca (Ed Hew) (11/13/89)
In article <4470@solo10.cs.vu.nl> maart@cs.vu.nl (Maarten Litmaath) writes: >I like the proposal, but there's another point to consider: what if there's >a close finish of 2 (or more) groups? >Example: > YES NO > sci.aquaria 666 500 > news.aquaria 665 500 > >Shouldn't both groups be created? Institute a required 2/3 minimum majority in addition to the minimum 100 yes-no plurality, and the above problem goes away, as it should. >(Monica Cellio) | maart@cs.vu.nl, mcsun!botter!maart Ed. A. Hew Authorized Technical Trainer Xeni/Con Corporation work: edhew@xenicon.uucp -or- ..!{uunet!}utai!lsuc!xenicon!edhew ->home: edhew@xenitec.on.ca -or- ..!{uunet!}watmath!xenitec!edhew
peter@ficc.uu.net (Peter da Silva) (11/14/89)
?? An alternative suggestion is that if there are more YES than NO votes by ?? whatever margin is decided upon, the name with the most YES votes will be ?? created (or used, if it is an existing group). Me: This is a reasonable alternative, but to take the sci.aquaria farce as an Me: example, it's not really meaningful to combine votes for sci.aquaria and for Me: rec.aquaria as mutual "yes" votes. A mutual hate society, maybe. > Richard Sexton and Oleg Kiselev aside, most people voted FOR sci.aquaria on > the merits of the "aquaria", not the "sci". I noticed. > I submit that in fact, the votes > could be combined to reflect interest in a newsgroup on the subject under > *any* name -- even in the case of *.aquaria. In this case they would have voted for *both* groups. Combining all the votes for all the groups would then count these folks *twice*. Can you say "let's stuff the ballot box by putting lots of names in?". Sure you can. -- `-_-' Peter da Silva <peter@ficc.uu.net> <peter@sugar.hackercorp.com>. 'U` -------------- +1 713 274 5180. "*Real* wizards don't whine about how they paid their dues" -- Quentin Johnson quent@atanasoff.cs.iastate.edu
richard@gryphon.COM (Richard Sexton) (11/14/89)
In article <1989Nov12.162221.10189@NCoast.ORG> allbery@ncoast.ORG (Brandon S. Allbery) writes: > >most people voted FOR sci.aquaria on the merits of the "aquaria", not the "sci" This is your opinion, which is a good thing, because if you had stated that as a fact, I would have been duty bound to point out that you are wrong. -- This article was made from the finest quality words and sentances. Minor imperfections in syntax, like the grain in fine leather, serve to enhance it's beauty. richard@gryphon.COM {routing site}!gryphon!richard
" Maynard) (11/14/89)
In article <22222@gryphon.COM> richard@gryphon.COM (Richard Sexton) writes: >In article <1989Nov12.162221.10189@NCoast.ORG> allbery@ncoast.ORG (Brandon S. Allbery) writes: >>most people voted FOR sci.aquaria on the merits of the "aquaria", not the "sci" >This is your opinion, which is a good thing, because if you had stated that >as a fact, I would have been duty bound to point out that you are wrong. Prove it. -- Jay Maynard, EMT-P, K5ZC, PP-ASEL | Never ascribe to malice that which can jay@splut.conmicro.com (eieio)| adequately be explained by stupidity. {attctc,bellcore}!texbell!splut!jay +---------------------------------------- Shall we try for comp.protocols.tcp-ip.eniac next, Richard? - Brandon Allbery
jeffd@ficc.uu.net (jeff daiell) (11/14/89)
Richard Sexton writes: > > > >most people voted FOR sci.aquaria on the merits of the > >"aquaria", not the "sci" > > This is your opinion, which is a good thing, because if you had stated that > as a fact, I would have been duty bound to point out that you are wrong. ... and I will feel duty bound to point out that YOU are wrong. But let's ask those who voted on sci.aquaria: If you voted yes, please send me your response to this question: I voted YES to get an aquaria group, regardless of name. ___ I would have voted NO to rec.aquaria. ___ And if you voted no, please send me your response to this question: I voted NO because of the name. ___ I would have voted NO to any aquaria group. ___ Thanks. Jeff -- "'Tis not too late to seek a newer world." -- Alfred, Lord Tennyson
SLOANE@kuhub.cc.ukans.edu (Bob Sloane) (11/14/89)
In article <6950@ficc.uu.net>, peter@ficc.uu.net (Peter da Silva) writes: > In this case they would have voted for *both* groups. > > Combining all the votes for all the groups would then count these folks > *twice*. Can you say "let's stuff the ballot box by putting lots of names > in?". Sure you can. You don't need to add up the YES votes for all the groups. Any single PERSON who votes YES for any of the names could be counted as ONE vote in favor of the group charter or topic. The vote taker could report the total number of ballots received, and the number of ballots that contained at least one YES vote. These numbers would reflect the popularity of the charter, as opposed to the name. -- USmail: Bob Sloane, University of Kansas Computer Center, Lawrence, KS, 66045 E-mail: sloane@kuhub.cc.ukans.edu, sloane@ukanvax.bitnet, AT&T: (913)864-0444
allbery@NCoast.ORG (Brandon S. Allbery) (11/15/89)
As quoted from <22222@gryphon.COM> by richard@gryphon.COM (Richard Sexton): +--------------- | In article <1989Nov12.162221.10189@NCoast.ORG> allbery@ncoast.ORG (Brandon S. Allbery) writes: | >most people voted FOR sci.aquaria on the merits of the "aquaria", not the "sci" | | This is your opinion, which is a good thing, because if you had stated that | as a fact, I would have been duty bound to point out that you are wrong. +--------------- Are you claiming your response as a fact? If so, I am duty-bound to require proof. If your proof consists of the results of the poll you cited during the vote, don't bother responding. ++Brandon -- Brandon S. Allbery allbery@NCoast.ORG, BALLBERY (MCI Mail), ALLBERY (Delphi) uunet!hal.cwru.edu!ncoast!allbery ncoast!allbery@hal.cwru.edu bsa@telotech.uucp *(comp.sources.misc mail to comp-sources-misc[-request]@backbone.site, please)* *Third party vote-collection service: send mail to allbery@uunet.uu.net (ONLY)* expnet.all: Experiments in *net management and organization. Mail me for info.
dww@stl.stc.co.uk (David Wright) (11/15/89)
In article <4470@solo10.cs.vu.nl> maart@cs.vu.nl (Maarten Litmaath) writes: >I like the proposal, but there's another point to consider: what if there's >a close finish of 2 (or more) groups? >Example: > YES NO > sci.aquaria 666 500 > news.aquaria 665 500 > >Shouldn't both groups be created? This shows the weakness of MAUVE schemes - people cannot express a preference. The above example suggests that 665 people want any aquaria group, and even if they have a strong preference for one or other name, they would rather have the less preferred name than no group. But the effect is that the decision on which name to choose is taken by the one voter who voted for sci.aquaria only. With STV, everyone expresses their preference; if most people would rather have one name than another, that name will win. What is more, people who prefer unpopular name A, but would also be happy with name B, will probably have their second preference vote counted so get a say towards B (and so on). STV elections tend both to express people's choice well and also give more voters one of their choices (if not their first choice). Of course for a net vote there would also be a NO vote, and to 'win' the vote would have to produce more 'positive' votes than NO votes by a margin of >100. From the voter's point of view, STV is "as simple as 1-2-3" (to quote the Electoral Reform Society). People could simply vote (e.g.) 1 net.this 2 no 3 net.that 4 other.thing Note that in this case the voter wants the group to exist if called net.this, otherwise they'd prefer that there was no group at all. But if the group *is* created with another name (i.e. if neither 'net.this' or NO gets enough support), this voter would prefer the group to be called net.that. I think that letting NO voters still express a further opinion is quite reasonable, and consistant with the idea of giving most voters one of their choices, if not the first. I doubt that more than a few would abuse this by voting for silly choices, especially if we do not allow 'write in' votes (I used to think these were a good idea, but have been convinced by the arguments made against them). Note also that in an STV ballot, your lower choice CAN NOT affect the probability of your higher choice(s) winning; the lower choice only gets counted if your higher choice has already been eliminated. Having pushed the cause of STV, I'd better post a "proposal in 25 lines". Not tonight though :-) I'll do it soon. The rest of this article can be skipped by people who are not interested in the technicalities of STV; you don't need to know any of what follows in order to vote in an STV ballot. What we've been talking about on the net is actually a special case of STV, in which there is only one 'candidate' to be elected, (this special case is sometimes called the alternative vote system). STV is especially effective in electing a committee or group of councillors/representatives/etc. It is effective because it gives a result roughly proportional to the voter preference, and because most voters will get some of their choices (not necessarily their first ones) elected. For example, if electing a group of 5 people, where the candidates represent distinct views A or B, and the voters split 60/40 to view A, a 'first past the post' vote will tend to give each A candidate 60% of the vote (hence electing 5 A's), whereas STV will tend to elect 3 A and 2 B candidates. The reason it is called the 'Single Transferable Vote' is that each elector gets one vote only, but that vote can be transfered from a candidate who cannot 'use' it to one who can. In a typical vote for a committee, the surplus votes of winning candidates are transferred first (normally by transferring the unused fraction of each vote - e.g. if a candidate has got 150% of the votes required for election, their votes will be transfered at the value of .33, as .67 has already been 'used' in electing them). Then if no more candidates are elected, the candidate with the lowest vote is eliminated and their votes are transferred (at full value, as they haven't been 'used' yet). And so on, until the required number of candidates are elected. Sounds complicated to count, and it is a bit, but having assisted at a number of counts where up to 3000 ballot papers with 40 candidates for a committee of 15 are being counted, I can assure you that its quite possible to do -- though that did take several people all of a long day!. Fortunately the 'election' of a news group by STV would be much simpler to count, as we only want one winner, and would normally only have a few names (plus NO) to choose between. The ballots of the name with the least first preference votes would be transfered to the voter's next choice, repeating until a name has enough votes to pass the 'quota', i.e. over half of the (remaining) votes in this case. No doubt people who know awk better than I could write an automatic vote parser and counter, but even without it, manually or semi-manually counting the few hundred votes that a very successful net vote produces should be quite possible in an evening. It would also need a bit of editing (or another awk script) to prepare a schedule of how everyone voted; previous posters have suggested suitable formats, so I won't repeat them here. If required, special voting rules could be defined to make counting easy, such as are being used now for the [sci|comp].groupware vote. That's enough for now; I'll post the 25 line proposal RSN! -- 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 Living in a country without a written constitution means having to trust in the Good Will of the Government and the Generosity of Civil Servants.
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (11/16/89)
I hate to disagree with some rational people, but I think the idea of deciding the name before deciding to have a group is like naming the baby before getting pregnant. When a group name is in doubt a vote should be taken on the charter first, then a name should be chosen by simple majority vote. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "The world is filled with fools. They blindly follow their so-called 'reason' in the face of the church and common sense. Any fool can see that the world is flat!" - anon
shafer@elxsi.dfrf.nasa.gov (Mary Shafer) (11/16/89)
In article <1989Nov14.194231.11595@NCoast.ORG> allbery@NCoast.ORG (Brandon S. Allbery) writes: In article <1989Nov14.194231.11595@NCoast.ORG> allbery@NCoast.ORG (Brandon S. Allbery) writes: >As quoted from <22222@gryphon.COM> by richard@gryphon.COM (Richard Sexton): >+--------------- >| In article <1989Nov12.162221.10189@NCoast.ORG> allbery@ncoast.ORG (Brandon S. Allbery) writes: >| >most people voted FOR sci.aquaria on the merits of the "aquaria", not the "sci" >| >| This is your opinion, which is a good thing, because if you had stated that >| as a fact, I would have been duty bound to point out that you are wrong. >+--------------- >Are you claiming your response as a fact? If so, I am duty-bound to require >proof. If your proof consists of the results of the poll you cited during the >vote, don't bother responding. If all it takes is one example, here's proof that Richard is right. I didn't vote for sci.aquaria because I like fish or aquaria, I voted for it because the namespace fanatics made me angry with their rantings. (I'm an ichthyophobe who cannot ever bear to think about touching a fish. The only fish I like are in 3-1/2 oz cans, packed in water, and I'm not even too fond of them.) -- Mary Shafer shafer@elxsi.dfrf.nasa.gov ames!elxsi.dfrf.nasa.gov!shafer NASA Ames-Dryden Flight Research Facility, Edwards, CA Of course I don't speak for NASA
shafer@elxsi.dfrf.nasa.gov (Mary Shafer) (11/16/89)
In article <1989Nov14.194231.11595@NCoast.ORG> allbery@NCoast.ORG (Brandon S. Allbery) writes: In article <1989Nov14.194231.11595@NCoast.ORG> allbery@NCoast.ORG (Brandow S. Allbery) writes: >As quoted from <22222@gryphon.COM> by richard@gryphow.COM (Richard Sextow): >+--------------- >| In article <1989Nov12.162221.10189@NCoast.ORG> allbery@ncoast.ORG (Brandow S. Allbery) writes: >| >most people voted FOR sci.aquaria on the merits of the "aquaria", not the "sci" >| >| This is your opiniow, which is a good thing, because if you had stated that >| as a fact, I would have been duty bound to point out that you are wrowg. >+--------------- >Are you claiming your response as a fact? If so, I am duty-bound to require >proof. If your proof consists of the results of the poll you cited during the >vote, dow't bother responding. If all it takes is one example, here's proof that Richard is right. I didn't vote for sci.aquaria because I like fish or aquaria, I voted for it because the namespace fanatics made me angry with their rantings. (I'm an ichthyophobe who cannot ever bear to think about touching a fish. The only fish I like are in 3-1/2 oz cans, packed in water, and I'm not even too fond of them.) -- Mary Shafer shafer@elxsi.dfrf.nasa.gov ames!elxsi.dfrf.nasa.gov!shafer NASA Ames-Dryden Flight Research Facility, Edwards, CA Of course I don't speak for NASA
mesard@bbn.com (Wayne Mesard) (11/16/89)
davidsen@crdos1.UUCP (bill davidsen) writes: >When a group name is in doubt a vote >should be taken on the charter first, then a name should be chosen by >simple majority vote. That's essentially what we do now, except the name is decided (barring a recent example) by consensus rather than by an explicit vote. The problem is that the intent of the proponents of a group is often poorly correlated with what actually winds up in the group. <insert standard list of examples here.> It's my impression that the opponents of sci.you-know-what don't think that a technical aquarium group is a Bad Thing. Rather, they believe that the group we wind up with will be significantly different than the charter. -- void *Wayne_Mesard(); Mesard@BBN.COM BBN, Cambridge, MA
etxtsg@solsta.ericsson.se (Thomas Grennefors TX/DKF) (11/16/89)
>>>>> On 15 Nov 89 17:18:10 GMT, davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) said: > I hate to disagree with some rational people, but I think the idea of > deciding the name before deciding to have a group is like naming the > baby before getting pregnant. When a group name is in doubt a vote > should be taken on the charter first, then a name should be chosen by > simple majority vote. I think a name controversy when voting for a new group can make people, who like the object of the group, but dilslikes the name of the group, to wote no. I think that the name of the group should be decided before voting on the creation of that group. Look at the recent vote for rec.radio.shortwave. What could have been a major dissagrement on the name, was solved before the voting, and the group passed with 265 yes vs 10 no. Maby it should have had the same ratio of yes/no votes even if the name vote hadn't been done, but i dont think so. And if we decides on a 2 to 1 yes-no ratio shouldn't there be a minimum level of yes votes that has to be reached. If ther's only the 2-1 Ratio a group can pass with 10 yes 5 no, wich i dont think shows a big interest for that particular group. My proposal for a voting sheme is: 1. Call for discussion for X days. 2. Call for name vote if a name controversy exists else goto 4 3. Call for discussion Y days. 4. Call for votes for Z days, a pass with 2/3 yes votes and a minimum of N yes votes. The unknown variables X,Y,Z and N should be discussed. -- Thomas Grennefors etxtsg@solsta.ericsson.se Ericsson Telecom ,Karlstad, Sweden. Member of THHS krfc (tlps) --------------------------------------------------------------- There's too much month at the end of the money.
dww@stl.stc.co.uk (David Wright) (11/18/89)
In article <2812@cpoint.UUCP> alien@cpoint.UUCP (Alien Wells) writes:
# ... Perhaps there should be an entry for simply YES or NO ...
#
#>This permits one of the obvious STV pathologies unless the changes from
#>the previous paragraph are incorporated: it is possible to have a vote
#>in which several hundred people vote YES, no one casts a NO vote,
#>yet no newsgroup is created because no name received more than 100
#>YES votes.
#
#I think that this is actually a place where this proposal is significantly
#better than STV. If people are actually voting properly (ie: everything that
#they would find acceptable is voted YES), the only way that you could have
#all proposals fail is if each vote taken independantly would fail.
This shows a misunderstanding of how the STV process works. Assuming that
a NO vote was one of the options (and with the addition of a special ">100"
rule to match our present system), a proper STV ballot would achieve exactly
what you want. Assuming the voters were putting ALL their preferences in
order (including the NO if that was their view), then ballots for unpopular
first preferences would be transferred to the voter's second preference, and
so on, until one preference had over 50% of the (remaining) votes. And
that's it. Maximum voter choice - e.g. no need to not vote for your real
preference in case that was so unpopular that your vote was wasted (the
fault of the traditional voting system), nor the problem of MAUVE, that if
I vote YES to all three names (say) to ensure the group is created, I loose
the ability to influence WHICH name is chosen.
OK, OK, I'll write that 25 line proposal.....
--
Regards, "None shall be enslaved by poverty, ignorance or conformity"
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