There were 132 YES votes and 19 NO votes, for a total of 151 valid votes.
There were 3 abstains and 5 invalid ballots.
For group passage, YES votes must be at least 2/3 of all valid (YES and NO)
votes. There also must be at least 100 more YES votes than NO votes.
There is a five day discussion period after these results are posted. If no
serious allegations of voting irregularities are raised, the moderator of
news.announce.newgroups will create the group shortly thereafter.
sci.physics.strings String theory and related fields. (Moderated)
Voting closed at 23:59:59 UTC, 16 Mar 2004.
This vote was conducted by a neutral third party. Questions
about the proposed group should be directed to the proponent.
Proponent: Urs Schreiber <Urs.sc...@uni-essen.de>
Proponent: Lubos Motl <mo...@feynman.harvard.edu>
Proponent: Arvind Rajaraman <araj...@uci.edu>
Mentor: Ru Igarashi <r...@nucleus.usask.ca>
Votetaker: David E. Smith <da...@technopagan.org>
sci.physics.strings is primarily meant to be a forum for discussions
between the mainstream physicists who actively work in high energy
theoretical physics, especially string theory as well as other fields
that interact with the string theory community, including high-energy
phenomenology and high-energy experiments, and the laymen interested
in these questions.
The topics would include theoretical foundations of string theory,
dualities, physics of various backgrounds, quantum gravity,
mathematical methods and fields that are useful to study string
theory, stringy inspired models of particle physics beyond the
Standard Model, and the questions about their experimental
There are many string theorists who would like to discuss on the
internet such topics as those appearing in scientific papers on the
hep-th, hep-ph, and gr-qc archives at www.arxiv.org and who currently
don't find a suitable online discussion forum.
While several unmoderated online fora dedicated to string theory do
exist and have a large amount of traffic, experience shows that the
lack of moderation hampers the development of a decent discussion.
The only other moderated newsgroup that offers some room for such
discussions today is sci.physics.research. Recently, however, the
traffic on this newsgroup has become rather large, and the
discussions about string theory are never focused because a
significant fraction of the participants on SPR prefers various
alternative approaches to theoretical physics, and serious
discussions about string theory are usually interrupted by general
critical remarks of the participants who find string theory
The separation of string theory related issues from
sci.physics.research will hence avoid latent unproductive statements
of disagreement on the general relevance of string theory which was
felt to be unsatisfactory by all parties.
Moderation is clearly necessary and desirable in order to ensure an
interesting, productive, and focused discussion, just as moderation
in comparable newsgroups such as sci.physics.research or
sci.astro.research is generally believed to be beneficial.
sci.physics.strings is a moderated newsgroup for discussion of string
theory (a theory of quantum gravity and unification of forces) and
related fields of high-energy physics.
1) Posts must be related to the discussion of string theory, which
includes all the modern aspects of the theory (such as branes,
dualities, "M-theory"), as well as related fields, but excludes
non-stringy proposals for theories of grand unification, quantum
gravity and high-energy physics, unless a direct relevance for string
theory can be made explicit. Original and interesting criticism of
string theory is welcome.
2) The decisions of the moderators whether a posting is interesting
enough and appropriate for the newsgroup will reflect the views
generally held by typical members of the theoretical high-energy
physics community. But of course possibly unconventional and even
controversial stringy topics, as far as they are interesting for
physicists, are welcome.
Moderators will however not be responsible for the technical
correctness of accepted posts, though they may reject posts or send
the posting back suggesting revision when they note errors.
3) Posts may be rejected if they contain personal attacks, incitement
to retribution (i.e. flames and obvious trolling), or general
Clearly inappropriate postings include:
(1) Personal attacks;
(2) Discussion that isn't about or related to string theory
(3) Multiple messages that all say the same or very similar things; and
(4) Overly speculative postings and postings that are "not even wrong."
4) An article that is rejected will be tried to be returned (if that
is reasonable and feasible) to the author , perhaps with the
suggestion of a more appropriate newsgroup, or a suggestion of how to
revise it to make it acceptable.
5) Posts may include links to websites or personal pages that contain
relevant and on-topic information about string theory. Posts that
are, in the opinion of the moderator, primarily intended to solicit
business, sell a product, or further some commercial purpose, will be
6) Posting of binary files is prohibited. Short segments of source
code or quotes from papers may be posted; however, large projects and
complete papers should be archived elsewhere, and a link to the
download site included in the post.
MODERATOR INFO: sci.physics.strings
Moderator: Urs Schreiber <Urs.sc...@uni-essen.de>
Moderator: Lubos Motl <mo...@feynman.harvard.edu>
Moderator: Arvind Rajaraman <araj...@uci.edu>
Administrative contact address: s...@matfyz.cz
Article submission address: sci.physics.st...@matfyz.cz
END MODERATOR INFO.
The CFV was posted to the following newsgroups:
sci.physics.strings Final Vote Ack
a.ometa~web.de Alwiné Ometa
ad.renard~free.fr Alain RENARD
adamrhopper~hotmail.com Adam Hopper
ajt~Math.Berkeley.EDU A.J. Tolland
alanlewis~adelphia.net Alan Lewis
alberto~nuclecu.unam.mx Alberto Guijosa
amer.iqbal~rcn.com Amer Iqbal
andy.resnick~grc.nasa.gov Andrew Resnick
arajaram~uci.edu Arvind Rajaraman
arivero~unizar.es Alejandro Rivero
awalsh~sympatico.ca Anthony Michael Walsh
b.w.schofield~durham.ac.uk Ben Schofield
bds~ipp.mpg.de Bruce Scott TOK
bergv~uiuc.edu Maarten Bergvelt
bert~visi.com Bert Hyman
bob~cave.org Robert Wilkins
bobby~imsc.res.in BOBBY EZHUTHACHAN
borcis~users.ch Boris Borcic
bouvin~daimi.au.dk Niels Olof Bouvin
bray~iupui.edu Bruce David Ray
brycej~truman.edu Bryce Jones
buff~pobox.com William Denton
ceri~submonkey.net Ceri Davies
coolraven2000~web.de Karsten Schlender
david~farrar.com David Farrar
david_hume2004~yahoo.com David Hume
dgodinez~microsoft.com Douglas Godinez
dirk~hollywood.org Dirk Laureyssens
djurfeldt~nada.kth.se Mikael Djurfeldt
dpc~ipp.mpg.de David Coster
eforgy~hotmail.com Eric Forgy
Ekkehard~Uthke.de Ekkehard Uthke
fabinger~ias.edu Michal Fabinger
francis~strcprstskrzkrk.co.uk Francis Bursa
franck.thales~wanadoo.fr Franck T.
fynn11~gmx.de Robert Graham
ggw~wolves.durham.nc.us Gregory Woodbury
gmeiner~physik.hu-berlin.de Florian Gmeiner
hcheng~jacobi.harvard.edu Hsin-Chia Cheng
hees~comp.tamu.edu Hendrik van Hees
hendrik~enyo.de Hendrik Weimer
hllywd2~feynman.harvard.edu John Wang
hwabnig~aon.at Helmut Wabnig
ian~schwinger.harvard.edu Ian Low
induction~comcast.net Mike Valente
jacques.fric~tele2.fr Jacques Fric
jafferis~fas.harvard.edu Daniel Jafferis
jake~rset.net Jake Mannix
JamesL~Lugoj.com James Logajan
jbl~spamblocked.com J. B. Levin
jbrocks~oeb.harvard.edu Jochen Brocks
jgra~loc.gov James Graber
jjanhune+news~cc.hut.fi Salomon Janhunen
jjlxa22~xs4all.nl J. J. Lodder
jmalina~best.com JoAnn Malina
JoachimPimiskern~web.de Joachim Pimiskern
johan~naqua.se Johan Hermansson
john~stafford.net John J. Stafford
jones~physics.harvard.edu Gregory Chapman Jones
Jorma.Louko~maths.nottingham.ac.uk Jorma Louko
jpc~suespammers.org J. Porter Clark
kai.jelinek~ruhr-uni-bochum.de Kai Jelinek
kray~fas.harvard.edu Koushik Ray
kris~obsecurity.org Kris Kennaway
lech.zakrzewski~post.pl Lech 'obatal' Zakrzewski
lmend~ccny.cuny.edu Loren D. Mendelsohn
lsumner~gte.net Larry D. Sumner
Matej.Pavsic~ijs.si Matej Pavsic
medawar~panix.com Bassem Medawar
meyer~physics.harvard.edu Rob Meyer
mhatz~blackcat.com Mike Hatz
mitchtemporarily~hotmail.com Mitchell Porter
mjake~sirus.com Michael John
motl~feynman.harvard.edu Lubos Motl
mreece~midway.uchicago.edu Matt Reece
neitzke~fas.harvard.edu Andrew Neitzke
nhoffmann~despammed.com Norbert Hoffmann
nico_formanek~gmx.de Nico Formanek
olepetd~fysmat.ntnu.no Ole Petter Dybvik
paolobr~hotmail.com Paolo Branchesi
pascal~physics.utoronto.ca Pascal Vaudrevange
peet~physics.utoronto.ca Amanda Peet
Peter-Lawrence.Montgomery~cwi.nl Peter L. Montgomery
philipp~indy4.Theo-Phys.Uni-Essen.DE Philipp Kuhn
pjf~talen.it.net.au Peter Fimmel
ppisbetter~fastmail.fm David Burt
prinznirgendwo~gmx.net Fritz Walter
psvrcek~Princeton.EDU Peter Svrcek
ralf~macnews.de Ralf Bindel
ralphwallace~sbcglobal.net Ralph Wallace
refrost~dcwi.com Ralph Frost
rick~jarosh.net Rick Jarosh
rnews~river.com Richard Johnson
RONALD.J.RIEGERT~usa.dupont.com Ronald J. Riegert
S.Webb~open.ac.uk Stephen Webb
sarahman~seventh-sense-software.com Sabbir Rahman
sbfaulds~ihug.co.nz Stuart Brodie Faulds
senbrue~yahoo.de Kai Obst
shrao~nyx.net Shrisha Rao
shughes~iolfree.ie Shane Hughes
smi123th~innerx.com Frank Dodd (Tony) Smith, Jr.
smlucas~flashmail.com Steven Lucas
sparkyg~pacbell.net Sparky Gregory
strelka~gmx.net Michael König
stromme~mi.uib.no Stein Arild Strømme
swanson~theory.caltech.edu Ian Swanson
sweetser~alum.mit.edu Douglas Sweetser
tcfa300~ucl.ac.uk Nigel Seeley
tec~highstream.net Thomas Cuny
theis~hep.itp.tuwien.ac.at Ulrich Theis
thor.kottelin~elisa.fi Thor Kottelin
timo.herrala~metla.fi Timo Herrala
tobias~pa.msu.edu Peter Tobias
tobin~splorg.org Tobin Fricke
tomyee3~comcast.net Thomas Yee
trumbore~mindspring.com Chris Trumbore
Urs.Schreiber~uni-essen.de Urs Schreiber
volker.braun~physik.hu-berlin.de Volker Braun
wolfgang.lerche~cern.ch wolfgang lerche
xiyin~fas.harvard.edu Xi Yin
yaakov.krotin~sympatico.ca Yaakov Krotin
yangjade~hotmail.com Jie Yang
zed~resonant.org Zed Pobre
AbdulQat~sympatico.ca Abdul Qat
baccou~clipper.ens.fr P. Baccou
b_shagird~hotmail.com Buzurg Shagird
Fusioneer~direcway.com Charles Cagle
gfthomas~sympatico.ca G.F. Thomas
gprrspw~mindspring.com G.P. Ryan
helling~atdotde.de Robert Helling
hrundivbakshi~sympatico.ca Hrundi V. Bakshi
leonard~planck.com Leonard Cottrell
MaryJane1958~direcway.com Mary Jane Cagle
mmontcha~OregonVOS.net Matthew Montchalin
nurban~psu.edu Nathan Urban
psmyth~gmx.net Peter Smyth
rick~bcm.tmc.edu Richard Miller
ru.igarashi~usask.ca Ru Igarashi
spam_account~sympatico.ca Bill Pringlemeir
stainles~realtime.net Dwight Brown
suckfish~ihug.co.nz Ralph Loader
tomp~st.net.au Tom Perrett
Votes in error
dougbob~charter.net Brian Douglas
! Ack bounced
elikon~cox.net Richard Gaupsas
! No vote statement in message
klu~physics.muni.cz Josef Kluson
! Ack bounced
sandrakete~web.de Christoph Strembski
! Ack bounced
zimnyzenon~interia.pl Zenobiusz Zimny
! No ballot
Could I get some clarification on that?
> Pardon my ignorance, and tell me to mind my own business if appropriate, but
> I wouldn't think that a mentor for the formation of the group (in this case
> Ru) should vote, particularly a "no" vote. It just seems odd to me, thats
Dear Yowie, I was also surprised a bit. But the important question is
whether it is compatible with the rules? And I believe that it *is*
compatible, and both the proponents as well as the mentors are allowed to
vote, and not necessarily "yes". ;-) Am I wrong?
Ru has had doubts whether sci.physics.strings had the capacity to pass,
and it was probably a legitimate viewpoint for him because he is not
necessarily one of the proponents - his job was to help us. The attorneys
also sometimes doubt about their customers' innocence, don't they? ;-) Of
course, Ru was wrong.
E-mail: lu...@matfyz.cz fax: +1-617/496-0110 Web: http://lumo.matfyz.cz/
eFax: +1-801/454-1858 work: +1-617/496-8199 home: +1-617/868-4487 (call)
Well, the way I figure it, my job as mentor is to help shape up
a proposal to the best it can. Whether the proposal ends up being
good enough to me despite everyone's efforts is an altogether
different question. I may not like some aspect of how the charter
looks, even if it is as good as anyone can make it. In that case,
for the sake of consistency, as a news.groups regular (I have said
voting NO in such situations is appropriate), I vote against it.
Let me put it this way. Now I can't be accused of being biased
by helping a proposal come to fruition. :)
My standard proposals rant:
Quality, usefulness, merit, or non-newsgroups popularity of a topic
is more or less irrelevant in creating a new Big-8 newsgroup.
Usenet popularity is the primary consideration.
Well, thats true :-)
It just struck me as odd, thats all.
Thankyou for your openness.
I think it is fine. He gave good advice during the discussion; some of
it was accepted and some not.
I considered voting No because I have serious reservations about one of
the moderators. On the other hand, since I couldn't understand most of
the discussions regarding what might or might not be on topic, maybe I
also wasn't correctly understanding the tone of responses. I decided
not to vote at all, since I won't be using the group.
The reason why I questioned it in the first place, I guess, is that I
figured the votetaker and the mentor to be neutral third parties, who
therefore "shouldn't" vote.
I didn't follow the RFDs particularly well, because most of it went right
over my head anyway. I have no doubt that Ru gave his/her advice in as
unbiased way possible and if anything, helped stimulate discussion in a
contructive way. So I didn't doubt that Ru was *allowed* to vote, so much as
I thought it was a bit odd that s/he did. Had I been "mentor" (not that I'll
ever have the technical competance to act in that role!) I think I'd
disqualify myself from the voting process simply so relative newbies (like
me) couldn't call it into question.
Anyway, at least it shows that someone actually reads and pays attentions to
the results :-)
I am anxious to take part in this new newsgroup, but my company's news
server doesn't seem to have it available yet. It is not available at
google yet either.
Any ideas what is involved with getting my news server to make it available?
Are there any web sites that presently have it available?
I don't believe that it's available anywhere yet. It takes a few days
or weeks extra to set up all of the behind-the-scenes infrastructure
needed by a moderated newsgroup, so I don't expect to see the group for
a week or two yet.
Personal address: robkelk -at- jksrv -dot- com
Any opinions here are mine, not the Government's.
Speaking for myself, not the Ottawa News Administrators Group.
> Eric Alan Forgy <er...@nospam.com> wrote:
> > Are there any web sites that presently have it available?
> I don't believe that it's available anywhere yet. It takes a few days
> or weeks extra to set up all of the behind-the-scenes infrastructure
> needed by a moderated newsgroup, so I don't expect to see the
> group for a week or two yet.
The group is up on news.individual.net and I've been reading posts
since last week (I haven't tried to post and have no immediate plans
I just checked and was surprised to find it isn't up on Sympatico
yet (they're a large -- the largest I believe -- Canadian ISP). The
isc.org control messages were sent out for it on March 25, with
boosters on March 26 and April 1. Normally a Big 8 control
message would be virtually automatic is my understanding, so
I'm guessing it's because of the special configuration that needs
to be done for moderated groups. Perhaps Russ or someone
can post if that's it, or whether people really do need to write
their newsadmins to get a Big 8 moderated group carried these
no new newsgroup, unfortunately not even the "Big Eight" newsgroup, is
automatically added to *every* newsserver.
If your newsserver does not carry sci.physics.strings, go to
or click at "sci.physics.strings: Request creation" at
and fill in the name of your newsserver's domain (e.g. washington.edu).
Maybe it will convince them. Alternatively, send a polite mail to
addresses such as
where rutgers.edu is replaced by your domain or the domain of your
All the best
> Normally a Big 8 control message would be virtually automatic is my
> understanding, so I'm guessing it's because of the special configuration
> that needs to be done for moderated groups.
Nope. Moderated groups are no different than unmoderated groups for news
> Perhaps Russ or someone can post if that's it, or whether people really
> do need to write their newsadmins to get a Big 8 moderated group carried
> these days.
I have no idea why a particular server doesn't act on Big Eight control
messages. You'd have to ask them.
Personally, I automatically ignore all of those silly STUMP-generated
request messages. We've had four so far for sci.physics.strings, despite
the fact that we automatically added the group here within minutes of the
control message going out. That tells us that either the message is lying
when it says "A user of supernews.com ...", or that said user has a problem
with their news client that no amount of anonymous email requests will fix.
http://www.supernews.com - individual and corporate NNTP services
> KalElFan <KalE...@scifipi.com> writes:
> > Normally a Big 8 control message would be virtually automatic is my
> > understanding, so I'm guessing it's because of the special configuration
> > that needs to be done for moderated groups.
> Nope. Moderated groups are no different than unmoderated groups for
> news administrators.
I thought he meant that normally a newgroup would be virtually
automatically *sent* shortly after the vote result appeared (five
days or so), so that the newgroup was sent the date he named (March
25) reflected a delay so that the moderated group itself could be
configured properly, something that has routinely been done in
This may be reading way too much into what he wrote, but since he
has demonstrated the ability to be clueful in the past, I chose to
look for a way that what he wrote could *be* clueful.
Not that I remember whether March 25 was any more than five days
after the vote result appeared *anyway*.
> > Perhaps Russ or someone can post if that's it, or whether people
> > really do need to write their newsadmins to get a Big 8 moderated
> > group carried these days.
> I have no idea why a particular server doesn't act on Big Eight control
> messages. You'd have to ask them.
Granted, but still, the question here is a no-brainer. People really
do need to write their newsadmins to get a Big 8 group carried now;
people really did need to do so in 1995 when the Big 7 became the
Big 8 (on my then-ISP, normally a highly compliant one, the humanities.*
hierarchy was *not* added right away!); people really did need to
write their newsadmins to get Big 7 groups added as far back as the
Big 7 have existed.
The question always has to do with things like how many people, and
at how many news servers, and why and how those news servers are not
So basically, although KalElFan's question is easy to answer, the
answer is pretty meaningless, and Russ Allbery's comment gets to
the real issue: it's a nasty slog server by server.
This got worse in roughly the period 1997-1999 as many server operators
basically threw up their hands rather than deal with PGP authentication
of control messages, and as the more individualistic, the ones who
prided themselves on actually reading control messages, became unable
to deal with the HipCrime spews. I don't know when it got better before
that (propagation everywhere was definitely not a Big 7 group's
birthright from the start!). I think these days it's better than in
1999, because of the widespread availability of more or less compliant
alternative servers, and perhaps because the worse-run servers have been
getting weeded out more than others by the concentration of news server
operation in fewer hands.
> In article <87u101b...@windlord.stanford.edu>, Russ Allbery
> <r...@stanford.edu> wrote:
> > KalElFan <KalE...@scifipi.com> writes:
> > > Normally a Big 8 control message would be virtually automatic is my
> > > understanding, so I'm guessing it's because of the special configuration
> > > that needs to be done for moderated groups.
> > Nope. Moderated groups are no different than unmoderated groups for
> > news administrators.
> I thought he meant that normally a newgroup would be virtually
> automatically *sent* shortly after the vote...
No, I knew it was sent because I'd checked control.newgroup and
saw the control messages.
As for Russ's point that they're no different from a news admin's POV,
that surprises me. Messages get sent to the moderator as opposed
to just copied and propagated through Usenet, right? So that has to
be set up properly for starters, doesn't it? If Supernews and other
major providers didn't do that, their downstreams would receive
anything that anyone posted via Supernews.
Beyond that let's cut to the core issue: there either is or is not a
Usenet-wide difference in propagation, say in the first two weeks
after a control message is sent, that's achieved through the Big 8
process, versus alt.*. Are you saying there's no difference, Joe?
If that's what you're saying, then it's news to me. Virtually no
significant ISPs/NSPs automatically add alt.* groups nowadays
and haven't for maybe 10 years, but I've always understood there
were many ISPs /NSPs "automatically" adding Big 8 newsgroups
upon receipt of a newgroup control message from tale or now
his successors Russ et al.
I put "automatically" in quotes because I'm not necessarily
suggesting it's literally programmed to accept the newgroup,
just that they don't virtually all sit around and wait for user
requests the way they do with alt.* Particularly academic
institutions and ISPs large enough to have dedicated news
admins, my impression was that many, indeed most, will
add the Big 8 group fairly quickly.
> As for Russ's point that they're no different from a news admin's POV,
> that surprises me. Messages get sent to the moderator as opposed to
> just copied and propagated through Usenet, right? So that has to be set
> up properly for starters, doesn't it?
This is all handled automatically by software that acts on the control
message, or by admin action based on the control message. The only
difference to the administrator is literally a single character in the
creation command, or none at all if they're using PGP verification of
control messages. From the perspective of the individual news
administrator, no separate configuration is required for a moderated
More work is required for the n.a.n team to set things up, but that's what
we're here for, and we have five days after the end of the vote to get
that stuff set up. And of course a moderated group does require that the
moderators be set up to handle submissions, and there are times when the
creation was postponed until the moderators were ready, but we only do
that if the moderators specifically ask. For the most part, they're ready
when the time for the newgroup comes around.
> I put "automatically" in quotes because I'm not necessarily suggesting
> it's literally programmed to accept the newgroup, just that they don't
> virtually all sit around and wait for user requests the way they do with
Literally being programmed to accept the newsgroup is what the system is
designed to support. Whether any individual news site chooses to do that
or not is, of course, up to them.
> KalElFan <KalE...@scifipi.com> writes:
> > As for Russ's point that they're no different from a news admin's POV,
> > that surprises me. Messages get sent to the moderator as opposed to
> > just copied and propagated through Usenet, right? So that has to be set
> > up properly for starters, doesn't it?
> This is all handled automatically by software that acts on the control
> message, or by admin action based on the control message. The only
> difference to the administrator is literally a single character in the
> creation command...
I suspect few news admins nowadays have given up their control
over new group creation to automatic software. Which means that
single character difference kicks in. In this case it's a rarity because
almost all alt.* groups are unmoderated. The creation of a new
moderated group is something that may only happen once or twice
a year perhaps. For any given news admin, it may be a few years
if they personally missed the last couple. On the other hand, they
set up alt.* groups every week based on customer requests.
> > I put "automatically" in quotes because I'm not necessarily suggesting
> > it's literally programmed to accept the newgroup, just that they don't
> > virtually all sit around and wait for user requests the way they do with
> > alt.*
> Literally being programmed to accept the newsgroup is what the system is
> designed to support. Whether any individual news site chooses to do that
> or not is, of course, up to them.
I'm not sure why any of them would at this point, especially considering
how few Big 8 control messages go out compared to the total number
of groups they're asked to create and have to manually do so. I think
they would just eyeball the Big 8 message, perhaps match it up with
the CFV results announcement in nan, and create the group the way
they manually do for alt.* every week.
When it's moderated, all but the most experienced news admins
might not recall exactly how to do it with that extra character, or
might be more jittery about not getting complaints from people
expecting it to work properly. So maybe it gets delayed a bit,
as they research it or a more experienced person handles it.
In some cases maybe it even waits for someone to ask for it,
whereas a non-moderated group might not because it's like
99% of the groups set up.
Anyway, that's what I was thinking with the original post. That
there has to be some difference, and you've confirmed that. It
may not be a trivial difference to you and experienced admins,
but to the guys/gals handling the low-priority newsgroup requests
and the like, maybe that difference leads to delay.
> I suspect few news admins nowadays have given up their control over new
> group creation to automatic software.
It comes configured that way out of the box, at least if you install a
news server using most Linux distributions.
For the rest of what you're saying, I think we'd just be trading
speculation. What you're worrying about doesn't ring true to me, but I
have never done a survey of people who run news servers so I don't have
any particular facts to discuss.
> I suspect few news admins nowadays have given up their control
> over new group creation to automatic software.
As it happens, my brother has been a news-admin, or anyway has been
among the people handling news-admin duties, at two different companies.
The first veered wildly among auto-creation of all newgroups, personal
attention, and automatically creating valid newsgroups. The second,
the current one, as far as I know just copies the active file at the
ISC, which covers bunches of hierarchies, not just the Big 8. And
does this automatically, every x days, not on request or remembrance.
> Which means that single character difference kicks in.
(I.e., since what's under discussion here is whether propagation of
new moderated groups is slower than propagation of new unmoderated
groups, the single character difference involved in manually
newgrouping a moderated group in INN and related programs.)
> In this case it's a rarity because
> almost all alt.* groups are unmoderated. The creation of a new
> moderated group is something that may only happen once or twice
> a year perhaps. For any given news admin, it may be a few years
> if they personally missed the last couple. On the other hand, they
> set up alt.* groups every week based on customer requests.
I don't know what sort of server you personally use. I'm accustomed
to seeing whole *hierarchies* in the "new groups" list on a regular
basis, on the various ISPs I've used; I don't remember readfreenews.net
doing this quite so often, but I don't think it was that uncommon.
Many hierarchies have moderated groups in them.
Oh, also. You seem to assume that there are whole teams of news
admins doing newgrouping at every server. Nonsense. I'm sure there
are multiple news admins at the big sites, and they *may* split up
the work of newgrouping - but frankly, I'm not aware of a genuinely
big site that hasn't automated this anyway. (As far as I know,
Panix, from which I post, hasn't automated newgrouping; it wouldn't
surprise me even slightly if it were the biggest such site. I would
be absolutely floored to see Panix screw up setting up a moderated
group.) At smaller sites, there may be no designated individual
acting as news-admin; in this case, I'd expect that whoever had to
do a manual newgroup, for whatever reason, would look in a manual
or something, because *that* is a fairly default procedure. At
the genuinely small sites that people like to think of as the Usenet
norm (but that certainly do not now carry a substantial percentage
of posters), there would be only one news-admin.
[automating handling newgroups and rmgroups in the Big 8]
> I'm not sure why any of them would at this point, especially considering
> how few Big 8 control messages go out compared to the total number
> of groups they're asked to create and have to manually do so. I think
> they would just eyeball the Big 8 message, perhaps match it up with
> the CFV results announcement in nan, and create the group the way
> they manually do for alt.* every week.
I don't actually know how many newgroups go out in the Big 8 these
days. But anyone who was in the field in 1998 knows perfectly well
that it is *possible* for someone to send several thousand in a
Why on *earth* would they not want to automate something with that
kind of track record?
Where did you get the idea that the only control messages people get
are legitimate ones?
(My own interest in Big 8 propagation dates mostly from when, in 1998,
I was lead proponent for a group that had the misfortune of being
newgrouped during the first big forged-newgroup storm, and that as
a result was widely seen as a bogus newsgroup. Please do take my
word on this.)
Look, last time I posted I was chary of sketching a history of Big 8
propagation, because I haven't properly researched the subject, and
I haven't even properly researched the second half of the period at
*all*. But you're clearly not as clueful as I thought, so I'm just
going to go out on a limb and sketch one anyway. I welcome, of
course, corrections from people who have them to offer; for what
it's worth, one relatively obvious way to study the matter
systematically would be to consult the propagation numbers in the
old Arbitron surveys. Anyway:
In Olden Times, people created groups by posting to them, and
news admins decided whether to keep the groups so created by
various means. After a couple of years of their working more or
less in the dark, in late 1981 people started making lists of
newsgroups to guide them, and after another few years (I think
1984), they came up with something called the "checkgroups" message,
enabling news admins who trusted the person issuing the checkgroups
to automate creation and deletion of newsgroups. Around this time,
also, the concept of the control message was invented, so instead of
creating groups by posting to them you had to format the post in a
special way, and many sites restricted who could make this kind of
By this time there are two or four hundred official newsgroups, and
it is commonplace for smaller sites not to carry all of them. In
addition, by this time there are *numerous* regional, local,
linguistic, etc. hierarchies. Out of a desire to manage the handling
of official newsgroups by similar criteria to the handling of local
ones, there is a push for some sort of classification of the official
groups; in 1986-1987, this leads to Great Renaming that creates the
In the Big 7 as created, there is an explicit hierarchy of value.
comp.*, sci.*, and news.* are the Good Hierarchies that everyone
should take. They are transmitted to Europe.
misc.* and rec.* are the OK Hierarchies that sites with enough
storage, and with indulgent management, should take. I think
they are not transmitted to Europe.
soc.* is the Bad Hierarchy that only really generous sites should
take. It definitely doesn't go to Europe.
talk.* is the Eeeeeviiil Hierarchy that only obsessively complete
sites should take. It definitely doesn't go to Europe.
Let's be clear here: This sorting out is the *purpose* of the
Great Renaming. It is not an afterthought.
That said, at this time there was an increasing trend towards
transmitting Usenet via Internet connections, and the work on
transmitting via satellite was under way. Transmission costs were,
at that time, the main costs; I believe this is still true today.
So at the very same time that the Great Renaming was happening,
one of the chief motivating factors behind differential propagation
Also, around this time, people started forming private companies
to provide network service to people, and for these companies,
"full Usenet access" was at that time an important selling point.
So a class of server appeared for which carrying *all* the official
groups was taken for granted.
As a result, by sometime around 1990, an attitude began to gain
ground on Usenet that held that a *good* server would, barring
unusual circumstances, carry the entire Big 7. The previous
view has never entirely gone away, but I think no later than
about 1991 or 1992, the *predominant* reaction to the news that
some server didn't carry some official group had shifted *closer*
to "How naughty of them to ignore the Official List" than to "Good
for them that they're paying attention to their own situation".
(The *main* attitude has always been "How is that any of my
business?" I'm talking about what else gets said.)
This is the regime under which the Big 8 created, oh, about
fifteen hundred newsgroups in 1992-1997.
Several things combined to bring this period to an end. One is
that attentive news-admins went away. As internet service providers
found that Usenet was not, after all, their main selling point,
they stopped paying for people to work on it much of the time.
As Usenet itself got a worse and worse reputation, fewer people
wanted to get onto it, and even non-profit servers found that it
made less sense to allocate hours to it.
Meanwhile, among what attentive news-admins there *were*, a
pervasive sense developed that Usenet was profoundly broken. This
had more to do with things like spam and flaming than with the
newsgroup creation system, but to the extent that the news-admins
*did* tune in to news.groups, they constantly came across people -
among them both you and me - saying that this system was in profound
disrepair. This did not present an incentive to the news-admins to
honour the system's results.
*Meanwhile*, and critically - what brought the situation from slow
to rapid change - the vandals ramped up their attacks on the system
to the point of flooding the net with newgroups and rmgroups. The
main way news-admins were advised to deal with these floods was to
start processing all Big 8 control messages via a program called
PGPverify. But this program was reputed to be hard to set up.
Also, as far as I know, PGPverify worked only on INN at this time,
but for the first time in nearly a decade, there were new programs
being offered for running news servers, and many sites switched.
It is not an accident that the resulting era of relatively poor
Big 8 propagation coincided with the collapse of Big 8 newsgroup
That said, the era in question *also* ended. It turned out that
while Usenet was not for everyone, there were *enough* people
interested in it to keep a fair number of companies running along,
or to justify a national-level ISP in dedicating at least a few
people's time to Usenet alone. So with the average site going from
bad to worse, a market appeared for *good* sites to try to sell their
services. Meanwhile, Usenet turned out to be a good size for testing
things; as I understand it, both groups.google.com and
*.readfreenews.net owe their existence to testing. I don't know even
that much about the purposes behind individual.net, but the upshot is
that by today, there's a bunch of options for people whose local ISP
is asleep at the Usenet wheel.
At the same time, people found ways to adapt to the vandals. It's
not really, for example, all that *hard* to sync your active file
with the one the ISC maintains - I'm a poor programmer, but this is
something *I* could do. It's just that before 1998, few sites had
any real reason to do that, and in the immediate wake of the floods,
many posters were still at sites where there were no longer news-
admins who would know how. Nowadays, though, this sort of thing is
a solved problem, and is widely know to be so. I don't know, but
do assume, that something similar holds for the non-INN programs
that were new in 1998 but no longer are.
So although, today, there *may* still be poor propagation of the
Big 8 by number of *sites*, the majority of *posters* are at sites
that actually handle the Big 8 reasonably well.
Whether this will result in a revival of Big 8 newsgroup creation,
or even removal, remains to be seen. Results to date are not
> *Meanwhile*, and critically - what brought the situation from slow to
> rapid change - the vandals ramped up their attacks on the system to the
> point of flooding the net with newgroups and rmgroups. The main way
> news-admins were advised to deal with these floods was to start
> processing all Big 8 control messages via a program called PGPverify.
> But this program was reputed to be hard to set up.
It was also, prior to controlchan for INN, something that could still
allow and even make worse huge load spikes on the server while it tried to
process all the forged control messages.
> Also, as far as I know, PGPverify worked only on INN at this time, but
> for the first time in nearly a decade, there were new programs being
> offered for running news servers, and many sites switched.
tale tried pretty hard to make pgpverify work for as many different news
servers as possible, and there were patches to C News (for instance), but
yeah, I think D News was the only one among the new set that had
instructions there for it. I don't remember exactly how Diablo creates
newsgroups, but I do think it supports something similar these days.
> At the same time, people found ways to adapt to the vandals. It's not
> really, for example, all that *hard* to sync your active file with the
> one the ISC maintains - I'm a poor programmer, but this is something *I*
> could do.
chongo wrote actsync to help out with this with INN at least, and that
plus tale's wrapper around it has shipped with INN for quite some time.
[re newgroup control message abuses]
> Why on *earth* would they not want to automate something
> with that kind of track record?
They're 100% automated in this sense: they automatically don't
do anything with Usenet unless they're prodded to. For exactly
the reason you cited (hundreds of bogus newgroups, thousands
over the course of a year -- I was aware of that and don't know
why you assume I wasn't), they don't have things automatically
My eyes glazed over at the above quote point and I stopped skimming
your long post. You're either misunderstanding or imagining things I'm
not saying. For example right before the above you'd interpreted me
to be saying many dedicated news admins were common, when I
think no dedicated news admin is more common even with some
Usenet has become an extremely low priority for most of them. It's
a low-level customer service guy/gal that handles the crap -- requests,
complaints, and so on -- along with non-Usenet customer service.
It's the low-level tech who handles the Usenet server and software
problems as one part of his/her duties. A more senior admin has
Usenet among many duties and so on. Supernews, Giganews, etc.
may be said to have people working full time on Usenet, and maybe
a few other massive providers, but for 90%++ of ISPs (broadly
defined) I suspect it's a distinctly part-time job.
Someone asks for a group, it goes in the queue and a week or two
later a handful of new groups from the latest batch of requests show
up. Almost all of them are alt.* or other non-Big 8 hierarchies. If a
non-moderated new Big 8 group can easily get lumped in with those,
it might appear with the new batch.
Maybe it gets its own batch if somebody recognizes that the Big 8
is supposed to be a special hierarchy in terms of the hoops necessary
to get a group going. But new Big 8 groups happen so infrequently,
even that recognition per se may have declined. You look at some
of these rec.* groups and they're worse crap than many alt.* ones
in terms of the signal to noise ratio, so that doesn't help much either.
When there is no automation in terms of an automatic setting up
of groups, and you have the above type of situation where Usenet
is viewed as the crap you really wish you didn't have to spend any
time on, I think the moderated newgroups aren't really viewed as
Big 8 at all. They're viewed as a special kind of group where
some customers savvy enough to use them to filter noise might
want it. "Might want it" will get you no moderated strings group
on Sympatico, Canada's largest ISP (I just checked and they
still haven't added it), more than a month after the control message
If I ask for it they'll add it, and I'm sure many Sympatico users would
see it pop up the next batch and have an interest. But I never got
around to asking because this server I'm posting from is the main
one I've been using the last few months, and they did add it right
away. Maybe I'll request it on Sympatico but...
It's instructive that Usenet has become such a low priority that there
really isn't much distinction between Big 8 and alt or other hierarchies,
for many (and maybe almost all at this point, for all I know) ISPs. It's
really all just the same low-priority Usenet.