jeffrey@algor2.algorists.com (Jeffrey Kegler) (11/20/89)
Consider, as examples, comp.lang.c and comp.unix.wizards. These groups have 1) high value; 2) even higher bandwidth and 3) a fair degree of abuse. Suppose the following were done. Create new newsgroup and designate a moderator. Have the moderator do the moderation partially via an automated interface. One way the interface could behave is as follows. A number of posters are "authorities". One type of good choice for authorities is people who answer lots of questions, usually in an accurate manner. The interface has a list of authorities and passes all their stuff through, untouched and without delay or interaction with the moderator. Postings by others are dealt with by the moderator. He can deal with the remainder by a combination of strategies. Stuff that he deems inappropriate he rejects (informing the poster) or asks for a rewrite of. If the newsgroup volume is high, he puts questions in a waiting list (informing the poster of this). If a poster submits reasonably good stuff, he is added to the "authorities" list for automatic pass through. I will be interested in hearing from the net.wise on the (in)advisability of the above. -- Jeffrey Kegler, Independent UNIX Consultant, Algorists, Inc. jeffrey@algor2.ALGORISTS.COM or uunet!algor2!jeffrey 1762 Wainwright DR, Reston VA 22090
chip@ateng.com (Chip Salzenberg) (11/21/89)
According to jeffrey@algor2.algorists.com (Jeffrey Kegler): >Consider, as examples, comp.lang.c and comp.unix.wizards. > >[Proposed:] >The interface has a list of authorities and passes all their stuff >through, untouched and without delay or interaction with the >moderator. Postings by others are dealt with by the moderator. I like this idea. I'd read these groups more often if I knew that the correct answers from Doug Gwyn, Guy Harris, Chris Torek, Henry Spencer, et al weren't often contradicted and sometimes flamed by ignorant newbies, especially those who flame hardest because they have Good Intentions. On the other hand, we'd have to be very sure to arrange for a backup moderator... -- You may redistribute this article only to those who may freely do likewise. Chip Salzenberg at A T Engineering; <chip@ateng.com> or <uunet!ateng!chip> "Did I ever tell you the Jim Gladding story about the binoculars?"
tneff@bfmny0.UU.NET (Tom Neff) (11/21/89)
In article <25684AE6.26256@ateng.com> chip@ateng.com (Chip Salzenberg) writes: > I'd read these groups more often if I knew that the >correct answers from Doug Gwyn, Guy Harris, Chris Torek, Henry Spencer, et >al weren't often contradicted and sometimes flamed by ignorant newbies, >especially those who flame hardest because they have Good Intentions. Keeps 'em honest though. Even "experts" make mistakes. Usenet would be a poorer place if we introduced the caste system. I mean, why stop there? You could have all sorts of intermediate privilege levels: always allowed to CROSS-post, allowed to cross-post but only via moderator, always allowed to post with nondefault expire, ditto only via moderator, NEVER allowed to post, always allowed to post LONG articles, ditto via moderator, etc etc etc. Future net.kafkaesque.flunkies await these developments with relish... As we learned to ask in my middle school social studies class (during a brief aberrant phase of liberality in education): "Who decides?" -- Canada -- a few acres of snow. ^v^v^ Tom Neff -- Voltaire v^v^v tneff@bfmny0.UU.NET
linimon@attctc.Dallas.TX.US (Mark Linimon) (11/21/89)
In article <1989Nov19.223239.26100@algor2.algorists.com> jeffrey@algor2.ALGORISTS.COM (Jeffrey Kegler) writes: >[on semi-moderated groups] >One way the interface could behave is as follows. A number of posters >are "authorities". One type of good choice for authorities is people >who answer lots of questions, usually in an accurate manner. 1. My first observation is that the pro-anarchy forces would have a hard time with this; a compromise would be to create a seperate newsgroup which the moderator(s) approved selected articles to. This is essentially what Greg's doing now in news.announce.newgroups; I don't know if he uses cross-posting or not. Note that I am not necessarily agreeing or disagreeing with this view, just making an observation. 2. The more general case, which has been mentioned by several folks, is to expand the KILL file concept to a more general one: essentially a simple expert system where you could set the rules, such as "assign articles by default to 0" "add to articles from Spaf +10" "add to articles from Joe Flamer -10" "add to articles with PC in the title -7" "set current viewing threshhold to +5" and so forth. Then everyone could assign their own numeric "authoritative- ness". In fact, I could see hierarchies of moderators, i.e. "assign articles by default to 0" "use weightset last posted by Joe Flamer * -10" etc. Of course this would require a new generation of newsreader; volunteers? Mark Linimon attctc!linimon
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (11/21/89)
A group in which there are self-styled athorities and the moderator rejects postings which disagree with them? Boy that sounds great, let's get rid of all that dissent. The athorities know best, and shouldn't be bothered with rabble. Perhaps you might have noticed that the above mentioned group has been guilty of posting a flame or two? Why bother to have a group if dissenting opinions are to be banned? Let's start a 'partyline' heirarchy for these, like partyline.lang.c. -- 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
jeffrey@algor2.algorists.com (Jeffrey Kegler) (11/21/89)
Tom> Article <14940@bfmny0.UU.NET> tneff@bfmny0.UU.NET (Tom Neff) Chip> Article <25684AE6.26256@ateng.com> chip@ateng.com (Chip Salzenberg) Chip> I'd read these groups more often if I knew that the Chip> correct answers from Doug Gwyn, Guy Harris, Chris Torek, Henry Spencer, et Chip> al weren't often contradicted and sometimes flamed by ignorant newbies, Chip> especially those who flame hardest because they have Good Intentions. Tom> Keeps 'em honest though. Even "experts" make mistakes. Usenet would be Tom> a poorer place if we introduced the caste system. I mean, why stop Tom> there? ... Much larger than the problems of finding someone to keep the net "authorities" honest, would be to keep them from flaming each other. I anyone on the above list were to err (unlikely, but so was the Berlin Wall falling this year), the others would be very fast to correct him (gently I hope). A couple of clarifications: Semi-moderated groups should never be granted a monopoly. That is, if a semi-moderated version of comp.lang.c starts up, the old group should remain. Also multiple semi-moderated groups on the same topic should be allowed (and probably encouraged). The competition will keep moderation from getting stiffling. Second, the people allowed "pass-through" posting, the "authorities" as I called them, would not be limited to net.luminaries like the above. Lesser lights who contribute to the discussion would also be "authorities". Perhaps a better term than authorities would be "pass-through posters". -- Jeffrey Kegler, Independent UNIX Consultant, Algorists, Inc. jeffrey@algor2.ALGORISTS.COM or uunet!algor2!jeffrey 1762 Wainwright DR, Reston VA 22090
chip@ateng.com (Chip Salzenberg) (11/22/89)
According to tneff@bfmny0.UU.NET (Tom Neff): >Keeps 'em honest though. Even "experts" make mistakes. Sure. I don't mind the correct corrections. It's the incorrect corrections that get my goat. >As we learned to ask in my middle school social studies class (during a >brief aberrant phase of liberality in education): "Who decides?" We're not talking about political discourse here, just techical discussion. So a moderation who knows C or Unix well enough to filter out junk would be a big help. -- You may redistribute this article only to those who may freely do likewise. Chip Salzenberg at A T Engineering; <chip@ateng.com> or <uunet!ateng!chip> "Did I ever tell you the Jim Gladding story about the binoculars?"
martin@mwtech.UUCP (Martin Weitzel) (11/22/89)
In article <25684AE6.26256@ateng.com> chip@ateng.com (Chip Salzenberg) writes: >According to jeffrey@algor2.algorists.com (Jeffrey Kegler): >>Consider, as examples, comp.lang.c and comp.unix.wizards. >> >>[Proposed:] >>The interface has a list of authorities and passes all their stuff >>through, untouched and without delay or interaction with the >>moderator. Postings by others are dealt with by the moderator. I fear, this could invite some individuals to try themselves on forgery, so that they appear as 'authority'. You may argue, that someone who wants discussions on 'pronounciation' and 'how to tell a wizzard' does not have the knowledge to do forgery, but I allways watched, that the knowledge among the "bad guys" spreads fast and that there is more energy to reach a goal, if there is some kind of price to win. So, writing obsolete stuff to a news group might occur more often, if it is *not* as easy as now. The list of authorities can soon become worthless, if there is no way of having an 'reliable signature' (I have some public/private key scheme in mind?). >I like this idea. I'd read these groups more often if I knew that the >correct answers from Doug Gwyn, Guy Harris, Chris Torek, Henry Spencer, et [rest deleted] I fully agree. A future enhancement could be to automate the process of adding/deleting people from this 'authority' list by sending a 'rating' to the selection interface mentioned above (eg +3 for 'very useful' to -3 for 'very obsolete'). The 'rn'-programs could be modified to accept this rating after one has read an article (I said, *future* enhancement ...). There could also be some 'ageing', so that after some time disqualified people get qualified again and negative ratings might be discarded if they do not come from persons, who themselves have accumulated a certain amount af positive rating. But the point that is crucial in any case is the reliable signature. <Martin Weitzel -- martin@mwtech.UUCP)
allbery@NCoast.ORG (Brandon S. Allbery) (11/22/89)
Deja vu... Both this suggestion: As quoted from <1989Nov19.223239.26100@algor2.algorists.com> jeffrey@algor2.ALGORISTS.COM (Jeffrey Kegler) writes: +--------------- | [on semi-moderated groups] | One way the interface could behave is as follows. A number of posters | are "authorities". One type of good choice for authorities is people | who answer lots of questions, usually in an accurate manner. +--------------- And this suggestion: As quoted from <10275@attctc.Dallas.TX.US> by linimon@attctc.Dallas.TX.US (Mark Linimon): +--------------- | 2. The more general case, which has been mentioned by several folks, is | to expand the KILL file concept to a more general one: essentially | a simple expert system where you could set the rules, such as | | "assign articles by default to 0" | "add to articles from Spaf +10" | "add to articles from Joe Flamer -10" | "add to articles with PC in the title -7" | "set current viewing threshhold to +5" +--------------- Have been suggested by Brad Templeton. In fact, he has already *written* the software described above; it's called "newsclip". Unfortunately, I think he made it commercial, but someone could always write a PD version. (Me? I think not; I've got programming projects coming out my wazoo already.) ++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.
jamesd@qiclab.UUCP (James Deibele) (11/24/89)
In article <14940@bfmny0.UU.NET> tneff@bfmny0.UU.NET (Tom Neff) writes: >In article <25684AE6.26256@ateng.com> chip@ateng.com (Chip Salzenberg) writes: >> I'd read these groups more often if I knew that the >>correct answers from Doug Gwyn, Guy Harris, Chris Torek, Henry Spencer, et >>al weren't often contradicted and sometimes flamed by ignorant newbies, >>especially those who flame hardest because they have Good Intentions. > >Keeps 'em honest though. Even "experts" make mistakes. Usenet would be >a poorer place if we introduced the caste system. I mean, why stop >there? You could have all sorts of intermediate privilege levels: >always allowed to CROSS-post, allowed to cross-post but only via >moderator, always allowed to post with nondefault expire, ditto only via >moderator, NEVER allowed to post, always allowed to post LONG articles, >ditto via moderator, etc etc etc. Future net.kafkaesque.flunkies await >these developments with relish... > Agreed wholeheartedly. When asking a question, you have to be prepared to sift through the replies, sorting them as to their reliability. It would be nice to have some sort of automagic "authority" rating on each post (lessee, he gets a 3 on movies, a 7 on C compilers, and a 0 on relationships), but it just ain't possible. Say you had a C question, and K&R both spoke up and said "this is the way it is" but your C compiler vendor did things their own way, so K&R would really be saying "this is the way it should be." Some bozo who happened to use the same compiler might have a higher rating, etc. That's one of the real problems when entering into a new field: whose advice is reliable and whose advice should be tossed in the bit bucket. Given enough time, you can usually see a consensus of sorts build, as in "well, I think you're wrong on this nit, but basically I agree." The problem is that it takes time for your BS filters to build up enough to protect you. There've been times when I really had wished that someone would take and filter the stuff, then resell the knowledge thus gained for a reasonable price: case in point would be using an RLL (or Perstor) controller with a drive. If it works, you get 1.5 (or 1.9) the disk space you would normally have, which is a good deal <if it works>. But neither the controller manufacturer or the drive manufacturere will probably guarantee that the combination will work. It would be nice to send a message to someone who would report back that it worked 25 times, eventually failed 4 times, and failed immediately twice. As the same questions get asked over and over, the urge to respond dwindles, leaving the newbies answering instead of the "old pros." The only practical solution seems to be keeping the noise level down so that most of the old pros stick around for a reasonable amount of time as the newbies mature. (Or so it seems IMHO.) -- James Deibele jamesd@qiclab BBSs: (503) 760-1473 or (503) 761-7451 TECHBooks: The Computer Book Specialists --- Voice: (503) 646-8257 12600 SW 1st Beaverton, OR 97005 --- Book reviewers wanted for computer science & electronics - contact us for more information.
anw@maths.nott.ac.uk (Dr A. N. Walker) (11/25/89)
In article <1989Nov19.223239.26100@algor2.algorists.com> jeffrey@algor2.ALGORISTS.COM (Jeffrey Kegler) writes: [Proposal for automatic pass-through of known experts, moderation of other postings in certain groups.] This is not so different from the proposal I made a while back (and which raised just one, rather sniffy, ripple), which I therefore repeat. Basically, my idea was that in the "certain groups", articles whose subject began "Re: " but which lacked an "Expert: " header would be bounced. For "Expert" read "Approved" if you like, and for "bounced" read "sent to a moderator" if you like. Unlike present-day "Approved" headers, "Expert" headers would be retained, by default. Consequences: a) New topics arise freely, as at present. b) Chris, Henry, Doug, Guy, Jeffrey, Andy [:-)], etc., who know what they are doing, can reply easily. c) Once [CHDGJA]* has replied, anyone can follow-up that reply easily. d) Any net.neophyte who leaps in before [CH...]* gets diverted. Thus, a successful reply must either be expert or at least be in response to an expert. e) The software mods are easy, and only need to be made by backbone sites for the scheme to work adequately. f) Anyone could self-proclaim expert status, but those who did so and then messed up would rightly look very foolish. -- Andy Walker, Maths Dept., Nott'm Univ., UK. anw@maths.nott.ac.uk