Three new INET groups coming
Michael C. Berch
mailing list communities, three new newsgroups in the inet distribution
will be created in the next few days. These groups are:
comp.soft-sys.nextstep The NeXTstep computing environment.
comp.unix.solaris Discussions about the Solaris operating system.
comp.os.msdos.4dos The 4DOS command processor for MS-DOS.
These groups are peered with mailing lists at indycms.iupui.edu
administered by John Harlan (JBHa...@indyvax.iupui.edu) and will be
gatewayed here at presto.ig.com, joining a number of other
globally-gatewayed groups (as well as the bionet.* groups on
genbank.bio.net down the hall). I expect there to be additional
groups added in the next few weeks.
Newgroups with distribution "inet" will issue in a day or two and will
come from me, with an Approved header by Erik Fair (fa...@apple.com).
Please create these groups locally when they arrive.
For the INET distribution,
--
Michael C. Berch
IntelliGenetics, Inc.
m...@presto.ig.com
Bill Bogstad
>As part of the continuing efforts to forge links between the Usenet and
>mailing list communities, three new newsgroups in the inet distribution
>will be created in the next few days. These groups are:
>
>comp.soft-sys.nextstep The NeXTstep computing environment.
>comp.unix.solaris Discussions about the Solaris operating system.
>comp.os.msdos.4dos The 4DOS command processor for MS-DOS.
These all sound like reasonable groups and I might vote for them if
they went through the normal procedures for creating USENET groups.
However, it's hard enough explaining to people what the difference is
between the main USENET hierarchies and hierarchies like alt, gnu, etc.
Allowing some people the privilege to create newsgroups when they want to
and telling others they must go through proper procedures is just too
confusing.
The situation with "inet" groups has been discussed over and over
and I thought the consensus was that it was a mistake and we should formally
create all of the current "inet" groups and do away with this practice.
I'll be at the USENIX conference next week and if there is a USENIX BOF will
bring this issue up again. If someone wants me to relay comments, please
email them to me. Discuss it in this newsgroup as well if you like. If
necessary, I will even volunteer to count the votes for the 60 odd
groups that will need to be "created".
As far as these 3 groups are concerned, I would prefer if you
hold off on creating them until after next week; but you certainly
have (IMHO) bad precedent for going ahead.
Bill Bogstad
bog...@cs.jhu.edu
For those of you (myself included) who weren't sure what groups are part of
the inet distribution, the following is gleaned from a recent Checkgroups
postings by Gene Spafford:
comp.ai.edu Applications of Artificial Intelligence to Education.
comp.ai.vision Artificial Intelligence Vision Research. (Moderated)
comp.dcom.lans.hyperchannel Hyperchannel networks within an IP network.
comp.editors Topics related to computerized text editing.
comp.edu.composition Writing instruction in computer-based classrooms.
comp.lang.asm370 Programming in IBM System/370 Assembly Language.
comp.lang.clu The CLU language & related topics.
comp.lang.forth.mac The CSI MacForth programming environment.
comp.lang.icon Topics related to the ICON programming language.
comp.lang.idl IDL (Interface Description Language) related topics.
comp.lang.lisp.franz The Franz Lisp programming language.
comp.lang.lisp.x The XLISP language system.
comp.lang.rexx The REXX command language.
comp.lang.scheme.c The Scheme language environment.
comp.lang.visual Visual programming languages.
comp.lsi.cad Electrical Computer Aided Design.
comp.mail.multi-media Multimedia Mail.
comp.music Applications of computers in music research.
comp.org.eff.news News from the Electronic Frontiers Foundation. (Moderated)
comp.org.eff.talk Discussion of EFF goals, strategies, etc.
comp.os.aos Topics related to Data General's AOS/VS.
comp.os.cpm.amethyst Discussion of Amethyst, CP/M-80 software package.
comp.os.rsts Topics related to the PDP-11 RSTS/E operating system.
comp.os.v The V distributed operating system from Stanford.
comp.periphs.printers Information on printers.
comp.protocols.iso.dev-environ The ISO Development Environment.
comp.protocols.iso.x400 X400 mail protocol discussions.
comp.protocols.iso.x400.gateway X400 mail gateway discussions. (Moderated)
comp.protocols.pcnet Topics related to PCNET (a personal computer network).
comp.protocols.snmp The Simple Network Management Protocol.
comp.protocols.tcp-ip.domains Topics related to Domain Style names.
comp.protocols.time.ntp The network time protocol.
comp.security.announce Announcements from the CERT about security. (Moderated)
comp.soft-sys.andrew The Andrew system from CMU.
comp.std.announce Announcements about standards activities. (Moderated)
comp.sys.cdc Control Data Corporation Computers (e.g., Cybers).
comp.sys.handhelds Handheld computers and programmable calculators.
comp.sys.intel.ipsc310 Anything related to Xenix on an Intel 310.
comp.sys.northstar Northstar microcomputer users.
comp.sys.super Supercomputers.
comp.sys.ti.explorer The Texas Instruments Explorer.
comp.sys.zenith Heath terminals and related Zenith products.
comp.terminals.bitgraph The BB&N BitGraph Terminal.
comp.terminals.tty5620 AT&T Dot Mapped Display Terminals (5620 and BLIT).
comp.theory Theoretical Computer Science.
comp.theory.cell-automata Discussion of all aspects of cellular automata.
comp.theory.dynamic-sys Ergodic Theory and Dynamical Systems.
comp.theory.self-org-sys Topics related to self-organization.
comp.unix.cray Cray computers and their operating systems.
comp.windows.x.announce X Consortium announcements. (Moderated)
comp.windows.x.motif The Motif GUI for the X Window System.
news.software.nntp The Network News Transfer Protocol.
rec.games.vectrex The Vectrex game system.
rec.mag.fsfnet A Science Fiction "fanzine." (Moderated)
sci.bio.technology Any topic relating to biotechnology.
sci.math.num-analysis Numerical Analysis.
sci.philosophy.meta Discussions within the scope of "MetaPhilosophy."
soc.culture.esperanto The neutral international language Esperanto.
Special DDN groups ...
ddn.mgt-bulletin The DDN Management Bulletin from NIC.DDN.MIL (Moderated)
ddn.newsletter The DDN Newsletter from NIC.DDN.MIL (Moderated)
Ran Atkinson
> The situation with "inet" groups has been discussed over and
> over and I thought the consensus was that it was a mistake and we
> should formally create all of the current "inet" groups and do away
> with this practice.
Your perception of the consensus is not correct. The consensus is
that the "inet" newsgroups are A Good Thing. The point of having INET
newsgroups is to make transport of mailing lists more efficient. It
does this very well.
INET newsgroups are not part of USENET and so no amount of
discussion of them in news.groups or news.config will change their
status. Carrying INET newsgroups is completely voluntary. If you
don't want to carry them, then don't. Please don't try to go out on
your own and tell the rest of us what to do.
Ran
atki...@itd.nrl.navy.mil
(USENET and Internet oldtimer
-- I lived on net 10 and was on USENET before The Great Renaming).
Bill Bogstad
>In article <1992Jun6.1...@blaze.cs.jhu.edu> bog...@rhombus.cs.jhu.edu (Bill Bogstad) writes:
>
>> The situation with "inet" groups has been discussed over and
>> over and I thought the consensus was that it was a mistake and we
>> should formally create all of the current "inet" groups and do away
>> with this practice.
>
> Your perception of the consensus is not correct. The consensus is
>that the "inet" newsgroups are A Good Thing. The point of having INET
>newsgroups is to make transport of mailing lists more efficient. It
>does this very well.
I have nothing against the current groups that are part of the INET
distribution. Nor am I suggesting that any of them be removed
(grandfathering). Transporting mailing lists via NNTP is a "Good Thing".
This very poorly publicized policy of mixing hierarchies and the ways in
which they are controlled is not. BTW, I've been flamed in email about my
misuse of the word "consensus" which if you look it up is most commonly
defined to mean unanimity. Obviously, there is no consensus since you and I
disagree. (Assumming of course, that we are part of the same group....)
> INET newsgroups are not part of USENET and so no amount of
>discussion of them in news.groups or news.config will change their
>status. Carrying INET newsgroups is completely voluntary. If you
>don't want to carry them, then don't. Please don't try to go out on
>your own and tell the rest of us what to do.
I already have email from the administrator of one public access
USENET site who didn't even know that something called an "INET"
distribution existed. His impression was that any groups under the standard
set of hierarchies were controlled by discussions in the news.* groups. He
uses the newsgroup file off of uunet and gets at least one of the inet
groups (comp.editors) on his system. It is hard enough to explain the
different ways in which different hierarchies work without combining the most
formalized and one of the least formalized ones. His suggestion was that if
NNTP sites want the ability to easily create groups there should be an
"inet" hieararchy. Given the problems that occurred during the creation of
the us/usa hierarchy this could cause its own problems involving
distribution leakage and distributions vs. hierarchies. Anyone with a
suggestion for a better name?
Unless I get strong opposition (like threatening to cut off my NNTP
USENET feed :-), I think that I will go through the RFD/CFV procedure for
each one of the INET groups that currently appears to reside in USENET. I
also want to stop creation of new INET groups. Suggestions on ways to do
this while allowing the more efficient distribution of mailing lists via NNTP
would be appreciated. Can anyone say anything about the BITNET "bit"
hiearchy?
Bill Bogstad
bog...@cs.jhu.edu
>Ran
>atki...@itd.nrl.navy.mil
>(USENET and Internet oldtimer
> -- I lived on net 10 and was on USENET before The Great Renaming).
Well, you have me half beat there. I never lived on net 10, but was
on USENET long before the renaming (around 82 or 83 actually). My first net
account was on a machine with a 9600 baud leased line to an IMP (84?). And
does/should any of this matter?
Frank Peters
mail hierarchy and put groups created specifically as mailing list
gateways under it. That hierarchy could have a quicker creation
procedure tailored to the issue of creating a mailing list gateway.
On the one hand it does seem confusing to have some groups in the
'core' hierarchies require a voting procedure while others are just
created with no such formal process.
On the other hand, it seems a tad excessive to require a whole formal
RFD/CFV process to create usenet links for already thriving mailing
lists. Many of the questions that a usenet vote are already intended
to answer (popularity, will it have readers) are already answered by
the existence of a successful list.
--
Frank Peters - UNIX Systems Programmer - Mississippi State University
Internet: f...@CC.MsState.Edu - Phone: (601)325-7030 - FAX: (601)325-8921
Joe Buck
>
>> The situation with "inet" groups has been discussed over and
>> over and I thought the consensus was that it was a mistake and we
>> should formally create all of the current "inet" groups and do away
>> with this practice.
>
> Your perception of the consensus is not correct. The consensus is
>that the "inet" newsgroups are A Good Thing. The point of having INET
>newsgroups is to make transport of mailing lists more efficient. It
>does this very well.
If there were a consensus in either direction, you two wouldn't be
arguing. There clearly is no consensus. In any case, the mechanism for
making an inet group from a mailing list is totally arbitrary; it is based
on the whim of one person, and that one person is increasingly out of
touch with the current state of Usenet.
INET groups greatly complicate the administration of news sites for those
not on the Internet because of the large amount of leakage. Unmoderated
INET groups don't work at all because there is no mechanism to add
Distribution: inet to the headers, since anyone can post.
Moderated INET groups can work, since every posted article goes through
the gateway.
> INET newsgroups are not part of USENET and so no amount of
>discussion of them in news.groups or news.config will change their
>status. Carrying INET newsgroups is completely voluntary. If you
>don't want to carry them, then don't. Please don't try to go out on
>your own and tell the rest of us what to do.
If a news administrator on the Internet doesn't want to carry INET groups,
how is this to be accomplished without a great deal of work? "!inet"
will not do the job, because most articles in unmoderated INET groups do
not have an inet distribution.
INET would have worked if a separate top-level hierarchy had been created
for them. It would also work, though less well, if all INET groups were
moderated. Unmoderated INET groups don't work at all. If you administer
a site that is part of the inet distribution, you are in no position to
comment on this, because the nasties, caused by the large numbers of
leaks, affect only those that aren't in inet.
--
Joe Buck jb...@ohm.berkeley.edu
Jamie Gritton
> Unless I get strong opposition (like threatening to cut off my NNTP
> USENET feed :-), I think that I will go through the RFD/CFV procedure for
> each one of the INET groups that currently appears to reside in USENET. I
> also want to stop creation of new INET groups.
One bit of opposition you may want to watch out for is the maintainers
of the lists that are gatewayed to these groups. If a list maintainer is
against the status change, that could seriously hamper your chances of
changing it (of couse I don't know why anyone would be opposed to it).
But as a precaution I would contact the individual maintainers and see
what they think.
Generally, I against the confusion of the inet groups. What stops J.
Random User from creating groups with names under the regular Usenet
hierarchies and claiming it's OK because they're really "jrandomuser"
groups? Granted Erik Fair isn't J. Random User, but the point remains.
BTW, the current RFD for comp.org.eff.* is just such a move. I started
it at the request of the list "owner". If you're against inet groups sharing
Usenet's namespace, you'll want to vote "yes" on both these groups.
> Well, you have me half beat there. I never lived on net 10, but was
> on USENET long before the renaming (around 82 or 83 actually). My first net
> account was on a machine with a 9600 baud leased line to an IMP (84?). And
> does/should any of this matter?
I've only been around a few years. I concur: so what? I can learn my
Usenet history from the old timers. They talk about it enough.
--
James Gritton - gri...@byu.edu - I disclaim
Ed Wilts
>
> If a news administrator on the Internet doesn't want to carry INET groups,
> how is this to be accomplished without a great deal of work? "!inet"
> will not do the job, because most articles in unmoderated INET groups do
> not have an inet distribution.
It's trivial. Simply process the monthly checkgroups posting without INET
groups. Your INET groups will go away...
> Joe Buck jb...@ohm.berkeley.edu
--
Ed Wilts, BC Systems Corp., 4000 Seymour Place, Victoria, B.C., Canada, V8X 4S8
EWi...@Galaxy.Gov.BC.CA | Ed.W...@BCSystems.Gov.BC.CA | (604) 389-3430
Joe Buck
>In article <10tom1...@agate.berkeley.edu>, jb...@forney.berkeley.edu (Joe Buck) writes:
>>
>> If a news administrator on the Internet doesn't want to carry INET groups,
>> how is this to be accomplished without a great deal of work? "!inet"
>> will not do the job, because most articles in unmoderated INET groups do
>> not have an inet distribution.
>
>It's trivial. Simply process the monthly checkgroups posting without INET
>groups. Your INET groups will go away...
No, they don't go away. The articles keep coming, and they appear in
junk. Since we're creating two or three new groups a week, it's difficult
for the news administrator to tell whether he/she missed the creation of
a valid group, or it's inet leakage. So he/she creates some of the groups
that seem to have heavy leakage traffic (e.g. comp.editors). Then, when
the next checkgroups message is run, the process starts all over again.
Meanwhile, the admin gets complaints from local users: why did
comp.editors go away? It had useful traffic.
In the end, this process repeats itself to the point where some inet
groups become de-facto net-wide groups, and the administrator can no
longer use the (non-INET) checkgroups message (it would drop groups that
your users want), and cannot even receive the INET checkgroups message,
since it is distributed with "Distribution: inet".
I'm writing all this from my personal experience at administering a
non-Internet Usenet site, in my previous stint in the "real world". If
you don't have the corresponding experience, you aren't qualified to
dispute what I'm telling you here. The fact that the creator of INET does
not have the experience of running a non-Internet site explains why he
himself hasn't noticed this problem.
This problem doesn't occur for moderated groups, since every posted
article has a Distribution: inet header so there isn't leakage. I
therefore have no objection to moderated INET groups. I'd like to see all
unmoderated INET groups either voted into the standard list, converted to
moderated (the "moderator" could simply be a mail reflector that makes
sure to attach a Distribution: header), or dropped.
--
Joe Buck jb...@ohm.berkeley.edu
Bob Snyder
>>>>> (Ran Atkinson) writes:
Ran> In article <1992Jun6.1...@blaze.cs.jhu.edu>
Ran> bog...@rhombus.cs.jhu.edu (Bill Bogstad) writes:
> The situation with "inet" groups has been discussed over and
> over and I thought the consensus was that it was a mistake and we
> should formally create all of the current "inet" groups and do away
> with this practice.
Ran> Your perception of the consensus is not correct. The consensus is
Ran> that the "inet" newsgroups are A Good Thing. The point of having INET
Ran> newsgroups is to make transport of mailing lists more efficient. It
Ran> does this very well.
Ran> INET newsgroups are not part of USENET and so no amount of
Ran> discussion of them in news.groups or news.config will change their
Ran> status. Carrying INET newsgroups is completely voluntary. If you
Ran> don't want to carry them, then don't. Please don't try to go out on
Ran> your own and tell the rest of us what to do.
I would have no problems with the INET groups if they were truely not part of
Usenet. But they are. They are list4d under the mainstream hierarchies,
and from what I've seen there is no difference between INET groups and
the other groups from the user's point of view other than one is voted upon
and the other is not.
INET should either be incorporated "officially" into the mainstream usenet,
and voted upon (and I doubt highly that any would fail a vote), or moved off
into a seperate hierarchy, like "inet.*."
Ran> (USENET and Internet oldtimer
Ran> -- I lived on net 10 and was on USENET before The Great Renaming).
I fail to see how this affect your position, one way or the other....
Bob
--
Bob Snyder rsn...@fergie.dnet.ge.com, or rsn...@atl.dnet.ge.com
Computing Services ^^^^^^^^^^^^^^^^^^^^^ if it bounces, try ^^^^^^^^
GE Aerospace, Advanced Technology Labs (609) 866-6672 FX: (609) 866-6397
Romain Kang
sites that would like to chime in. For the same reason, I've included
the bulk of the parent article at the bottom.]
As another former non-INET Usenet admin, let me add my voice to Joe's.
My feed was an Internet site, probably one of the best run university
computing organizations around. There was nothing practical they or I
could do to keep INET groups off of my site.
When I last checked around February of 1991, less than 10% of the
articles in INET groups had the correct Distribution: in them. The
rest spilled off onto unsuspecting UUCP sites. The lack of interest
from most of the Usenet community suggests widespread ignorance and
apathy.
For those who care, putting in a top level "inet" name for new INET
groups would make the lives of non-INET sites much easier, at little
cost to the sites that do want the INET groups. Or will the INET
boosters also be ignorant and apathetic?
Romain
Bill Bogstad
>On the other hand, it seems a tad excessive to require a whole formal
>RFD/CFV process to create usenet links for already thriving mailing
>lists. Many of the questions that a usenet vote are already intended
>to answer (popularity, will it have readers) are already answered by
>the existence of a successful list.
If you are talking about the current INET newsgroups, I agree that
having an RFD/CVF might be overkill; but hopefully it will silence any
complaints from sticklers for the guidelines (which I have a tendency to be
myself). If you are talking about creating new INET groups at the request of
people who run mailing lists; I disagree that an RFD/CVF is not required.
One of the things that the RFD/CVF does is to find the most appropriate name
for a newgroup (another one of my pet peeves). The name of a mailing list
may not in anyway correspond to normal naming conventions for USENET groups.
A group may already exist that carries similar traffic and a bidirectional
gateway to this already existing group might be more appropriate. The
RFD/CFV process allows the collective knowledge (ignorance?) of the net
input into these decisions. The existence of a thriving mailing list has
always been considered (by many people) a strong reason for creation of a
newsgroup; but it is not the only factor on which I base my vote.
Bill Bogstad
Bill Bogstad
>bog...@rhombus.cs.jhu.edu (Bill Bogstad) writes:
>
> One bit of opposition you may want to watch out for is the maintainers
>of the lists that are gatewayed to these groups. If a list maintainer is
>against the status change, that could seriously hamper your chances of
>changing it (of couse I don't know why anyone would be opposed to it).
>But as a precaution I would contact the individual maintainers and see
>what they think.
I would be very surprised if maintainers of the actual mailing lists
wanted to restrict them to machines which only receive the INET distribution.
Particularily since, as Joe Buck has stated, many of the postings in those
groups don't have inet distribution lines anyway and the maintainers of the
lists can't control whether or not a machine on the Internet forwards
messages to none Internet hosts. For example, I can't imagine why
comp.windows.x.announce is an inet group other then someone didn't want to
bother going through the RFD/CVF process to create it. I'm not proposing
that the moderation status of the current INET groups be changed, just that
they stop being INET groups and that we don't create any more groups with
this mechanism. In any case, I guess I'll dig up the address of the
maintainers of the lists and let them know what is going on.
This brought something else to mind relative to a previous posting
that claimed that INET groups aren't part of USENET. If INET groups aren't
part of USENET; and USENET goes through its standard process to create a
newsgroup in its namespace (which just so happens to have the same name as
an INET group); whose responsibility is it to make sure those two namespaces
don't leak into each other or conflict. Or in a somewhat related example,
there is an USENET group called comp.apps.spreadsheets. There is an INET
groups called comp.editors. I think that editors are just as important as
spreadsheet programs and want to RFD/CVF an USENET group called
comp.apps.editors. Should the existence of a non-USENET group with similar
subject matter in any way effect the creation of a USENET group? (In what
appears to most users to be the same namespace, but really isn't.) In the
case of alt.* groups, a thriving ALT group has actually been used as an
argument for creation of a USENET group. (Admittedly usually with the
assumption that the ALT group would wither away....)
Bill Bogstad
bog...@cs.jhu.edu
Ross Ridge
> The situation with "inet" groups has been discussed over and
> over and I thought the consensus was that it was a mistake and we
> should formally create all of the current "inet" groups and do away
> with this practice.
Ran Atkinson writes:
> INET newsgroups are not part of USENET and so no amount of
>discussion of them in news.groups or news.config will change their
>status. Carrying INET newsgroups is completely voluntary. If you
>don't want to carry them, then don't. Please don't try to go out on
>your own and tell the rest of us what to do.
If INET newsgroups aren't part of Usenet why are they using the Usenet
name space? If they aren't Usenet groups then no one should be
should be upset if send rmgroup messages for them, since they don't
exist here. The fact of the matter "INET" has very rudely ingored
all standard Usenet precidence and protocol and is then one telling
us what do. Would you find it acceptable if Fidonet started creating
there own Usenet groups? Would you mind if I started creating groups
for NANET? You know, I think I'll start by creating comp.os.msdos.4dos
for the NANET 4DOS conference. I hope nobody minds.
Ross Ridge
--
Ross Ridge - The Great HTMU l/ //
[OO][oo]
ro...@zooid.guild.org /()\/()/
uunet.ca!zooid!ross db //
Peter da Silva
>that the "inet" newsgroups are A Good Thing.
The consensus *where*? On news.groups the consensus is that the inet groups
*as currently implemented* are a bad thing. By using the same name space
as Usenet groups, they effectively prevent the formation of Usenet groups
while having no formal method for reorganizing them. The perrenial
comp.editors thread is a classic example of this.
>The point of having INET
>newsgroups is to make transport of mailing lists more efficient. It
>does this very well.
It would do this just as well in an "inet" hierarchy without messing up
the larger Usenet. Please stop and think of the effect of INET on Usenet
as well as the effect of Usenet on INET. Please consider either making
this move or abandoning the INET groups as a separate entity.
--
`-_-'
Have you hugged your wolf today? 'U`
Peter da Silva, Taronga Park BBS, Houston, TX +1 713 568 0480/1032
Eliot Lear
>The consensus *where*? On news.groups the consensus is that the inet groups
>*as currently implemented* are a bad thing.
Excuse me? There is never a consensus on news.groups. Even if there
appears to be one, there is a silent minority (as anyone who is silent
on news.groups is *definitely* a minority).
;-)
--
Eliot Lear
[le...@sgi.com]
Gerry Swetsky
>If there were a consensus in either direction, you two wouldn't be
>arguing. There clearly is no consensus.
So how does "inet" get away with this? As I stated before, what's
to keep me from making vpnet a sort of god of newsgroups and
creating them as I wish? I still think comp.embroidery is a good
idea. (NOT!)
>In any case, the mechanism for
>making an inet group from a mailing list is totally arbitrary; it is based
>on the whim of one person, and that one person is increasingly out of
>touch with the current state of Usenet.
And quite obviously has a rather large ego.
>If a news administrator on the Internet doesn't want to carry INET groups,
>how is this to be accomplished without a great deal of work? "!inet"
>will not do the job, because most articles in unmoderated INET groups do
>not have an inet distribution.
We use the "newsgroups" file from ftp.uu.net to determine which
newsgroups are valid and which aren't. However, I just checked the
file and it seems the groups created by "inet" are labelled okay by
uunet. Why?
>INET would have worked if a separate top-level hierarchy had been created
>for them.
Well why wasn't this done at the onset of this debacle?
--
============================================================================
| Help stamp out stupid .signature files! Gerry Swetsky WB9EBO |
| vpnet - Public access Unix and Usenet |
| Home (708)833-8122 vpnet (708)833-8126 lis...@vpnet.chi.il.us |
============================================================================
Gerry Swetsky
>It's trivial. Simply process the monthly checkgroups posting without INET
>groups. Your INET groups will go away...
Obviously this person knows nothing of the workings of uucp feeds.
Gerry Swetsky
> Your perception of the consensus is not correct. The consensus is
>that the "inet" newsgroups are A Good Thing. The point of having INET
>newsgroups is to make transport of mailing lists more efficient. It
>does this very well.
As far as I'm concerned, this is in violation of everything I've
learned about the creation of Usenet newsgroups.
> INET newsgroups are not part of USENET and so no amount of
>discussion of them in news.groups or news.config will change their
>status.
If they are under the same hierarchy as the Usenet newsgroups, how
do I tell them from Usenet newsgroups? Why not create these groups
under an unique hierarchy?
Gee, if you can do it, that means that I can too! I mean, what
makes you so all-fired high and powerful? I think I'll create
comp.embroidery.
>Carrying INET newsgroups is completely voluntary. If you
>don't want to carry them, then don't. Please don't try to go out on
>your own and tell the rest of us what to do.
Why the h*** not? You seem to be telling us what to do! I mean if
it looks like a Usenet newsgroup and it smells like a Usenet
newsgroup, then it must be a Usenet newsgroup! Give us a break
will you?
Bill Bogstad
>In article <26...@ra.nrl.navy.mil> atki...@itd.nrl.navy.mil (Ran Atkinson) writes:
>
>> Your perception of the consensus is not correct. The consensus is
>>that the "inet" newsgroups are A Good Thing. The point of having INET
>>newsgroups is to make transport of mailing lists more efficient. It
>>does this very well.
>
> As far as I'm concerned, this is in violation of everything I've
> learned about the creation of Usenet newsgroups.
> [suggestion of a creation of a comp.embroidery newsgroup.]
As the person who originally complained about the creation of new
INET groups, I would like to ask that people don't get too carried away
yet. I brought this issue up at the USENIX USENET BOF last night (many of
the "important" new administrators were there) and it seemed fairly clear
that the majority of them thought that the creation of new USENET groups
using the INET mechanisms was not a good idea. Some suggested that there
should be an easier way that a long standing mailing list become gatewayed
to a USENET group. I expect more discussion to occur at the NNTP and maybe
even some concrete proposals will come out of it. USENET talks about
problems for a long time (INET has been considered a problem for sometime)
until somebody finally gets worked up enought to do something about them.
As to why, this method of creating newsgroups ever was used, well
USENET has changed over the years and at one time it didn't seem like such a
bad idea to some people and others just weren't willing to put in the effort
to convince people that it was a bad idea. A relatively small group of
people is responsible for the work that has gone into both the news software
and the policies and guidelines that currently control the net. Sometimes
those people don't want to spend any more of their uncompensated time.
There are alot of good ideas out there, but not many people willing to put
in the considerable time to put them into place. If I actually have to go
through a RFD/CVF for the 60 or so INET groups, I'll feel like I've finally
done something to balance my debt to those individuals. But there are many
other areas that could use work, maybe you (generically) can think about
what you could do to make the net a better place. (Geez that sounds
syrupy...)
Bill Bogstad
-----
I speak for myself on USENET.
George William Herbert
the people involved (and read parts of what happened) so I'm going to take
a stab at explaining the genesis and reasons for INET.
There used to be groups net.*, which were all there was, which
was run by a small group (aka the Backbone Cabal) who decided after internal
discussions wether things like new groups would happen. They listened to
outsiders but weren't accountable to them.
net.* was replaced with the current namespace due to overwhelming
sensibility of doing things this way. At about the same time, the current
idea of holding a "vote" to determine wether a new newsgroup would be
implimented became the default way of doing things; the cabal more or less
faded into the background. But not quite.
You see, they were still the people who were actually running the
news services at the "backbone" news sites. Some of them rightly
felt that there were times that their good judgement could let something
happen quickly, easily, and without 6 weeks flame in news.groups (well,
the flames mostly came more recently 8-) (mostly). So, they reinstated
the Cabal in a slightly different form... the Inet distribution.
Inet is simply a superset (or perhaps nearly totally overlapping
set; are any "regular usenet" groups not carried at an "inet" site?) of
normal everyday Usenet. It has no effect on normal operation of Usenet,
except in the (rare) case of namespace overlap/disagreement, which has
happened so rarely as to be only describable on a case-by-case basis.
Inet provides the "inet people" a way to do sensible things
without bothering to ask the whole world first. IMHO, as long as it's
clear that Inet doesn't dictate to sites off its own distribution
(which is clearly the attitude of those I've communicated with)
and the people doing "inet" things don't cause problems, there
is no problem. Yes, it violates the "pure democratic spirit"
of the current elections method, but only if you consider "usenet"
all-encompassing and exclusive.
Is Inet necessary? No. Is it useful? Yes. Is it bad? The
previous two answers were easy, this one is the argument.
I can see a network flexible enough for both voting and inet, as long
as everyone is sensible about it. Which they have been, more or less
uniformly.
I would think that more things like Inet are going to pop up
in the near future, rather than less. Perhaps seperating them
with namespace rather than distribution is healthy, for eventually
someone will cause a problem with one, but that doesn't mean that
the existing Inet model or people are bad.
-george william herbert
g...@lurnix.com g...@soda.berkeley.edu g...@ocf.berkeley.edu g...@gnu.ai.mit.edu
Bob Snyder
>>>>> g...@soda.berkeley.edu (George William Herbert) writes:
George> Is Inet necessary? No. Is it useful? Yes. Is it
George> bad? The previous two answers were easy, this one is the
George> argument. I can see a network flexible enough for both voting
George> and inet, as long as everyone is sensible about it. Which
George> they have been, more or less uniformly.
George> I would think that more things like Inet are going to
George> pop up in the near future, rather than less. Perhaps
George> seperating them with namespace rather than distribution is
George> healthy, for eventually someone will cause a problem with one,
George> but that doesn't mean that the existing Inet model or people
George> are bad.
I'm definately not arguing against the usefulness of the inet groups. I
think the groups on the most part are good, and should exist. My qualm is
with where. The Inet people claim they are not a part of Usenet. Well, they
are putting their groups into the mainstream hierarchies, and that, in my
view, is part of Usenet.
The guidelines state that the creation of groups under the namespace
of comp, news, sci, misc, soc, talk, and rec must follow the voting
procedures. Inet is sticking groups in here without going through
this procedure. This is a "bad thing."
Inet is depending on the part of news that is perhaps the most broken,
the Distribution line. "Inet messages should only go out with Inet
distribution." Well, most sites will pass on traffic they recieve
like this, whether or not the distribution line says they should. And
post people who post to inet groups don't have a clue that they are
doing this, much less that they should limit distribution.... Staying
off the Inet distribution is painfully hard, considering that most
people pass comp.* to their neighbors, and if I don't have the inet
groups created, I suddenly get huge amounts of "junk" arriving daily.
The answer? Move the inet groups to a new top-level domain. inet.*.
You can control your own top level domain, and set up your own rules
for creating new newsgroups, and no one will dispute your right to do
so. Keep trying to create groups under the mainstream hierarchies,
and you will have a fight every time. (Not that Mr. Berch seems to
notice, or even care, that most people commenting on the matter think
it's a really bad idea)
The current groups are good, and should be voted upon or
grandfathered, but please, no more new Inet groups under the
mainstream namespace.
Lazlo Nibble
> Inet provides the "inet people" a way to do sensible things without
> bothering to ask the whole world first. IMHO, as long as it's clear
> that Inet doesn't dictate to sites off its own distribution (which is
> clearly the attitude of those I've communicated with) and the people
> doing "inet" things don't cause problems, there is no problem.
Nobody here is likely to dispute the right of the inet people (and anyone
else, for that matter) to do what they want with their own machines, but
distributions leak, and this *does* cause problems for people outside the
inet distribution.
Given this problem, there are two possible ways out: fix the leaky
distributions or find a more reliable way to control distribution of the
inet groups. The former solution I think most people will admit is a
worthy but unlikely goal; the latter would work fine if a new 'inet.'
hierarchy was created. Aside from the work it would take to implement a
new hierarchy, I don't see that this is any worse a way of handling the
inet groups than they way that they're handled now. It would certainly
make life easier for a lot of admins.
--
Lazlo (la...@triton.unm.edu)
George William Herbert
Once the only way to create a newsgroup was for the group of
reasonable news admins to just do it.
With Inet, we (*) still have that capability.
Explicitly writing that out of Usenet, while at the same time
admitting that those news admins are the ones who are responsible for/
in charge of Usenet itself, seems like a pointless thing to do.
Until the net.democracy solidifies considerably, Inet will serve as
a reasonable safety valve for Usenet as a whole.
I don't think it's time yet for Inet to stop. I think that some
agreement clarifying that a properly held vote could overrule any
conflicts that might arise is a good idea, to show that there's some
movement away from Inet's reasons for being, but stopping it right now
is probably premature.
But that's just my opinion. I trust the people, really... 8-)
(*) We in this case is referring to netnews as a whole, in the general
sense, not the Inet people ... I'm not one of them, though I did once
offer to take some of ucbvax off Erik Fair's hands 8-) I am not and
don't speak for the people running Inet.
-george william herbert
g...@lurnix.com g...@soda.berkeley.edu g...@ocf.berkeley.edu g...@gnu.ai.mit.edu
"There is no such thing as a good organization, Grasshopper.
There are only good people. If you think you have a good
organization, you have good people and don't know it.
If you think you have good people, you're lucky."
-maniac
Gerry Swetsky
>Nobody here is likely to dispute the right of the inet people (and anyone
>else, for that matter) to do what they want with their own machines, but
>distributions leak, and this *does* cause problems for people outside the
>inet distribution.
Just checked comp.editors in our spool directory. Only one
article had a Distribution line and it said "na". So much for
distributions.
>the latter would work fine if a new 'inet.'
>hierarchy was created.
But the INET powers have already stated here point blank that they
refuse to listen to any suggestions we have! How can we now make
suggestions?
Don Lane
>
>> Inet provides the "inet people" a way to do sensible things without
>> bothering to ask the whole world first. IMHO, as long as it's clear
>> that Inet doesn't dictate to sites off its own distribution (which is
>> clearly the attitude of those I've communicated with) and the people
>> doing "inet" things don't cause problems, there is no problem.
>
I've been made aware of the "inet people" and the newsgroups that are part
of this "net". Just what newsgroups are part and parcel of "inet" that
are included in the group of seven. I'm all in favour of the grandfather
clause, but let's face it grandfather, "times are a-changing".
>Nobody here is likely to dispute the right of the inet people (and anyone
>else, for that matter) to do what they want with their own machines, but
>distributions leak, and this *does* cause problems for people outside the
>inet distribution.
>
What distributions? As far as I can tell there is no distinction between
the "inet" newsgroups and the normal "UsenNet newsgroups.
>Given this problem, there are two possible ways out: fix the leaky
>distributions or find a more reliable way to control distribution of the
>inet groups. The former solution I think most people will admit is a
>worthy but unlikely goal; the latter would work fine if a new 'inet.'
>hierarchy was created. Aside from the work it would take to implement a
>new hierarchy, I don't see that this is any worse a way of handling the
>inet groups than they way that they're handled now. It would certainly
>make life easier for a lot of admins.
>
This would be the answer... create a "inet" hierarchy and abide by the
newsgroup creation policy? of Usenet.
>--
>Lazlo (la...@triton.unm.edu)
--
Don Lane, System Administrator
Pacific Information Systems UUCP: ... {softwords.bc.ca}!infopac!dlane
P.O. Box 5952, Stn B, Victoria, BC, Canada V8R 6S8
Jamie Gritton
> But the INET powers have already stated here point blank that they
> refuse to listen to any suggestions we have! How can we now make
> suggestions?
Make suggestions, but not to the INET powers. I suggest, for
example, that news admins not accept these three new inet groups until
they get their own hierarchy. I also appluad the idea that Bill
Bogstad had of running an RFD/CFV on all of the existing inet groups.
Peter da Silva
> But the INET powers have already stated here point blank that they
> refuse to listen to any suggestions we have! How can we now make
> suggestions?
Well, we could create a "UUCP" distribution... taking the same care that
the "INET" people have to make sure it doesn't leak... and create a few
groups without discussing it with them, or in news.groups. Start with a
"news.uucp" group for configuration discussions. Make one of the rules for
voting in "UUCP" groups being that you not be in the internet. See how
they like it...
Michael C. Berch
> > But the INET powers have already stated here point blank that they
> > refuse to listen to any suggestions we have! How can we now make
> > suggestions?
>
> Make suggestions, but not to the INET powers. I suggest, for
> example, that news admins not accept these three new inet groups until
> they get their own hierarchy. I also appluad the idea that Bill
> Bogstad had of running an RFD/CFV on all of the existing inet groups.
No one is *asking* you to accept the new inet groups unless you are
already part of the inet distribution. In the latter case you are
already getting 60+ other inet groups, and are rather strongly
benefiting from the existence of inet (whether you know it or not).
As for RFD/CFV for the existing groups, feel free -- there is no reason
why these groups cannot be shared with mainstream Usenet, and those
that enjoy sufficient popularity can easily be "promoted" if desired.
--
Michael C. Berch
m...@presto.ig.com
Michael C. Berch
> > But the INET powers have already stated here point blank that they
> > refuse to listen to any suggestions we have! How can we now make
> > suggestions?
>
> Well, we could create a "UUCP" distribution... taking the same care that
> the "INET" people have to make sure it doesn't leak... and create a few
> groups without discussing it with them, or in news.groups. Start with a
> "news.uucp" group for configuration discussions. Make one of the rules for
> voting in "UUCP" groups being that you not be in the internet. See how
> they like it...
The above is undoubtedly for rhetorical value only, but by all means, feel
free to create such a distribution if you think that it is needed or
desirable. You will not hear an objection from me.
Gerry Swetsky
>
>No one is *asking* you to accept the new inet groups unless you are
>already part of the inet distribution.
*WHAT* INET distribution?
I told you they weren't listening, folks! We want ONE Usenet, not
two!
Rich Salz
> But the INET powers have already stated here point blank that they
> refuse to listen to any suggestions we have!
No "INET power" has even posted any response on the matter here.
The consensus at the two BOF's held during Usenix seemed to be:
One vote, merge all current inet groups into standard namespace.
Second question, what to do for the future? I think it is reasonable
to follow the standard rules. Erik Fair says no. I believe Erik is
out of touch, and unlikely to get back in touch. The war is over, Unix
and (more importantly) netnews has won. I think we should grandfather
in the current groups, and make new groups follow the procedures.
There is an info hierarchy.
/r$
LCDR Michael E. Dobson
only these sites didn't know they were "special". Without all articles in
the "INET" groups containing a "Distribution: inet" header there was no
practical way to prevent that from happening. Fortunately the newgroup
messages were distributed with that header so systems using explicit
distributions in their sys files will not propagate the newgroup messages
to non-inet sites but what about subsequent articles?
--
LCDR Mike Dobson, Sys Admin for | Internet: rd...@nmrdc1.nmrdc.nnmc.navy.mil
nmrdc1.nmrdc.nnmc.navy.mil | UUCP: ...uunet!mimsy!nmrdc1!rdc30
AT&T 3B2/600G Sys V R 3.2.3 | BITNET: dobson@usuhsb or nrd0mxd@vmnmdsc
WIN/TCP for 3B2 R3.2 | MCI-Mail: 377-2719 or 00037...@mcimail.com
Eliot Lear
>The guidelines state that the creation of groups under the namespace
>of comp, news, sci, misc, soc, talk, and rec must follow the voting
>procedures. Inet is sticking groups in here without going through
>this procedure. This is a "bad thing."
Inet predates the guidelines by many years. Listen, kids. So far I
think we've seen about four new inet groups over the past two years.
This is not a mechanism that's being abused. Why fuss?
--
Eliot Lear
[le...@sgi.com]
Willard Dawson
Which must surely mean that M.C. Berch is on the internet (sic). Of
course you wouldn't hear from him! It would be technically impossible,
unless some leakage did occur.
I happen to think it is a neat idea. I grow tired of hearing how the
dismemberment of the NFSnet et al would lead to the destruction of Usenet.
It would be difficult, however, to prevent crossover from happening.
--
Willard Dawson, BellSouth Advanced Networks, 1100 Johnson Ferry Road, Ste. 900
Atlanta, GA 30342, Voice: +1 404 303 2343, UUCP: gatech!vdbsan!willard!dawson
Internet -> X.400: /G=W/S=DAWSON/O=BSAN/ADMD=BELLSOUTH/C=US/@sprint.com
Standard disclaimer applies.
Bob Snyder
>>>>> Salz) writes:
Rich> The consensus at the two BOF's held during Usenix seemed to be:
Rich> One vote, merge all current inet groups into standard namespace.
Rich> Second question, what to do for the future? I think it is
Rich> reasonable to follow the standard rules. Erik Fair says no. I
Rich> believe Erik is out of touch, and unlikely to get back in touch.
Rich> The war is over, Unix and (more importantly) netnews has won. I
Rich> think we should grandfather in the current groups, and make new
Rich> groups follow the procedures.
So what is it to be then? A CFV for the grandfathering of current INET
groups and prohibiting of future non-voted upon groups? I think this would
make the most sense...
Eliot Lear
>So what is it to be then? A CFV for the grandfathering of current INET
>groups and prohibiting of future non-voted upon groups? I think this would
>make the most sense...
Can we please not jump off of any bridges (yet)?
BY ITSELF a vote will not and shouldn't affect INET. Since INET has
never been subject the votes, guidelines, etc, and in fact predates
any of the current policies, a vote means very little to those who
carry the distribution. In order to shut down INET, people have to
stop accepting both the distribution and the groups involved. And
they have to stop honoring the newgroups. I for one will not do that,
since Erik and Michael have shown great restraint and wisdom in the
past, whereas the popular vote has often yielded stunningly bad
results.
I also think it's important to point out one problem with carrying any
newsgroup (yes, it's been pointed out many times before). There are
newsgroups that exist in more than one form in more than one
hierarchy. It's important that such newsgroups not be joined commonly
to one mailing list, or else what article gets posted to what
newsgroup can be somewhat random. Someone who is an INET meister can
prevent this class of problem, and this roll is perfectly suited for
the current INET charter, since it consists only of newsgroups that
are gatewayed to mailing lists.
Now, I say again: unless you think something is *really* broken, let's
not fuss.
--
Eliot Lear
[le...@sgi.com]
Peter da Silva
>The above is undoubtedly for rhetorical value only, but by all means, feel
>free to create such a distribution if you think that it is needed or
>desirable. You will not hear an objection from me.
OK, I think I'll start by issuing global rmgroups for all "INET" groups,
taking the same care that the "INET" people do to make sure they don't
leak out of the "UUCP" distribution.
Bob Snyder
>>>>> le...@yeager.corp.sgi.com (Eliot Lear) writes:
Eliot> Can we please not jump off of any bridges (yet)?
Certainly. I try not to be completely unreasonable. However, when no one is
willing to take up the cause of INET to argue for it's existence when there
is fairly strong opposition to it...
Eliot> BY ITSELF a vote will not and shouldn't affect INET. Since
Eliot> INET has never been subject the votes, guidelines, etc, and in
Eliot> fact predates any of the current policies, a vote means very
Eliot> little to those who carry the distribution. In order to shut
But most people, I think, aren't aware that they are carrying the
distribution. What I've seen (and experienced) is that people assume the
current INET groups were created under the guidelines. All you ever hear
of INET normally is the checkgroup message, which comes in 2 different
flavors: with, and without.
I also challenge the concept of an INET-only distribution. The
Distribution: header is broken, has been broken, and probably will not
be fixed for some time to come, if ever. Even if it did work, no one
on unmoderated groups actually gives their articles an "INET"
distribution. Since many sites simply have an entire hierarchy
enabled (all of comp.*, and so on), and pass that along, even if I do not
honor the newgroup, I will get pounded with it's traffic and my news software
will complain regularly until I create the groups.
Eliot> down INET, people have to stop accepting both the distribution
Eliot> and the groups involved. And they have to stop honoring the
Eliot> newgroups. I for one will not do that, since Erik and Michael
Eliot> have shown great restraint and wisdom in the past, whereas the
Eliot> popular vote has often yielded stunningly bad results.
I'm a firm believer in the popular vote; I disagree that a small number of
news admins know what is best for the entirety of Usenet. I don't want to
get rid of INET. I think the groups created are, from what I've seen,
appropriate and right. What I object to is their existance intermixed with
the other groups of usenet. I would love to see, and would carry, an inet.*
hierarchy. It is then obvious that the rules of "normal" usenet don't
apply.
Eliot> I also think it's important to point out one problem with carrying any
Eliot> newsgroup (yes, it's been pointed out many times before). There are
Eliot> newsgroups that exist in more than one form in more than one
Eliot> hierarchy. It's important that such newsgroups not be joined commonly
Eliot> to one mailing list, or else what article gets posted to what
Eliot> newsgroup can be somewhat random. Someone who is an INET meister can
Eliot> prevent this class of problem, and this roll is perfectly suited for
Eliot> the current INET charter, since it consists only of newsgroups that
Eliot> are gatewayed to mailing lists.
I'm not sure I'm understand this; are you worried about mailing lists being
routed onto multiple newsgroups? If so, I would think the list maintainer
would control this, and I don't see how the existance of INET solves this.
And again, I'm not calling for the disbanding of INET. What I think should
happen is the current groups be grandfathered but any new groups not
be placed in the mainstream hierarchy.
Eliot> Now, I say again: unless you think something is *really*
Eliot> broken, let's not fuss.
I think enough is brok*
Eliot> broken, let's not fuss.
I think enough is broken to warrant fixing.
Scott Hazen Mueller
It's rather badly broken:
wizard% cd ~news/comp/unix/solaris
wizard% ls
10 12 3 5 7 9
11 2 4 6 8
wizard% grep -i distribution *
12:Distribution: usa
12 articles in one of the 3 new inet groups, on an Internet site (not Zorch,
but my workplace), and what are the distributions?
There is no distinction between INET and Usenet at an INET site, and no
mechanism whatsoever for enforcing the 'proper' Distribution header. The
most reasonable resolution is for Usenet to absorb the current INET groups,
and for new INET groups (if any) to either live in a separate hierarchy or
to pass a Usenet vote.
--
Scott Hazen Mueller sc...@zorch.SF-Bay.ORG or (ames|pyramid|vsi1)!zorch!scott
Moderator, ba.announce; submissions to ba-an...@Zorch.SF-Bay.ORG.
Moderator, rec.arts.sf.announce; submissions to sf-an...@Zorch.SF-Bay.ORG.
SF-Bay Public-Access Unix 408-254-0246/1032/9882 login newuser password public
Eliot Lear
>Certainly. I try not to be completely unreasonable. However, when no one is
>willing to take up the cause of INET to argue for it's existence when there
>is fairly strong opposition to it...
First of all, lack of comment in this forum does not imply anything
other than perhaps a lack of interest in the topic. Second, I do
believe that Michael and I are speaking out for INET.
>But most people, I think, aren't aware that they are carrying the
>distribution. What I've seen (and experienced) is that people assume the
>current INET groups were created under the guidelines. All you ever hear
>of INET normally is the checkgroup message, which comes in 2 different
>flavors: with, and without.
Generic comment: I'm really beginning to think that a Distribution
line should be manditory, and that there be associated with any
newsgroup a default distribution. A lot of this stuff could be fixed
in the posting agents. I agree, though, that a fix would take a very
long time to get distributed.
>I'm a firm believer in the popular vote; I disagree that a small number of
>news admins know what is best for the entirety of Usenet.
Then don't you think we ought to increase the number of people
required to show interest in a newsgroup, given that there are, oh
some estimated > 1,000,000 readers on the net?
>I think the groups created are, from what I've seen,
>appropriate and right. What I object to is their existance intermixed with
>the other groups of usenet. I would love to see, and would carry, an inet.*
>hierarchy. It is then obvious that the rules of "normal" usenet don't
>apply.
How about we create a woods-guidelines.* hierarchy, since inet was
here first? Alright, I'm mostly kidding. But we are now bickering
about who owns the namespace, and I'd like to avoid that.
>I'm not sure I'm understand this; are you worried about mailing lists being
>routed onto multiple newsgroups? If so, I would think the list maintainer
>would control this, and I don't see how the existance of INET solves this.
In many cases there is little or no intervention from the list
maintainer, thanks to automated list servers like LISTSERV. In
getting people to use netnews, it's important to recognize useful
mailing lists and convert them (spread the gospel).
>I think enough is broken to warrant fixing.
Then how about this compromise. Let's let Erik use his discretion
when creating groups that would normally go in the Inet distribution.
Let's request that he create the groups with normal distribution.
Until Erik shows poor judgement, let's not fuss about the
popular/cabal'o'one argument.
--
Eliot Lear
[le...@sgi.com]
Neil Rickert
>
>Generic comment: I'm really beginning to think that a Distribution
>line should be manditory, and that there be associated with any
>newsgroup a default distribution.
For all the griping, the real reason "Distribution:" is getting such
criticism is that groups not in the active file are still propagated
through junk. If unsubscribed groups were not propagated, many fewer
articles would leak very far beyond their intended range.
What we really need is a new flag in the 'sys' file entry to control
whether junked articles are forwarded to a neighbor.
Tom Limoncelli
>Certainly. I try not to be completely unreasonable. However, when no one is
>willing to take up the cause of INET to argue for it's existence when there
>is fairly strong opposition to it...
Of the people that "strongly oppose" the INET distribution, I have
found that 40% of them liked it once someone explained the historical
events that lead to it. The other 60% didn't like it because of the
leaks. Leaks of ANY distribution are the fault the person at YOUR site
that gave wrong information to the person at your ajacent cites.
I don't think INET should be grandfathered into the mainstream. I
think people running NNTP over the internet should vote on if we really
want to grandfather any of those so-called "mainstream" newsgroups. I
mean, the Official Usenet Board of Directors has never petitioned the
NNTP Grand Review Board about this. Heck, has Kibo, Supreme Ruler of
HappyNet ever even offered an opinion?
Tom
(who accepts INET at this internet site, and on his home UUCP-only site)
--
Tom Limoncelli -- t...@warren.mentorg.com (work) -- t...@plts.uucp (play)
"Tax cuts don't do any good for someone who is out of work," Gov. Bruce
Sundlun, D-RI, said in denouncing Bush's economic rescue plan as woefully
inadequate. "They don't need a tax cut. They need a job." Feb 3, 1992
Joe Buck
>Of the people that "strongly oppose" the INET distribution, I have
>found that 40% of them liked it once someone explained the historical
>events that lead to it. The other 60% didn't like it because of the
>leaks. Leaks of ANY distribution are the fault the person at YOUR site
>that gave wrong information to the person at your ajacent cites.
Oh, rot. Those who administer or have administered sites on the UUCP-net
are pretty much unanimously opposed. In any case, while there is an
inet distribution, the inet groups are constructed in such a way as
to be almost impossible to stop leaks.
I claim that everyone who has direct experience of the problems associated
with inet is opposed to its current implementation. Why? Site
administrators naturally assume that "!inet" will prevent transmission of
inet groups. It doesn't, because 95%+ of articles in the inet groups do
not have Distribution: inet.
OK, you say, every Internet site that feeds any UUCP site could add
60 different "!group" strings to each sys line, and fix that sys line
every time Erik decides to make a new inet group. If anyone anywhere
on the net fails to do this, then we have Tom Limoncelli's word that
it is totally the fault of some person that gave wrong information.
Right.
You, for example, don't know what INET is, since you've called it a
"distribution". It isn't a distribution; "na" is a distribution. INET is
a set of groups.
NOTE THE DISTINCTION: I am opposed to INET's current implementation.
I am not opposed to the concept of a set of groups that Erik Fair or
anyone else gateways into a set of newsgroups. I simply assert that
some change needs to be made. Any of the following would do:
1. Construct some mechanism that forces a "Distribution: inet" header
onto every article. The easiest way to do this is to make every group
moderated.
2. Move all INET groups into the standard hierarchy.
3. (A mix of the above). Move only the unmoderated INET groups into the
standard hierarchy.
4. Rename the groups to begin with the prefix "inet.".
I find the status quo, where the same hierarchies have two different
mechanisms for adding groups with an extremely faulty method of separating
the two, to be unacceptable.
--
Joe Buck jb...@ohm.berkeley.edu
Peter da Silva
>Inet predates the guidelines by many years. Listen, kids. So far I
>think we've seen about four new inet groups over the past two years.
>This is not a mechanism that's being abused. Why fuss?
Because the total volume of the net is getting to the point where splitting
at least some of the inet groups is becoming a necessity... and we can't do
it because they're inet groups.
Bob Sloane
le...@yeager.corp.sgi.com (Eliot Lear) writes:
> Then how about this compromise. Let's let Erik use his discretion
> when creating groups that would normally go in the Inet distribution.
> Let's request that he create the groups with normal distribution.
> Until Erik shows poor judgement, let's not fuss about the
> popular/cabal'o'one argument.
What does Erik have to do with this discussion? The groups were
announced and created by m...@presto.ig.com (Michael C. Berch) not Erik
Fair. Does this mean that anyone can create an inet group? If so, I
have a couple of mailing lists that I could create groups for. If not,
then who can create them? If I create an inet group, and someone else
thinks it "shows poor judgement" then will all inet group creations be
stopped, or just mine? This whole issue seems to be leading us more
toward the scheme used in alt for creating group, ie, everyone for
themselves. Is that what you really want?
--
USmail: Bob Sloane, University of Kansas Computer Center, Lawrence, KS, 66045
E-mail: slo...@kuhub.cc.ukans.edu, slo...@ukanvax.bitnet, AT&T: (913)864-0444
Bob Snyder
>>>>> le...@yeager.corp.sgi.com (Eliot Lear) writes:
Eliot> First of all, lack of comment in this forum does not imply anything
Eliot> other than perhaps a lack of interest in the topic. Second, I do
Eliot> believe that Michael and I are speaking out for INET.
The only lack of comment I saw was for those supporting the status quo.
Nobody was arguing for it (until recently); and the general opinion seemed
against the status quo. Recently both you and Michael have strated speaking
for it, so...
[...]
Eliot> Generic comment: I'm really beginning to think that a Distribution
Eliot> line should be manditory, and that there be associated with any
Eliot> newsgroup a default distribution. A lot of this stuff could be fixed
Eliot> in the posting agents. I agree, though, that a fix would take a very
Eliot> long time to get distributed.
I agree Distribution needs to be fixed. If it worked, that would satisfy
a lot of the problems I have with things as they are. There still would be
a namespace question, but...
>I'm a firm believer in the popular vote; I disagree that a small number of
>news admins know what is best for the entirety of Usenet.
Eliot> Then don't you think we ought to increase the number of people
Eliot> required to show interest in a newsgroup, given that there are, oh
Eliot> some estimated > 1,000,000 readers on the net?
I don't see what this has to do with the INET situation, but I think it
deserves consideration. I've been watching that discussion take
place, but haven't gotten involved in it.
>I'm not sure I'm understand this; are you worried about mailing lists being
>routed onto multiple newsgroups? If so, I would think the list maintainer
>would control this, and I don't see how the existance of INET solves this.
Eliot> In many cases there is little or no intervention from the list
Eliot> maintainer, thanks to automated list servers like LISTSERV. In
Eliot> getting people to use netnews, it's important to recognize useful
Eliot> mailing lists and convert them (spread the gospel).
bit.* seems to be thriving fairly well... I'm not argueing about if
INET should or shouldn't exist; I think that it deserves and needs
it's own namespace, seperate from the mainstream "voting" usenet.
>I think enough is broken to warrant fixing.
Eliot> Then how about this compromise. Let's let Erik use his discretion
Eliot> when creating groups that would normally go in the Inet distribution.
Eliot> Let's request that he create the groups with normal distribution.
Eliot> Until Erik shows poor judgement, let's not fuss about the
Eliot> popular/cabal'o'one argument.
That's not a compromise, that's the status quo. :-)
The Distribution is broken, so distributing with or without
Distribution: INET has no effect. This is like saying, "These rules
apply to you, unless your name is Erik or Michael." I'm not arguing
Erik's judgement; I just don't think he should be exempt from what the
rest of usenet has adopted as their guidelines.
Eliot Lear
>What does Erik have to do with this discussion? The groups were
>announced and created by m...@presto.ig.com (Michael C. Berch) not Erik
>Fair.
Michael is acting on Erik's authority with his knowledge and consent.
> Does this mean that anyone can create an inet group?
No.
--
Eliot Lear
[le...@sgi.com]
Michael C. Berch
> >The above is undoubtedly for rhetorical value only, but by all means, feel
> >free to create such a distribution if you think that it is needed or
> >desirable. You will not hear an objection from me.
>
> OK, I think I'll start by issuing global rmgroups for all "INET" groups,
> taking the same care that the "INET" people do to make sure they don't
> leak out of the "UUCP" distribution.
But inet control messages do not, and have never, leaked. Every inet
newgroup message goes out with Distribution: inet. Inet control
messages are only issued by two people. If there is a
site redistributing messages with munged or deleted Distribution
headers, please let me know and I will go into major Net Police mode.
"Leakage" of the inet distribution is limited to non-control messages,
and is easily dealt with on a per-site basis.
It would be improper for you to issue "global rmgroups" on inet groups
(assuming you mean "without Distribution: uucp" on them) just as it
would be improper for me to issue global rmgroups (or newgroups) for
global newsgroups (or "uucp" groups, for that matter).
Tom Limoncelli
>In article <1992Jun16.1...@Warren.MENTORG.COM> t...@Warren.MENTORG.COM (Tom Limoncelli) writes:
>>Of the people that "strongly oppose" the INET distribution, I have
>>found that 40% of them liked it once someone explained the historical
>>events that lead to it. The other 60% didn't like it because of the
>>leaks. Leaks of ANY distribution are the fault the person at YOUR site
>>that gave wrong information to the person at your ajacent cites.
>You, for example, don't know what INET is, since you've called it a
>"distribution". It isn't a distribution; "na" is a distribution. INET is
>a set of groups.
It is a distribution in the abstract sense, and that's how I meant it.
If I meant otherwise I would have written it I would have capitalized
the "d" as I usually do. (And to borrow from alt.comics.buffalo-roam,
and I would have used "the special subby-wubby voice").
I agree that modifying the software to include "Distribution: inet"
would be an improvement, but you never communicated that you would be
acceptable to you. I think we could reach a compromize and agree to
encourage Erik to add "Distribution: inet" at the gateway.
Of course, by being reasonable and backing down to accept a compromize
I have broken the most serious rule of "Rampant Useless Flame War
Mongering". But then again, you missed a great opportunity to do a
spelling flame on my typo of "adjacent".
times_usenet_has_seen_this_debate++
Tom
Gerry Swetsky
>think we've seen about four new inet groups over the past two years.
>This is not a mechanism that's being abused. Why fuss?
We don't like not being heard.
We don't like being heard and not being listened to.
We don't like being listened to and not understood.
Look......
$ ls /news/spool/comp/editors|wc -l
9
$ grep Distribution /news/spool/comp/editors/*
3318:Distribution: na
3321:Distribution: na
Does that help?
ONE USENET!!!
Joe Buck
>I agree that modifying the software to include "Distribution: inet"
>would be an improvement, but you never communicated that you would be
>acceptable to you. I think we could reach a compromize and agree to
>encourage Erik to add "Distribution: inet" at the gateway.
>Of course, by being reasonable and backing down to accept a compromize
>I have broken the most serious rule of "Rampant Useless Flame War
>Mongering". But then again, you missed a great opportunity to do a
>spelling flame on my typo of "adjacent".
You aren't reading, Tom. I offered several different possible compromises.
But your argument above reveals your technical ignorance. It's funny
how you claimed that only the ignorant objected to INET. Other than
Eliot and Michael, it seems to me that only the [partially] ignorant
defend it.
Erik *already* attaches "Distribution: inet" at the gateway! He did from
day one. The problem is that, for unmoderated groups, less than 5% of the
articles pass through the gateway. The remainder are generated by people
posting to the groups from their NNTP sites. These people don't even know
that they are posting to INET groups. Solutions that require everyone
everywhere to modify software (Eliot proposed one such solution) don't
work; people won't install the software.
The conclusion is that INET works perfectly well for moderated groups.
I do not object to INET for moderated groups; Erik may make any number
of them without a peep from me. All their articles go through the
gateway.
My preferred compromise is to "grandfather in" the existing unmoderated
INET groups, leaving the moderated ones alone. I would also be happy
to have Erik continue creating unmoderated groups if he wishes, *but
in a different namespace*. I suggest "inet.comp.whatever" for INET
comp groups. I would have no problem with Erik continuing to create
new moderated INET groups as well.
Requiring every site on the INET/non-INET boundary to add a huge
!this,!that line isn't acceptable, as you'll never get people to do it.
If someone can come up with an alternate way to get "Distribution: inet"
attached to every article, fine with me. Henry Spencer will no doubt
complain bitterly if you propose rewriting headers, however.
There is another technical solution that actually works: simply send
inet everywhere. Everyone would then be in inet. But in this case,
two sets of adminstrative rules seem hard to maintain. Why not just
formalize what Erik is doing and have a mechanism that permits large
mailing lists to be converted to newsgroups using a shorter process,
something more neutral than the opinion of one or two individuals as
to the merits of a group?
--
Joe Buck jb...@ohm.berkeley.edu
Christopher Davis
mcb> But inet control messages do not, and have never, leaked. Every inet
mcb> newgroup message goes out with Distribution: inet. Inet control
mcb> messages are only issued by two people.
HA! HA!!!!
I've seen comp.org.eff.news, an inet group, get unmoderated by happy
leaking newgroup dorks more than once.
Let's face it, folks, having two sets of groups in the same part of the
namespace is risky at best, and a most extreme headache (or worse!) at
worst. I created alt.comp.acad-freedom.{talk,news} in alt because I
feel that 'inet' is broken as it stands now, and I didn't want to break
it any farther. There's an RFD in now to move the comp.org.eff.* groups
into the mainstream, as well.
Let's fix inet by either absorbing it or putting it under a separate
hierarchy (like inet.* or inet-lists.* or something) already. Then we
can cleanly either feed it or weed it...
--
Christopher Davis * c...@eff.org * System Administrator, EFF * +1 617 864 0665
Samizdata isn't that different from Samizdat. -- Dan'l Danehy-Oakes
Tim Frost
>But they are already being shared by the vast majority of mainstream Usenet
>only these sites didn't know they were "special". Without all articles in
>the "INET" groups containing a "Distribution: inet" header there was no
>practical way to prevent that from happening. Fortunately the newgroup
>messages were distributed with that header so systems using explicit
>distributions in their sys files will not propagate the newgroup messages
>to non-inet sites but what about subsequent articles?
This would explain the number of articles in junk at my site in
New Zealand:
Newsgroups: comp.unix.solaris
Newsgroups: comp.os.msdos.4dos
Newsgroups: comp.dcom.lans
Newsgroups: comp.os.os2
Newsgroups: comp.unix.appleIIgs
Plus reports from the news system about receipt of articles for
unknown newsgroups.
How many of the above groups are INET groups?
Tim
--
________________ Tim Frost, Technical Consultant, GCS Ltd __________________
_________ P.O. Box 11-642, Manners Street, Wellington, New Zealand. _________
______________ Voice: +64 4 801-8000 Fax: +64 4 801-8888 __________________
______________________ Email: t...@animal.gcs.co.nz ___________________________
LCDR Michael E. Dobson
[some flamelettes and good discussion of the current problems with INET,
suggestion to grandfather non-moderated groups, aggreement to let current
and new moderated groups go through, etc deleted to save $$]
>
>There is another technical solution that actually works: simply send
>inet everywhere. Everyone would then be in inet. But in this case,
>two sets of adminstrative rules seem hard to maintain. Why not just
>formalize what Erik is doing and have a mechanism that permits large
>mailing lists to be converted to newsgroups using a shorter process,
>something more neutral than the opinion of one or two individuals as
>to the merits of a group?
>
This is a very good idea IMO. A large amount of what is contained is of
interest to USENET as a whole. Converting large mailing lists to
newsgroups is not the same as starting a new newsgroup from scratch or
splitting an existing group into subgroups. However a word of caution, we
all remember the nastiness with misc.activism.progressive which grew from a
mailing list.
Bengt Larsson
>In article <1992Jun15....@nmrdc1.nmrdc.nnmc.navy.mil> rd...@nmrdc1.nmrdc.nnmc.navy.mil (LCDR Michael E. Dobson) writes:
>>But they are already being shared by the vast majority of mainstream Usenet
>>only these sites didn't know they were "special". Without all articles in
>>the "INET" groups containing a "Distribution: inet" header there was no
>>practical way to prevent that from happening. Fortunately the newgroup
>>messages were distributed with that header so systems using explicit
>>distributions in their sys files will not propagate the newgroup messages
>>to non-inet sites but what about subsequent articles?
>
>This would explain the number of articles in junk at my site in
>New Zealand:
>
>Newsgroups: comp.unix.solaris
>Newsgroups: comp.os.msdos.4dos
>Newsgroups: comp.dcom.lans
>Newsgroups: comp.os.os2
>Newsgroups: comp.unix.appleIIgs
At least comp.unix.solaris was made by Michael C. Berch in a control message
with inet distribution. I just looked it up. This is confusing (trn telling
me there is a new comp.unix group when there was a discussion about
sysv groups just a while ago). One more for inet.*.
Bengt L.
--
Bengt Larsson - ben...@maths.lth.se
Dave Heath
people have a problem with the status quo, but are willing to support an
inet.* hierarchy, with some kind of grandfathering of groups that already
exist. It seems to me that the problem is that several people have their
ego's riding on the ability to create their own groups and that no amount
of rational discussion is going to change this.
-dave heath
Anyone got a scratch monkey?
>inet everywhere. Everyone would then be in inet. But in this case,
>two sets of adminstrative rules seem hard to maintain. Why not just
>formalize what Erik is doing and have a mechanism that permits large
>mailing lists to be converted to newsgroups using a shorter process,
>something more neutral than the opinion of one or two individuals as
>to the merits of a group?
>
If we were to split the inet groups, then there probably wind up being a
inet.comp.editors and a comp.editors (for example) Or people would wind up
getting inet.comp.editors through mail and if enough people did that, then
everyone may as well propagate the inet group because there would be less
overhead.
I believe an existing mailing list with a high (enough) readership should
probably have an easier time to get created as a mainstream group, whereas
a group like comp.sys.sliderule (:-)) should have to be voted onas who knows
how many people have a sliderule and would be interested in such a group.
(Yes I have a couple of sliderules sitting around here somewhere.)
>--
>Joe Buck jb...@ohm.berkeley.edu
-----Robert Gerber
--
+--------------------------+----------------------------------------------+
| Robert | rob...@bear.stonemarche.org |
+--------------------------+----------------------------------------------+
Bob Snyder
>>>>> rob...@bear.stonemarche.org (Anyone got a scratch monkey?)
>>>>> writes:
Robert> I agree with Joe here. usenet and inet are both world wide
Robert> distributions. If we were to split the inet groups, then
Robert> there probably wind up being a inet.comp.editors and a
Robert> comp.editors (for example) Or people would wind up getting
Robert> inet.comp.editors through mail and if enough people did that,
Robert> then everyone may as well propagate the inet group because
Robert> there would be less overhead.
I don't believe anyone has suggested that the current groups be moved
out of the mainstream hierarchy. Generally people have been saying
the groups should wither be simply grandfathered and allowed to remain
as they are, or to do a massive CFV on them all to have them get
"official" standing. I don't think the groups that currently exist
should be moved. They have been adopted by the mainstream usenet.
I do think, however, that any new groups should be created under a
seperate hierarchy (inet.* would be a good choice), so that we have
one set of rules for the namespace, and those who don't want to carry
the inet groups don't have to.
Ross Ridge
>Now, I say again: unless you think something is *really* broken, let's
>not fuss.
Nothing is broken? Ok, I'll just start gatewaying the NANET 4DOS
conference to comp.os.msdos.4dos. I've all the precedent I need.
Ross Ridge
--
Ross Ridge - The Great HTMU l/ //
[OO][oo]
ro...@zooid.guild.org /()\/()/
uunet.ca!zooid!ross db //
Bill Bogstad
>In <RSNYDER.92...@nutrimat.fergie.dnet.ge.com> rsn...@fergie.dnet.ge.com (Bob Snyder) writes:
>>So what is it to be then? A CFV for the grandfathering of current INET
>>groups and prohibiting of future non-voted upon groups? I think this would
>>make the most sense...
>
>Can we please not jump off of any bridges (yet)?
>
>BY ITSELF a vote will not and shouldn't affect INET. Since INET has
>never been subject the votes, guidelines, etc, and in fact predates
>any of the current policies, a vote means very little to those who
Yes, but a vote does affect USENET. One of the problems is that it
is often hard for users, administrators and software to tell the difference
between INET and USENET. Let's say for example that I decided to post a
call for votes for a USENET group. Maybe an unmoderated
comp.windows.x.announce. Or perhaps a moderated comp.editors. How about a
moderated rec.games.vectrex? What if at some later point USENET decided to
split one of these groups and delete the base group? What if???? I've
already stated that I will (if neccesary) do a RFD/CVF to create USENET
groups that are effectively indistinguishable from their INET counterparts.
Will INET (whatever that is) honor the results of decisions made about what
are USENET groups?
You have said in other messages that this isn't a major problem
since this INET doesn't create new groups frequently. Well, that is a big
part of why nobody has said anything about it for sometime. It was
something that could more or less be leaved with. Now we have three new
INET groups and the originator of the INET system publically stated at the
NNTP BOF at USENIX that he would have more free time now to "do INET
correctly". I (and others) spoke to him privately afterwards and it isn't
very clear what doing it "correctly" means or what his future plans are. He
seems to have no interest in talking to people about the current problems
with the technology by which INET is distributed or the political and
administrative problems of trying have two groups share a single namespace.
If it wasn't for the fact that INET groups were created in the past and its
creator has in the past made contributions the whole idea of shared
namespaces would be shutdown immediately. It seems that the only way to
stop the INET problem is to treat the people behind it the same way that we
treat news administrators like John Palmer (spelling?). You ignore their
postings and clean up after their messes. (This of course assumes that the
majority of involved news administrators thinks INET is a bad idea. Maybe
I'm just a member of a vocal minority....)
>In order to shut down INET, people have to
>stop accepting both the distribution and the groups involved. And
>they have to stop honoring the newgroups. I for one will not do that,
>since Erik and Michael have shown great restraint and wisdom in the
>past, whereas the popular vote has often yielded stunningly bad
>results.
rec.games.vectrex, although seemingly properly named, doesn't from
its traffic appear to have justified getting the fast track to group creation
which it received as part of the INET distribution. There are other groups
that have received small amounts of traffic and probably would have failed a
vote. Take a look at the number of active INET groups vs. the ones that
might as well not exist. I don't think the failure of voting is quite as
bad as you think. Yes, there have been some bad results of voting some of
which have been simply ignored by news administrators. I even think this is
a good thing. I consider this to be a safety valve, but don't see why it is
neccessary to invoke it on a regular basis. If the current methods for
creating groups don't work lets fix them not throw them away. (The argument
that there should be a "fast track" for existing mailing lists does have
merit. However, the RFD/CVF does more then just decide on a groups
creation. It also decides which hierarchy and name the group will by. I
for one think that these are important issues.)
Bill Bogstad
Bill Bogstad
>Of the people that "strongly oppose" the INET distribution, I have
>found that 40% of them liked it once someone explained the historical
>events that lead to it. The other 60% didn't like it because of the
>leaks. Leaks of ANY distribution are the fault the person at YOUR site
>that gave wrong information to the person at your ajacent cites.
As has been stated before, if the groups don't have a distribution
line you can't control their distribution with it...
>I don't think INET should be grandfathered into the mainstream. I
>think people running NNTP over the internet should vote on if we really
>want to grandfather any of those so-called "mainstream" newsgroups. I
>mean, the Official Usenet Board of Directors has never petitioned the
>NNTP Grand Review Board about this. Heck, has Kibo, Supreme Ruler of
>HappyNet ever even offered an opinion?
Excuse me? When has the INET review board queried the USENET
guidelines group? What if USENET decides it wants to create a group that
just so happens to have the same name? Just what are you INET people going
to do about it?
Bill Bogstad
P.S. Just so you know, my site is an NNTP site and is on the Internet.
Does this mean that I get to be on the INET review board?
Bill Bogstad
>would be an improvement, but you never communicated that you would be
>acceptable to you. I think we could reach a compromize and agree to
>encourage Erik to add "Distribution: inet" at the gateway.
This would solve half the problem from my perspective, but it is not
feasible. Many (most?) of the postings in INET groups don't go through the
gateway. Either all the USENET administrators on the Internet that get INET
groups need to install software that has default distribution lines or you
have to train all the users on those sites to manually add the lines.
Neither one is possible. I say half because you still have the problem of
two different organizations sharing the same namespace and trying to provide
similar services. You are undoubtedly going to have conflicts at some point
about names if nothing else.
>times_usenet_has_seen_this_debate++
I think it is only up to 3 or 4 at most now. Probably because
most people don't know what INET exists...
Bill Bogstad
Peter da Silva
>But inet control messages do not, and have never, leaked.
Yes they do. You're missing one important fact: C News sites typically
set up SYS files that look like:
hierarchy-list/all,!local,!regional
This site will distribute inet messages and inet control messages if it
receives them. There are so many sites set up like this, and it's so much
easier than tracking all the possible distributions, that it's never going
to be fixed. Distributions are unavoidably and unfixably broken. Arguing
that proper use of distributions will solve this problem is like arguing
that the coast guard can stop the drug trade.
>It would be improper for you to issue "global rmgroups" on inet groups
>(assuming you mean "without Distribution: uucp" on them)
No, I would put "distribution: uucp" on them, and they would still leak
into inet. Do you get it now?
Peter da Silva
>This is a very good idea IMO. A large amount of what is contained is of
>interest to USENET as a whole. Converting large mailing lists to
>newsgroups is not the same as starting a new newsgroup from scratch or
>splitting an existing group into subgroups. However a word of caution, we
>all remember the nastiness with misc.activism.progressive which grew from a
>mailing list.
How about this: for mailing lists, you drop the 100 vote restriction (there
is obvious interest in the topic) and just have a 2/3 majority vote on the
name?
Peter da Silva
>the groups should wither be simply grandfathered and allowed to remain
>as they are, or to do a massive CFV on them all to have them get
>"official" standing.
I offered to do that a few months back, but Erik objected in email.
James Kibo Parry
>think people running NNTP over the internet should vote on if we really
>want to grandfather any of those so-called "mainstream" newsgroups. I
>mean, the Official Usenet Board of Directors has never petitioned the
>NNTP Grand Review Board about this. Heck, has Kibo, Supreme Ruler of
>HappyNet ever even offered an opinion?
Actually, several HappyNet groups are abuzz about it. (You really
should get with the program and get it at your site, Tom. Hardly anyone
here reads the old net any more.)
-- K.
Brian Kantor
>Distributions are unavoidably and unfixably broken.
Oh, they can be fixed - or at least bandaided. You just set up a
script that takes any article that arrives at your site that shouldn't
have by virtue of its distribution, and delete it from your spool.
You're never bothered with stuff in distributions you shouldn't be
getting, and neither are any of the people downstream of you.
(I've convinced Rich to put this capability into INN, so you won't even
need the script.)
If you want to particularly obsessive about it, you issue a netwide
cancellation for the article. That way the sites that are doing the
wrong thing will eventually get pressured into fixing things, as people
become incandescently angry about what happens to their postings.
Imminent death of the net predicted. Film at 11. (or 8 am the next day if
you're connected by uucp.)
- Brian
Brian Kantor
>How about this: for mailing lists, you drop the 100 vote restriction (there
>is obvious interest in the topic) and just have a 2/3 majority vote on the
>name?
Why bother? The people doing the gatewaying of the group are going to
pick the name they think is right anyway, and it's likely to stick.
Given that, how is this different from Erik just creating the group in
the first place, if there's no voting on it's existence?
Golly.
- Brian
Dave Barr
>In article <O3K...@taronga.com> pe...@taronga.com (Peter da Silva) writes:
>>How about this: for mailing lists, you drop the 100 vote restriction (there
>>is obvious interest in the topic) and just have a 2/3 majority vote on the
>>name?
>
>Why bother? The people doing the gatewaying of the group are going to
>pick the name they think is right anyway, and it's likely to stick.
Yeah, but too often the list members don't pick a very good name.
--Dave
--
Dave Barr | loose: v. to set free, or not securely fastened.
dsb...@cs.psu.edu | lose: v. to miss from one's posession.
Peter da Silva
>In article <G3K...@taronga.com> pe...@taronga.com (Peter da Silva) writes:
>>Distributions are unavoidably and unfixably broken.
>Oh, they can be fixed - or at least bandaided. You just set up a
>script that takes any article that arrives at your site that shouldn't
>have by virtue of its distribution, and delete it from your spool.
I can already do that in my sys file, if I want to continually monitor
all the distributions that wash over my site. Much easier to just put
"all,!local" in and let them flow. And I'm aware of the problem: many
system admins aren't. Any solution to the problem that requires every
site change its software is not going to work.
So, we have the simple fact that groups created by Erik and Michael are
de-facto universally distributed usenet groups. But since they're not
"really" in the usenet hierarchy we can't do anything about name screwups
or making judicious splits.
Peter da Silva
>>How about this: for mailing lists, you drop the 100 vote restriction (there
>>is obvious interest in the topic) and just have a 2/3 majority vote on the
>>name?
>Why bother? The people doing the gatewaying of the group are going to
>pick the name they think is right anyway, and it's likely to stick.
It hasn't in the past. Remember misc.activism.general?
Eliot Lear
>It hasn't in the past. Remember misc.activism.general?
Remember sci.engr?
Oh. That exists. I wonder who picked that name...
--
Eliot Lear
[le...@sgi.com]
dil...@ccwf.cc.utexas.edu
There seem to be some misconceptions about the meaning of the words "manga"
and "anime" as well as a few other points which have got splattered in the
mud-slinging. :) Here is an attempt to clarify for the non-otaku:
(otaku is a japanese word used to describe die-hard fans)
du...@thismoment.Corp.Sun.COM (Duane Day) writes:
> Given that anime are a type or subset of manga, and
There are _many_ anime fans who do not know or care much, if anything about
manga. As far as the american market is concerned, neither is really a
subset of the other.
> Given that there is an existing newsgroup named rec.arts.manga, and
> Given that the newsgroup names are organized in a hierarchical fashion,
If hierarchical orginazition is so important, why isn't rec.arts.startrek.*
in rec.arts.sf.* ?
> it is proposed that rec.arts.anime be renamed to rec.arts.manga.anime.
Really, as far as I am able to determine, the two subjects are about equal,
to english-speaking audiences. Anime is not really percieved as a subset
of manga.
<from the rec.arts.manga Welcome message>
>>o WHAT IS REC.ARTS.MANGA?
>>A news group for the discussion of manga, or Japanese style comic.
What makes a comic "Japanese style"? Well, that can be hard to define, I can
almost instantly tell by looking at the artwork of a book, but it is hard to
explain. What is the difference between Bruce Springstein and Brahms? Well,
I can almost instantly tell by listening to the music, and I could say things
about the differnece in insturmentation, composition, tonal structure and so
forth, but to really understand, you have to listen to the music. Manga _do_
have a distinct and recgonizible style.
J...@SLACVM.SLAC.STANFORD.EDU writes:
> So it would seem to me that those of you that have argued all along that
> MANGAS are COMICS were telling the truth all along.
>
> And that those that claimed that MANGAS are not COMICS were just showing
>the all-too-common arrogance and ignorance of using a Japanese word wrong.
No, the word manga, in Japanese, translates to comic. However, when an
english-speaking manga fan uses the word, s/he means a specific style of
art and approach to storytelling, that is destinct from wahat someone means
whan they talk about comic books.
In article <l3sga9...@jethro.Corp.Sun.COM> du...@thismoment.Corp.Sun.COM (Du
ane Day) writes:
>>Someone somewhere, I don't remember exactly who it was :-), mentioned to me
>>that a poster from Japan referred to anime as manga TV shows or something
>>like that. Do you want to tell me that such a person coming to the net for
>>the first time wouldn't have an easier time finding rec.arts.manga.anime than
>>rec.arts.anime? Who would have a more difficult time finding it in the
>>manga hierarchy?
However, the perceptions of people from the culture in which the artform
originated and the english-speaking fans who now worship it are different.
du...@thismoment.Corp.Sun.COM (Duane Day) writes:
> Answer the question, Ken. How would renaming the anime group to
> rec.arts.manga.anime equate to "nuking" it or "victimizing" it or
> hurting it or its readers in any way, shape or form?
If someone goes looking for a group about anime, they could be completely
ignorant of the existence of manga, and thus miss the group entirely.
bey...@infonode.ingr.com (Maurice Beyke) writes:
> I don't have to. They're two different media. Print manga and comics are
> the same medium, i.e drawings, possibly with captions, usually displayed in
> a sequential order.
But there is a specific stylistic defference which seperates american-style
comics and japanese-style comics. There are also issues which manga fans
must concern themselves with that are unimprotant to comic book readers.
For most english-speaking manga fans, manga are a lot of pretty pictures
with a lot of incomprehensible scribbles associated with them. Translations
of manga stories are a regular, and nescary feature of any manga newsgroup.
I hope this clears up some of the misconceptions which I percieve to exist.
Really, manga and anime belong at the same level of the net hierarchicy,
perhaps in a japanese-culture subgroup. On the other hand, the number of
american-produced manga-style books is growing, and putting manga in a japanese
sub-branch might be misleading. Anime and manga are very closely interrelated
and share a number of problems, such as translations, artists and so forth.
One is really not a subset of the other.
--
\ _
- o We were once so close to Heaven, Peter came out and gave us medals,
^ declaring us the nicest of the Damned. - They Might Be Giants
Alex dil...@sleepy.cc.utexas.edu
I Go Elmo
> du...@thismoment.Corp.Sun.COM (Duane Day) writes:
>> Given that anime are a type or subset of manga, and
> There are _many_ anime fans who do not know or care much, if anything about
> manga. As far as the american market is concerned, neither is really a
> subset of the other.
According to the rec.arts.manga charter, anime _is_ a type or subset of
manga. That's the crux of the whole argument about manga/anime/hierarchies/ad
infinitum.
--
"You don't own my fear."
--Annie Lennox
Life...
>dil...@ccwf.cc.utexas.edu writes:
>>du...@thismoment.Corp.Sun.COM (Duane Day) writes:
>>>Given that anime are a type or subset of manga, and
>>There are _many_ anime fans who do not know or care much, if anything about
>>manga. As far as the american market is concerned, neither is really a
>>subset of the other.
> According to the rec.arts.manga charter, anime _is_ a type or subset of
>manga. That's the crux of the whole argument about manga/anime/hierarchies/ad
>infinitum.
That, and the apparent bait-and-switch on the part of rec.arts.manga by
voting for one charter and running with another. The only way .manga
should be permitted to change their charter without an explicit vote for
the charter change is by forming a subgroup, and as the topic in
question is the annexed .anime topic, .manga.anime is apparently
appropriate.
Basically what some people are saying is that the .manga group made the
mess, they're just making them realize that it stinks. (This from an
outsider.)
>elmo (mor...@fnal.fnal.gov,mor...@physics.rice.edu)
So why doesn't someone just start an RFD to correct the problem and
actually try to get something done regarding this? Is it going to be a
rename of rec.arts.anime, a rename of rec.arts.manga, a vote for .manga
charter change, what?
--
/// ____ \\\ |"Sorry, did I say something wrong? Pardon me for
| |/ / \ \| | | breathing, which I never do anyway so I don't even
\\_|\____/|_// | know why I bother to say it oh God I'm so depressed.
\_)\\/ | Here's another one of those self-satisfied doors.
gberigan `-' cse.unl.edu | Life... don't talk to me about life...."
Alex Dillard
>mor...@fnalo.fnal.gov (I Go Elmo) writes:
>> According to the rec.arts.manga charter, anime _is_ a type or subset of
>>manga. That's the crux of the whole argument about manga/anime/hierarchies/ad
>>infinitum.
>That, and the apparent bait-and-switch on the part of rec.arts.manga by
>voting for one charter and running with another. The only way .manga
>should be permitted to change their charter without an explicit vote for
>the charter change is by forming a subgroup, and as the topic in
>question is the annexed .anime topic, .manga.anime is apparently
>appropriate.
>Basically what some people are saying is that the .manga group made the
>mess, they're just making them realize that it stinks. (This from an
>outsider.)
I guess my point is that, if there is going to be a renaming of the anime
group, I think the manga group needs to be renamed as well. Most
new anime fans won't know what the word manga means. Perhaps the
Japanese needs to be removed entirely, and rec.arts.japanese.animation
should be created alongside r.a.japanese.comics. Then again, I find some
of the connotations of 'comics' insulting to manga, so I don't really
like the idea, but it sounds better than rec.arts.manga.anime, even
though the r.a.manga charter makes this 'right'.
>>elmo (mor...@fnal.fnal.gov,mor...@physics.rice.edu)
>So why doesn't someone just start an RFD to correct the problem and
>actually try to get something done regarding this? Is it going to be a
>rename of rec.arts.anime, a rename of rec.arts.manga, a vote for .manga
>charter change, what?
Maybe everyone is afraid of the flames, I know I am, besides I don't
see any real harm in letting things go as they are for a while, while
the emotions die down.
Bruce Becker
|In article <1992Jun20.0...@becker.GTS.ORG> b...@becker.GTS.ORG (Bruce Becker) writes:
|> Pure puppy poop. Around here the inet distribution
|> is passed around just like the sacred "real" ones
|> without any comment whatsoever. In fact in earlier
|> times I (among others) went out of my way to ensure
|> access to the inet groups on this most definitely
|> UUCP site...
|
|You also go out of your way to ensure that no alt group can ever be
|removed. You obviously enjoy promoting chaos, rather than anarchy, so
|why am I not surprised?
oh peter you are such fun to be around
obviously you don't read what I post but
instead just substitute what you want to
think is happening
if that's not promoting chaos, etc etc
--
,u, Bruce Becker Toronto, Ontario
a /i/ Internet: b...@becker.gts.org Uucp: ...!lsuc!becker!bdb
`\o\-e "You could have stopped me if only you'd done what I said"
_< /_ - Governments Everywhere
Bruce Becker
|In article <1992Jun16.1...@Warren.MENTORG.COM> t...@Warren.MENTORG.COM (Tom Limoncelli) writes:
|>found that 40% of them liked it once someone explained the historical
|>events that lead to it. The other 60% didn't like it because of the
|>leaks. Leaks of ANY distribution are the fault the person at YOUR site
|>that gave wrong information to the person at your ajacent cites.
|
|are pretty much unanimously opposed.
Pure puppy poop. Around here the inet distribution
is passed around just like the sacred "real" ones
without any comment whatsoever. In fact in earlier
times I (among others) went out of my way to ensure
access to the inet groups on this most definitely
UUCP site...
|I claim that everyone who has direct experience of the problems associated
|with inet is opposed to its current implementation.
What problems? Did you get a hernia from trying
to lift a heavy active file or something?
Bruce Becker
|"Around here" the bit, vmsnet, gnu, etc.. groups are passed around just
|like the "real" ones too. However these distributions were polite enough
|to use their own name space and not try to use that of the "real" Usenet.
|If the GNU folks wants to greate a new gnu groups, no one cares, it's
|their name space to manage. INET is different, they think they're
|privileged, that long standing Usenet conventions do not apply to them,
|and so they can create mainstream Usenet groups at will.
INET *is* part of the longstanding Usenet
conventions and, yes, in a way they *are*
sort of privileged - it's just that things
have gone on for so long that some may have
forgotten, and there are many people coming
onstream now who never knew about it to begin
with...
Edward J. Huff
(Bruce Becker) wrote:
>
> INET *is* part of the longstanding Usenet
> conventions and, yes, in a way they *are*
> sort of privileged - it's just that things
> have gone on for so long that some may have
> forgotten, and there are many people coming
> onstream now who never knew about it to begin
> with...
>
If so, it ought to be mentioned in the "What is Usenet" document.
The majority of users, I understand, are new users, and what they
know comes from those documents in news.announce.newusers.
--
Edward J. Huff hu...@mcclb0.med.nyu.edu (212)998-8465
Keck Laboratory for Biomolecular Imaging
NYU Chemistry Deptartment, 31 Washington Place, New York NY 10003
Posted via the new Mac NewsWatcher v1.3d2
Bill Bogstad
>
> INET *is* part of the longstanding Usenet
> conventions and, yes, in a way they *are*
> sort of privileged - it's just that things
> have gone on for so long that some may have
> forgotten, and there are many people coming
> onstream now who never knew about it to begin
> with...
There were even older Usenet conventions. Regular group names
started with "net" and moderated groups start with "mod". Except I think
that gatewayed arpanet mailing lists started with "fa". The conventions
changed. A lot of people seem to feel that it is time to change this one.
Bill Bogstad
Peter da Silva
> is passed around just like the sacred "real" ones
> without any comment whatsoever. In fact in earlier
> times I (among others) went out of my way to ensure
> access to the inet groups on this most definitely
> UUCP site...
Of course you also go out of your way to ensure that no alt group can
ever be deleted, and claim that this is in the best interests of the net.
Your opinions are not typical of Usenet admins.
> What problems? Did you get a hernia from trying
> to lift a heavy active file or something?
We have badly named INET groups, related INET groups in wildly differing
places. Well named INET groups that need to be split and can't, and so on.
Lots of organizational problems.
I would probably carry inet.*, even if it wasn't misleadingly placed in
the usenet name space.
Peter da Silva
> obviously you don't read what I post but
> instead just substitute what you want to
> think is happening
No, actually, I do read what you post. You have made your position on
alt rmgroups abundantly clear. Your position here, that any damn group
that someone who happens to be a buddy of Erik Fair wants should be
created without objection, is completely in line with it.
Perhaps you could explain these positions a bit more clearly if you
think I'm misinterpreting or misrepresenting them.
BTW: redirecting followups to a random nonexistent group is a pretty
daft move. It just pisses people off and convinces nobody of anything.
Chip Salzenberg
>If so, it ought to be mentioned in the "What is Usenet" document.
>The majority of users, I understand, are new users, and what they
>know comes from those documents in news.announce.newusers.
More like "History and Sources", methinks. Along with the history of
the Great Renaming, etc.
--
Chip Salzenberg at Teltronics/TCT <ch...@tct.com>, <7371...@compuserve.com>
"Do Rush place weird subliminal backmasked messages in their songs to
compel unwilling geeks to commit evil .sig atrocities?" -- Dean Engelhardt
Gerry Swetsky
>
> There were even older Usenet conventions. Regular group names
>started with "net" and moderated groups start with "mod". Except I think
>that gatewayed arpanet mailing lists started with "fa". The conventions
>changed. A lot of people seem to feel that it is time to change this one.
I for one, but it seems that interest is going to die out on this
the same as before . Nothing has been changed, nothing is settled.
Soon INET will come along and newgroup more inane groups - whether
we're in the middle of a RFD/CFV on them or not. Obviously there's
nothing we sheep here can do about it.
I've written David Lawrence <ta...@rpi.edu> two letters concerning
this and didn't receive the courtesy of one reply. As far as I'm
concerned, the present RFD/CFV method of newsgroup generation is
now a joke!
--
============================================================================
| Help stamp out stupid .signature files! Gerry Swetsky WB9EBO |
| vpnet - Public access Unix and Usenet |
| Home (708)833-8122 vpnet (708)833-8126 lis...@vpnet.chi.il.us |
============================================================================
David Barr
> I for one, but it seems that interest is going to die out on this
> the same as before . Nothing has been changed, nothing is settled.
> Soon INET will come along and newgroup more inane groups - whether
> we're in the middle of a RFD/CFV on them or not. Obviously there's
> nothing we sheep here can do about it.
Sure there is. Refuse to to honor the newgroups, refuse to pass
messages posted there, send out (regular?) rmgroups, and cancel posts
made to INET groups. Any one site can take one or all of these steps.
Soon the INET powers will listen to reason.
> I've written David Lawrence <ta...@rpi.edu> two letters concerning
> this and didn't receive the courtesy of one reply. As far as I'm
> concerned, the present RFD/CFV method of newsgroup generation is
> now a joke!
Well if you just wrote to him to say that that the present method was
a joke, I wouldn't think it worthy of a reply either. What do you propose
instead of the current method? I agree it needs to be changed (as I
have posted here several times in the past), but I wouldn't quite call
it a "joke".
--Dave
--
System administrator, Population Research Institute (814)-863-7374
Jamie Gritton
>> Obviously there's
>> nothing we sheep here can do about it.
> Sure there is. Refuse to to honor the newgroups, refuse to pass
> messages posted there, send out (regular?) rmgroups, and cancel posts
> made to INET groups. Any one site can take one or all of these steps.
> Soon the INET powers will listen to reason.
The last thing we need is to make a martyr of the INET folks. Maybe
you're right -- they could eventually be persuaded to listen to
reason. But bogus rmgroups and cancels? Calling such tactics "reason"
is stretching things just a bit.
--
James Gritton - gri...@byu.edu - I disclaim
David Barr
> The last thing we need is to make a martyr of the INET folks. Maybe
>you're right -- they could eventually be persuaded to listen to
>reason. But bogus rmgroups and cancels? Calling such tactics "reason"
>is stretching things just a bit.
And bogus newgroups aren't? Okay, global cancels would be rather
rude, but lots of general cleaning up needs to be done out there, and it's
just not happening.
How many people still have comp.sys.{mac,ibm.pc,next,amiga} in their
active file? (Last time I checked most had over 50% distribution)
Global rmgroups don't have any automatic effect. It's just a reminder
to people that something is wrong.
"Gee, there's lots of traffic in foo.bar.baz, I wonder... Hmm.. sure looks
like a valid group, I'll add it to my active file."
In my eye, the INET groups are just as bogus as the ones left over from
year-old reorganizations.
--Dave
--
System Administrator, Population Research Institute ba...@darwin.psu.edu
One is the loneliest number.
Two is as lonely as one, just shifted to the left.
William C Fenner
>Refuse to to honor the newgroups, refuse to pass
>messages posted there, send out (regular?) rmgroups, and cancel posts
>made to INET groups.
You think all these groups should just go away?
comp.ai.edu Applications of Artificial Intelligence to Education.
comp.ai.vision Artificial Intelligence Vision Research. (Moderated)
comp.dcom.lans.hyperchannel Hyperchannel networks within an IP network.
comp.editors Topics related to computerized text editing.
comp.edu.composition Writing instruction in computer-based classrooms.
comp.lang.asm370 Programming in IBM System/370 Assembly Language.
comp.lang.clu The CLU language & related topics.
comp.lang.forth.mac The CSI MacForth programming environment.
comp.lang.icon Topics related to the ICON programming language.
comp.lang.idl IDL (Interface Description Language) related topics.
comp.lang.lisp.franz The Franz Lisp programming language.
comp.lang.lisp.x The XLISP language system.
comp.lang.rexx The REXX command language.
comp.lang.scheme.c The Scheme language environment.
comp.lang.visual Visual programming languages.
comp.lsi.cad Electrical Computer Aided Design.
comp.mail.multi-media Multimedia Mail.
comp.music Applications of computers in music research.
comp.org.eff.news News from the Electronic Frontiers Foundation. (Moderated)
comp.org.eff.talk Discussion of EFF goals, strategies, etc.
comp.os.aos Topics related to Data General's AOS/VS.
comp.os.cpm.amethyst Discussion of Amethyst, CP/M-80 software package.
comp.os.rsts Topics related to the PDP-11 RSTS/E operating system.
comp.os.v The V distributed operating system from Stanford.
comp.periphs.printers Information on printers.
comp.protocols.iso.dev-environ The ISO Development Environment.
comp.protocols.iso.x400 X400 mail protocol discussions.
comp.protocols.iso.x400.gateway X400 mail gateway discussions. (Moderated)
comp.protocols.pcnet Topics related to PCNET (a personal computer network).
comp.protocols.snmp The Simple Network Management Protocol.
comp.protocols.tcp-ip.domains Topics related to Domain Style names.
comp.protocols.time.ntp The network time protocol.
comp.security.announce Announcements from the CERT about security. (Moderated)
comp.soft-sys.andrew The Andrew system from CMU.
comp.std.announce Announcements about standards activities. (Moderated)
comp.sys.cdc Control Data Corporation Computers (e.g., Cybers).
comp.sys.handhelds Handheld computers and programmable calculators.
comp.sys.intel.ipsc310 Anything related to Xenix on an Intel 310.
comp.sys.northstar Northstar microcomputer users.
comp.sys.super Supercomputers.
comp.sys.ti.explorer The Texas Instruments Explorer.
comp.sys.zenith Heath terminals and related Zenith products.
comp.terminals.bitgraph The BB&N BitGraph Terminal.
comp.terminals.tty5620 AT&T Dot Mapped Display Terminals (5620 and BLIT).
comp.theory Theoretical Computer Science.
comp.theory.cell-automata Discussion of all aspects of cellular automata.
comp.theory.dynamic-sys Ergodic Theory and Dynamical Systems.
comp.theory.self-org-sys Topics related to self-organization.
comp.unix.cray Cray computers and their operating systems.
comp.windows.x.announce X Consortium announcements. (Moderated)
comp.windows.x.motif The Motif GUI for the X Window System.
news.software.nntp The Network News Transfer Protocol.
rec.games.vectrex The Vectrex game system.
rec.mag.fsfnet A Science Fiction "fanzine." (Moderated)
sci.bio.technology Any topic relating to biotechnology.
sci.math.num-analysis Numerical Analysis.
sci.philosophy.meta Discussions within the scope of "MetaPhilosophy."
soc.culture.esperanto The neutral international language Esperanto.
I read 3 of these; your .newsrc says you read 4. Stop and think about what it
is that you're attacking before you jump into the war.
You're attacking a percieved problem, which shows no signs of being an actual
problem. What problems did the 3 recently created inet groups cause, other
than an increas of traffic in news.groups? How many of those problems were
actually caused by poorly configured software?
Bill
--
Bill Fenner fen...@cmf.nrl.navy.mil
Bruce Becker
It's true that there is a contingent who
are making such noises, and there certainly
are also those who are fairly vocal about
leaving things alone.
Probably the biggest crowd is the "tempest
in a teapot" folks, whose dominant response
is unbridled apathy.
Of course those sneaky "leave-it-alone"-ers
will no doubt claim this last group as more
of their followers 8^)
George William Herbert
>[...]
>In my eye, the INET groups are just as bogus as the ones left over from
>year-old reorganizations.
The INET groups as a concept predate all the reorgs, they
predate much of the current news software, and apparently predate
the majority of the (vocal) net's time of presence on Usenet.
In case nobody caught it when the other ten people who said this
earlier said it, catch it now.
The opinion that the INET groups are as bogus as leftovers
is about as reasonable an opinion as stating that since INET was here
before the guidelines were, it's their net and we can all go fuck off
and use email.
INET has for a long time been a quiet but integral and structural
part of how the "Usenet" as a whole worked. That some people whose
memory is too short to remember a time when the "guidelines" were just
that (and when INET was used well and effeciently and quite a bit more
than it is now) is an indication that Usenet has to some degree lost
it's "oral history" and is now suffering a generation gap. The really
unfortunate problem is that nobody is willing to stop frothing
and acknowledge that.
No matter what the final solution to this "problem" is,
looking at how the net came to be this way is going to be
necessary to have it come out without long-lasting destructive
consequenses. Communicate a little, guys... that _is_ what
Usenet is all about, no?
-george william herbert
g...@soda.berkeley.edu g...@lurnix.com g...@far.too.many.other.places.right.now
Michael C. Berch
> I for one, but it seems that interest is going to die out on this
> the same as before . Nothing has been changed, nothing is settled.
> Soon INET will come along and newgroup more inane groups - whether
> we're in the middle of a RFD/CFV on them or not. Obviously there's
> nothing we sheep here can do about it.
Personally, I am wondering which ones of the existing 61 inet groups
Mr. Swetsky considers inane, but let's put that aside for the moment.
I should point out that both Erik and I read news.groups (and I follow
news.announce.newusers closely) and there is no danger that an inet
group will purposely or inadvertently interfere with an ongoing CFV.
[It would be pointless to do so, since if a group is globally created
it (1) will not need the special effort to deal with it as an inet group,
and (2) it is exceptionally easy to arrange for an Internet list
gateway at a later date.]
For that matter, whatever problems exist in naming or splitting inet
groups are not intractable; contrary to various peoples' comments,
(Peter da Silva most prominently) inet groups can be split, renamed, moved,
or whatever. I only run one of the inet gateways and can only accomplish
such things directly for those groups, but I daresay namespace problems
can be rectified in inet rather more easily than in the mainstream.
--
Michael C. Berch
m...@presto.ig.com
earthbound misfit, I
> jb...@forney.berkeley.edu (Joe Buck) writes:
>
> |Oh, rot. Those who administer or have administered sites on the UUCP-net
> |are pretty much unanimously opposed.
>
>
> Pure puppy poop. Around here the inet distribution
> is passed around just like the sacred "real" ones
> without any comment whatsoever. In fact in earlier
> times I (among others) went out of my way to ensure
> access to the inet groups on this most definitely
> UUCP site...
But that's the whole problem. To the average user, "INET" groups look just
like regular ones. They're in the big 7 heirarchies, so they're "regular
newsgroups". That is, of course, until a group needs to be split, or is
badly named, or whatever. Then we hit the problems, the regular process of
Usenet can't be applied to INET groups, because they're Not Usenet.
To put my opposition simply: comp.something is a name of a Usenet group.
If it's not Usenet, it shouldn't be in comp.
- k
--
Craig Harding kil...@freddy.acme.gen.nz ACME BBS: +64 6 3551342
"Jub'f Xvob?"
David Barr
>You think all these groups should just go away?
>
Obviously not. Yes, I'm aware of how many and how old the INET groups
are.
>You're attacking a percieved problem, which shows no signs of being an actual
>problem. What problems did the 3 recently created inet groups cause, other
>than an increas of traffic in news.groups?
They continue an outdated precidence. Groups that aren't voted on don't
belong in the "big five" anymore. Obviously the existing groups should
be incoporated, but to continue the INET tradition in its present form
makes no sense. If someone wants to make a nifty new group without
voting for it, then use alt. We can't continue to manage a dualistic
network in one namespace.
>How many of those problems were
>actually caused by poorly configured software?
Huh? You trying to defend the "Distribution: inet"? Good luck.
--Dave
--
System Administrator, Population Research Institute ba...@darwin.psu.edu
I'd rather be rich than stupid. | Do not taunt Happy Fun Ball.
Jamie Gritton
> The INET groups as a concept predate all the reorgs, they
> predate much of the current news software, and apparently predate
> the majority of the (vocal) net's time of presence on Usenet.
> In case nobody caught it when the other ten people who said this
> earlier said it, catch it now.
So I caught it already. We know. If this weren't the case, the
solution would be much easier -- just remove the groups. But I still
don't see the old=sacred connection. No chance for progress (which
should have progressed years ago)?
> That some people whose
> memory is too short to remember a time when the "guidelines" were just
> that (and when INET was used well and effeciently and quite a bit more
> than it is now) is an indication that Usenet has to some degree lost
> it's "oral history" and is now suffering a generation gap.
What good would it do me to remember the good ol' days when INET
worked? That's one for the history books. Today, the INET groups are a
thorn in the side of news admins and users trying to make sense of a
schizophrenic namespace. And that counts for us folks on the Internet too.
> No matter what the final solution to this "problem" is,
> looking at how the net came to be this way is going to be
> necessary to have it come out without long-lasting destructive
> consequenses.
Whatever the past was, the present is this: one namespace should
have one administration. If the INET people won't kindly use an inet.*
hierarchy, maybe we should use a net.* hierarchy. Anyone for a Grand
Unreorganization?
Peter da Silva
> The INET groups as a concept predate all the reorgs, they
>predate much of the current news software, and apparently predate
>the majority of the (vocal) net's time of presence on Usenet.
>In case nobody caught it when the other ten people who said this
>earlier said it, catch it now.
Yes, but back then they were the "fa" (from arpa) hierarchy. If they had
remained a separate hierarchy we wouldn't have a problem now.
Just because the mistake was made a long time ago doesn't mean it wasn't
a mistake, or that circumstances haven't changed. Sometimes you need to
fix stuff like this: that's why we have reorganizations (grand or otherwise)
in the first place.
Peter da Silva
>For that matter, whatever problems exist in naming or splitting inet
>groups are not intractable;
Problems don't need to be intractable to be problems. The point isn't that,
with some effort, the problems can't be dealt with. The point is that by
existing the INET groups cause more problems than they solve, and there is
a relatively simple fix that can be made that will remove many of the problems.
>contrary to various peoples' comments,
>(Peter da Silva most prominently) inet groups can be split, renamed, moved,
>or whatever.
With greater difficulty and less likelihood of success. Plus, there's no
review process to prevent name space problems in the first place: nobody
no matter how intelligent or well-intentioned can catch all possible names,
or always find the best one. That's one reason I've been pushing multi-way
votes.
>I only run one of the inet gateways and can only accomplish
>such things directly for those groups, but I daresay namespace problems
>can be rectified in inet rather more easily than in the mainstream.
OK, then fix comp.soft-sys.nexstep or comp.unix.solaris. Fix comp.editors.
I'm glad that you're actually listening. Could you respond with something
more specific? Like, a plan to handle INET reorganizations, or a mechanism
to make sure that INET group names are acceptable?
Peter da Silva
>problem. What problems did the 3 recently created inet groups cause, other
>than an increas of traffic in news.groups?
Nothing, yet. When someone wants to reorganize comp.sys.next, the various
comp.software* subhierarchies, comp.unix, or comp.sys.sun then the problem
will crop up. Since they're not Usenet groups, they can't be included in the
reorg.
Remember the various attempts to split comp.editors?
Eliot Lear
>Just because the mistake was made a long time ago doesn't mean it wasn't
>a mistake, or that circumstances haven't changed.
No no no no no. The mistake was in C-News. In allowing a naked
``all'' in the distributions side of sys entries, distributions have
been muddied beyond belief. This could be fixed. If you argue
against INET you are arguing against distrbutions, which are really a
good idea implemented poorly (right now).
--
Eliot Lear
[le...@sgi.com]