webber@porthos.rutgers.edu (Bob Webber) (06/20/88)
In article <4542@gryphon.CTS.COM>, richard@gryphon.CTS.COM (Richard Sexton) writes: > In article <28.UUL1.3#935@aocgl.UUCP> tmanos@aocgl.UUCP (Theodore W. Manos) writes: > >Do I finally hear some peace and quiet coming??? :-) Here's my vote too. > >I also will offer to moderate a group - even one I don't read! > ... > Hmm. Can I make the tacit assumption that if you don't read a group it's > because you arn't interested in the content ? Could I even go so far as > to say you don't know anything about the subject ? > > Hey, makes sense to me. > > I offer to moderate rec.guns. Really. Glad you finally saw the light. A moderator who actually read what they were moderator would be BIASED. Look at the vast improvement to news.announce.conferences now that we have a moderator who doesn't know the difference between a conference and a tradeshow. Of course, since I have an opinion about everything, it is difficult for me to find a group that I could moderate without bias. However, since I have the least opinion about the current news group creation process, I guess I could moderate news.groups. Anything for a good cause. By the way, since my moderation policy will be simple (all messages must contain a prime number of words in their text and signatures no longer than one line), there is no need for all of this stuff to flow thru me. I will dub knight-moderators at any site that has a user who can list the prime numbers between 300 and 30,000 and count up to one. Knight-moderators will be permitted to include their names on the approved lines instead of mine when posting messages (either theirs or those of others) into news.groups. Un-dubbed people who make such postings or dubbed people whose dubbing has been revoked who make such postins will be viewed as forgers and will be added to the list of people who receive a copy of every one of my postings as mail messages. -------- BOB (webber@athos.rutgers.edu ; rutgers!athos.rutgers.edu!webber)
davidbe@sco.COM (The Cat in the Hat) (06/21/88)
And I volunteer to moderate rec.music.rock (since it looks like it will be
created). I'm sure everyone will be completely understanding about my
beliefs in just what rock is...and will excuse the fact that I won't allow
any crossposting (because folk isn't rock, classical isn't rock, the Beatles
have their own newsgroup), that discussions of CD's won't be allowed,
and that funk, reggae, punk, and modern music won't be discussed because they
are "not rock". And I'll keep the volume way down by not allowing every third
posting, thereby cutting out a third of the volume in my group.
Hey, it'll be the one newsgroup I'll enjoy reading.
~~~~
But seriously folks...The consequences of moderating the net will be these:
Fewer new topics of discussion, the possiblity of never having a new topic
show up (i.e. Prince in the rec.music.rock example), and the extreme overload
of the moderator. Not to mention the EXTREME added cost in terms of connect
time and phone costs for the moderator.
Moderating newsgroups works best for groups with a specific agenda, a moderator
with the time to moderate, or a group that goes outside of the network.
It works worst for groups with HIGH volume, high emotional content (such as
talk.politics or news.groups) or high amounts of crossposting.
Universally applied it won't work at all.
Who wants to moderate talk.bizarre?
--
David Bedno (aka The Cat in the Hat) Now appearing at: davidbe@sco.COM -OR-
...!{uunet,decvax!microsoft,ucbvax!ucscc}!sco!davidbe -OR-
At home: 408-425-5266 At work: 408-425-7222 x5123 (I'm probably here...)
Disclaimer: Not SCO's opinions. At least not that they've told me.
"...think of thermonuclear fusion, our mighty but mischevious friend."andrew@teletron.UUCP (Andrew Scott) (06/23/88)
In article <659@scovert.sco.COM>, davidbe@sco.COM (The Cat in the Hat) writes: > > But seriously folks...The consequences of moderating the net will be these: > Fewer new topics of discussion, the possiblity of never having a new topic > show up (i.e. Prince in the rec.music.rock example), and the extreme overload > of the moderator. Not to mention the EXTREME added cost in terms of connect > time and phone costs for the moderator. The most important consequence would still be that the overall net bandwidth requirements would be sharply reduced, which I hope most everybody agrees would be a "good thing". Brian Reid posted some numbers a while ago that indicated that the net had grown from 80 Meg a month to 130 Meg a month since last October. The net simply cannot continue to grow at this pace. The overloaded moderator argument could be solved by having multiple moderators, perhaps by a regional basis. Alternatively, a person might moderate for a month at a time before turning over to another "moderator of the month". Neither of these solutions would be practical with the current news software system, but they could be implemented for a "MODNET". >Moderating newsgroups works best for groups with a specific agenda, a moderator > with the time to moderate, or a group that goes outside of the network. > It works worst for groups with HIGH volume, high emotional content (such as > talk.politics or news.groups) or high amounts of crossposting. > Universally applied it won't work at all. I think that USENET would be have an entirely different character if it was completely moderated. It seems to me that a lot of mindless posting goes on in many of the high volume newsgroups, postings that might not be made in the first place if the groups were known to be moderated. For an example, let's compare rec.humor and rec.humor.funny, since they represent what moderation does to a newsgroup. Many people in rec.humor post jokes that have been heard before, prolonged "run-on" puns that cease to be funny, and flames at people for screwing up the punchline. Postings in rec.humor.funny are (for the most part) intelligent and original. You get the idea. Take a look at many of the other high volume newsgroups. The same kind of things go on. For example, in rec.arts.startrek or comp.lang.c, we get a zillion answers whenever a neophyte asks a simple question. A moderator could post a monthly introductory posting of commonly asked questions - this alone would be a huge improvement. Another pro-moderation argument typically heard is the "low signal-noise ratio" argument. I wholeheartedly agree. However, noise is too polite a word for some of the crap I've seen in recent months. The net used to be a very civil place. It's very discouraging to see it disintegrate into a cesspool of flame-throwing and abuse. I've seen more rude postings in the last few weeks than I care to count. I most definitely believe that these kind of postings would not have even been composed if USENET were moderated. It seems as though on USENET it is permissible to be rude in a public place. I can't understand how this attitude manifested itself, but it is clearly to the detriment of the net. > Who wants to moderate talk.bizarre? Hopefully, the moderated USENET would have no need for a talk.bizarre in its current form. A moderated t.b could be just as fun and amusing, but without the bandwidth. I know the bizarroids will flame me for that comment, but I think it's time we decided what USENET is and what it should be. One thing I think it should *not* be is a global BBS. That means that we should get rid of "frivolous" newsgroups, or at least replace them with moderated equivalents. I'm also in the "binaries must go" camp. I think that *all* sources and binaries newsgroups (including comp.sources.unix) should be replaced by a comp.sources.announce type newsgroup, where announcements of new source releases could be made, including instructions on how to get them from an archive server. (perhaps uunet?) You may get the impression that I'm a hard-core technodweeb who only wants the technical comp.* groups and wants to see the rec, soc and talk groups vanish. Not so. I do think that many of the recreational groups would be far more enjoyable if they were moderated. I know I would subscribe to more of them if I knew I wouldn't have 70 unread postings awaiting me every morning, most of them throw-away. In fact, I'd like to moderate a rec group myself, if we did go 100% moderated. I'd like to do rec.sport.hockey. I really think that 100% moderated USENET is a way to solve many of our current problems: 1) much reduced bandwidth. This means that transmission charges will be less, disk requirements will reduce, and readers will be able to get a lot more out of USENET than they do now. 2) flames and abusive postings would be disallowed. They belong (and always have) in private mail, not in a public forum. 3) no more JJ type postings (and the resultant waste of bandwidth in beating the subject to death) I've said a lot, but I'm concerned at what USENET is degenerating into and I want to see it live a more healthy life. Comments? -- Andrew Scott andrew@teletron.uucp - or - {codas, ubc-cs, watmath, ..}!alberta!teletron!andrew
len@netsys.UUCP (Len Rose) (06/24/88)
In article <379@teletron.UUCP> andrew@teletron.UUCP (Andrew Scott) writes: [lots of intelligent remarks,which make sense] >I'm also in the "binaries must go" camp. I think that *all* sources and >binaries newsgroups (including comp.sources.unix) should be replaced by >a comp.sources.announce type newsgroup, where announcements of new source >releases could be made, including instructions on how to get them from an >archive server. (perhaps uunet?) This is what I have a problem with.. Surely not the SOURCE groups.. Source is what makes Usenet so worthwhile for most sites. Surely with the reduced traffic,source would not be a problem.. -- Len Rose - NetSys,Inc. 301-520-5677 len@ames.arc.nasa.gov or {ames,decuac,ihnp4}!netsys!len
sl@van-bc.UUCP (pri=-10 Stuart Lynne) (06/26/88)
In article <8659@netsys.UUCP> len@netsys.UUCP (Len Rose) writes: >In article <379@teletron.UUCP> andrew@teletron.UUCP (Andrew Scott) writes: > >[lots of intelligent remarks,which make sense] > >>I'm also in the "binaries must go" camp. I think that *all* sources and >>binaries newsgroups (including comp.sources.unix) should be replaced by >>a comp.sources.announce type newsgroup, where announcements of new source >>releases could be made, including instructions on how to get them from an >>archive server. (perhaps uunet?) > >This is what I have a problem with.. Surely not the SOURCE groups.. Source >is what makes Usenet so worthwhile for most sites. Surely with the reduced >traffic,source would not be a problem.. > Yes *SOURCES* too. But with a qualification. Before we get rid of either sources or binaries we must ensure that an adequate number of sites are setup to allow FTP or anonymous uucp downloading of them. I have have been laboriously save comp.sources.all for years, it fill's disks, tapes, it's hard to find things because it spread across several tapes, and the files all have cryptic names. At this point in time for most things I just look in uunet!~/ls-lR and download what I need. It's fast, cheap (via trailblazer), and I get to use all that hard disk space. Also with a central repository it's very easy for contributors to continually keep the files uptodate. A week ago I downloaded g++ sources for someone locally. Had a heck of a time until I grabbed another copy of ls-lR and found out that a new version of some of the files had arrived. While I don't advocate that everyone signup for uunet (at least until they get the Amdal installed :-) ) I do advocate that alternatives to mass distributions be found. For example how about a top level distribution called archive which ties together sources / binaries group moderators with a *small* number of archive sites. In other words use it to create a distributed sources / archive database. While each site might not carry everything in the archives, choosing to specialize in say amiga sources or xenix binaries; it should be possible to at least give the files the same name and position in the archive file system heirarchy. -- Stuart.Lynne@wimsey.bc.ca {ubc-cs,uunet}!van-bc!sl Vancouver,BC,604-937-7532
andrew@teletron.UUCP (Andrew Scott) (06/26/88)
In article <8659@netsys.UUCP>, len@netsys.UUCP (Len Rose) writes: > In article <379@teletron.UUCP> andrew@teletron.UUCP (Andrew Scott) writes: > >I'm also in the "binaries must go" camp. I think that *all* sources and > >binaries newsgroups (including comp.sources.unix) should be replaced by > >a comp.sources.announce type newsgroup, where announcements of new source > >releases could be made, including instructions on how to get them from an > >archive server. (perhaps uunet?) > > This is what I have a problem with.. Surely not the SOURCE groups.. Source > is what makes Usenet so worthwhile for most sites. Surely with the reduced > traffic,source would not be a problem.. The reason I included sources with binaries is that a lot of them are machine specific also. You'll get no argument from me that sources are probably the most valuable postings in all of USENET, but they're also large. For example, many Sun specific sources are posted to comp.sources.unix, such as the monster Postscript interpreter from last year. I'll bet that a good many sites just let it pass through, as they don't have Suns. With the addition of 386 machines and other high powered PCs to the net, not every source posting can be used at every site. Thus, it seems to make sense to archive them and let individual sites pick them up if they have use for them. Surely the costs will be cheaper overall than transmitting them through every site that carries comp.sources.unix. While we're at it, the comp.mail.maps postings are also immense. Perhaps future news software could have the auto-extraction built in as part of the code, and only post updates to the maps in a special form: %add site newsite.com %change link from_a to_b(DEAD) and so on. We wouldn't need to post the whole map for a region when a site makes a few small changes, nor would we have to post context diffs (which I believe work out to be as large as posting the whole things). It seems to me that a lot of sites are solving the problem of an increasingly larger USENET by buying faster modems and installing larger disks. Doesn't it make more sense to re-organize the software than having to resort to such brute-force methods? -- Andrew Scott andrew@teletron.uucp - or - {codas, ubc-cs, watmath, ..}!alberta!teletron!andrew
sl@van-bc.UUCP (pri=-10 Stuart Lynne) (06/27/88)
In article <383@teletron.UUCP> andrew@teletron.UUCP (Andrew Scott) writes: >In article <8659@netsys.UUCP>, len@netsys.UUCP (Len Rose) writes: >> In article <379@teletron.UUCP> andrew@teletron.UUCP (Andrew Scott) writes: >> >I'm also in the "binaries must go" camp. I think that *all* sources and >> >binaries newsgroups (including comp.sources.unix) should be replaced by >> >a comp.sources.announce type newsgroup, where announcements of new source >> >releases could be made, including instructions on how to get them from an >> >archive server. (perhaps uunet?) >The reason I included sources with binaries is that a lot of them are machine >specific also. You'll get no argument from me that sources are probably the >most valuable postings in all of USENET, but they're also large. >For example, many Sun specific sources are posted to comp.sources.unix, such >as the monster Postscript interpreter from last year. I'll bet that a good >many sites just let it pass through, as they don't have Suns. With the >addition of 386 machines and other high powered PCs to the net, not every >source posting can be used at every site. >Thus, it seems to make sense to archive them and let individual sites pick >them up if they have use for them. Surely the costs will be cheaper overall >than transmitting them through every site that carries comp.sources.unix. Not only that, but it is paid for *by the people who need them*! >It seems to me that a lot of sites are solving the problem of an increasingly >larger USENET by buying faster modems and installing larger disks. Doesn't >it make more sense to re-organize the software than having to resort to such >brute-force methods? Well there are worse ways. CPU cycles, RAM and hard disk bytes are *far* cheaper today than they where in the past. Are you advocating that we stick with yesterdays technology and dream up neat algorithms to allow us run XWindows (for example) on a Z80. The technology is there, use it. That's not to say that intelligent and creative thinking can't help. When I first brought up news on my Callan a year and a half ago my NewStone rating (average number of articles unbatched per minute) was about 2. And that was with only about a half a dozen sys file entries. I'm now up to about 10, with about two dozen sys file entries. Without any hardware changes. Just better system configuration. It kind of bugs me when people complain too much about the increased volume. I don't think the S/N ratio is too much worse than two years ago when I joined the net. The volume has roughly doubled in that time but the cost of CPU cycles, RAM and disk space has had a roughly corresponding drop. The software is more efficent now - especially if you run C news ( I don't but will convert later this year ). The cost of long distance communications has dropped - both in terms of $/min and increased efficency of using the available bandwidth (ie Trailblazers). Even though the news volume has doubled I suspect that sites who keep fairly current are probably spending very little more and possibly less now than they where two years ago. -- Stuart.Lynne@wimsey.bc.ca {ubc-cs,uunet}!van-bc!sl Vancouver,BC,604-937-7532
andrew@teletron.UUCP (Andrew Scott) (06/27/88)
In article <1835@van-bc.UUCP>, sl@van-bc.UUCP (pri=-10 Stuart Lynne) writes: > > It kind of bugs me when people complain too much about the increased volume. > I don't think the S/N ratio is too much worse than two years ago when I > joined the net. The volume has roughly doubled in that time but the cost of > CPU cycles, RAM and disk space has had a roughly corresponding drop. The > software is more efficent now - especially if you run C news ( I don't but > will convert later this year ). The cost of long distance communications > has dropped - both in terms of $/min and increased efficency of using the > available bandwidth (ie Trailblazers). It's not just a matter of computing horsepower as to why I think we should attempt to cut volume, although it is an important factor for many sites. The net is just too big to handle as a *reader*. You are correct in saying that the S/N ratio is about the same as it was two years ago, but the doubling of net bandwidth in that time means that twice as much N has to be waded through to get to the S. I simply don't have the time to read as much as I would like. Getting back to the original discussion, a fully moderated USENET would be easier to read because most of the N has been filtered out. I can't imagine that anybody objects to *that*, although I do understand some of the other objections to moderation. -- Andrew Scott andrew@teletron.uucp - or - {codas, ubc-cs, watmath, ..}!alberta!teletron!andrew