Monthly Archives: April 1999

What is Usenet? A second opinion. (from news.announce.newusers)

Date: Tue, 28 Dec 1999 09:00:23 GMT
Message-ID: <>
From: (Edward Vielmetti)
Subject: What is Usenet?  A second opinion.
Newsgroups: news.announce.newusers, news.admin.misc
Reply-To: (Edward Vielmetti)
Original-author: (Edward Vielmetti)
Original-date: 26 Dec 1991
Archive-name: usenet/what-is/part2
Last-change: 23 Apr 1999 by (Ed Vielmetti)
Changes-posted-to: news.misc,news.admin.misc,news.answers
The periodically posted "What is Usenet?" posting goes:
>Archive-name: what-is-usenet/part1
>Original-from: (Chip Salzenberg)
>The first thing to understand about Usenet is that it is widely
>misunderstood.  Every day on Usenet, the "blind men and the elephant"
>phenomenon is evident, in spades.  In my opinion, more flame wars
>arise because of a lack of understanding of the nature of Usenet than
>from any other source.  And consider that such flame wars arise, of
>necessity, among people who are on Usenet.  Imagine, then, how poorly
>understood Usenet must be by those outside!
Imagine, indeed, how poorly understood Usenet must have been by those who
had the determined will to explain what it is by what it is not?
"Usenet was not a bicycle.  Usenet was not a fish."
Any posting like this that hasn't been revised every few months has
become a quaint historical document, which at best yields a
faint notion how the net "should have been" and at worst tries
to shape how the Usenet "really was".
The first thing to understand about Usenet is that it was big.  Really big.
Netnews (and netnews-like things) had percolated into many more places
than were even known about by people who tracked such things.  There was no
grand unified list of everything that was out there, no way to know beforehand
who was going to read what you post, and no history books to guide you that
would let you know even a small piece of any of the in jokes that popped
up in most newsgroups.  Distrust any grand sweeping statements about
"Usenet", because you can always find a counterexample.  (Distrust this
message, too :-).
>Any essay on the nature of Usenet cannot ignore the erroneous
>impressions held by many Usenet users.  Therefore, this article will
>treat falsehoods first.  Keep reading for truth.  (Beauty, alas, is
>not relevant to Usenet.)
Any essay on the nature of Usenet that doesn't change every so often
to reflect its ever changing nature is erroneous.  Usenet was not a
matter of "truth", "beauty", "falsehood", "right", or "wrong", except
insofar as it was a conduit for people to talk about these and many
other things.
> 1. Usenet is not an organization.
Usenet was organized.  There were a number of people who contributed
to its continued organization -- people who posted lists of things,
people who collected "frequently asked questions" postings, people
who gave out or sold newsfeeds, people who kept archives of groups,
people who put those archives into web servers, people who turned
those archives into printed books, talk shows, and game shows.
This organization was accompanied by a certain amount of disorganization
-- news software that didn't always work just right, discussions
that wandered from place to place, parts of the net that resisted easy
classification.  Order and disorder were part of the same whole.
In the short run, the person or group who ran the system that you read
news from and the sites which that system exchanged news with controlled
who got a feed, which articles were propogated to what places and how
quickly, and who could post articles.  In the long run, there were a number
of alternatives for Usenet access, including companies which sold you
feeds for a fee, and user groups which provided feeds for their members;
while you were on your own right when you typed this in, over the long
haul there were many choices you had on how to deal with the net.
> 2. Usenet is not a democracy.
Usenet had some very "democratic" sorts of traditions.  Traffic was
ultimately generated by readers, and people who read news ultimately
controlled what was and wasn't discussed on the net.  While the
details of any individual person's news reading system limited or
constrained what was easy or convenient for them to do at the moment, in the
long run the decisions on what was or wasn't happening rested with the
On the other hand, there had been (and always will have been) people who
had been on the net longer than you or I had been, and who had a
strong sense of tradition and the way things were normally done.  There
were certain things which were simply "not done".  Any sort of decision
that involved counting the number of people yes or no on a particular
vote had to cope with the entrenched interests of those who weren't about to
change their habits, their posting software, or the formatting of
their headers just to satisfy a new idea.
> 3. Usenet is not fair.
Usenet was a fair, a cocktail party, a town meeting, the notes of a
secret cabal, the chatter in the hallway at a conference, the sounds of
a friday night fish fry, post-coital gossip, the conversations overhead
in an airplane waiting lounge that launched a company, and a bunch
of other things.
> 4. Usenet is not a right.
Usenet is a right, a left, a jab, and a sharp uppercut to the jaw.
The postman hits!  You have new mail.
> 5. Usenet is not a public utility.
Usenet was carried in large part over circuits provided by public
utilities, including the public switched phone network and lines
leased from public carriers.  In some countries the national
networking authority had some amount of monopoly power over the
provision of these services, and thus the flow of information was
controlled in some manner by the whims and desires (and pricing
structure) of the public utility.
Most Usenet sites were operated by organizations which were not public
utilities, not in the ordinary sense.  You rarely got your newsfeed
from National Telecom, it was more likely to be National U. or Private
Networking Inc.
> 6. Usenet is not an academic network.
Usenet was a network with many parts to it.  Some parts were academic,
some parts weren't.  Usenet was clearly not a commercial network like
Sprintnet or Tymnet, and it was not an academic network like BITNET.
But parts of BITNET were parts of Usenet, though some of the traffic on
Usenet violated the BITNET acceptable use guidelines, even though the
people who were actually on BITNET sites reading these groups didn't
necessarily mind that they were violating the guidelines.
Whew.  Usenet was a lot of networks, and none of them.  You name
another network, and it wasn't Usenet.
> 7. Usenet is not an advertising medium.
A man walks into a crowded theater and shouts, "ANYBODY WANT TO BUY A
CAR?"  The crowd stands up and shouts back, "WRONG THEATER!"
Ever since the first dinette set for sale in New Jersey was advertised
around the world, people had been using Usenet for personal and for
corporate gain.  If you were careful about it and didn't make people mad,
Usenet was an effective means of letting the world know about
things which you find valuable.  But take care...
- Marketing hype was flamed immediately.  If you needed to post a
press release, edit it first.
- Speak nice of your competitors.  If your product was better than
theirs, you didn't say theirs is "brain damaged", "broken", or "worthless".
After all someone else might have had the same opinion of your product.
- Dance around the issue.  Post relevant information (like price, availability
and features) but make sure you didn't send everything out.  If someone
wanted the hard sell let them request it from you by e-mail.
- Don't be an idiot.  If you sold toasters for a living, you didn't spout off
in net.breadcrumbs about an international conspiracy to poison pigeons
orchestrated by the secret Usenet Cabal; toaster-buyers got word
of your reputation for idiocy and avoided your toasters even if they were
the best in the market.
- Disclaimers are worthless.  If you posted from, and put a note
on the bottom "not the opinions of foobar inc.,", you may have satisfied the
lawyers but your corporate reputation was still affected.  To maintain
a separate net.identity, you posted from a different site.
> 8. Usenet is not the Internet.
It was very difficult to sustain the level of traffic that was
flowing on Usenet back then if it weren't for people sending news feeds
over dedicated circuits with TCP/IP on the Internet.  That's not
to say that if a sudden disease had wiped out all RS/6000s and Cisco
routers that formed the NSFnet backbone, CIX hub, and MAE East
interconnects, that some people wouldn't be inconvenienced or cut
off from the net entirely.  (Based on the reliability of the MAE
East, perhaps the "sudden disease" already hit?)
There was a certain symbiosis between netnews and Internet connections;
the cost of maintaining a full newsfeed with NNTP was so much less
than doing the same thing with dialup UUCP that sites which depended
enough on the information flowing through news were some of the most
eager to get on the Internet.
The Usenet was not the Internet.  Certain governments had laws which
prevented other countries from getting onto the Internet, but that
didn't stop netnews from flowing in and out.  Chances were pretty good
that a site which had a Usenet feed could send mail to you from the
Internet, but even that was not guaranteed in some odd cases (news feeds
sent on CD-ROM, for instance).
> 9. Usenet is not a UUCP network.
UUCP carried the first netnews traffic, and a considerable number
of sites got their newsfeed using UUCP.  But was also fed using
NNTP, mag tapes, CD-ROMs, and printed out on paper to be tacked up
on bulletin boards and pasted on refrigerators.
>10. Usenet is not a United States network.
A 1991 analysis of the top 1000 Usenet sites showed about 58% US
sites, 15% unknown, 8% Germany, 6% Canada, 2-3% each the UK, Japan,
and Australia, and the rest mostly scattered around Europe.
The state of California was the center of the net, with about 14% of
the mapped top sites there.  The Washington, DC area was also the center
of the net, with several large providers headquartered there.  You
could read netnews on all seven continents, including Antarctica.
If you were looking for a somewhat less US-centered view of the world,
you could have tried reading regional newsgroups from various different
states or groups from various far-away places (which depending on where
you are at could be Japanese, German, Canadian, or Australian).  There were a
lot of people out there who were different from you.
>11. Usenet is not a UNIX network.
Well...ok, if you didn't have a UNIX machine, you could read news.  In
fact, there were substantial sets of newsgroups (bit.*) which were
transported and gatewayed primarily through IBM VM systems, and a set
of newsgroups (vmsnet.*) which had major traffic through DEC VMS
systems.  Reasonable news relay software ran on Macs (uAccess), Amiga
(a C news port), MS-DOS (Waffle), and no doubt quite a few more.  I'm
was typing on a DOS machine when I first wrote this sentence, and it's
been edited on Macs and X terminals since then.
There was a certain culture about the net that grew up on Unix
machines, which occasionally ran into fierce clashes with the
culture that had grown up on IBM machines (LISTSERV), Commodore
64's (B1FF 1S A K00L D00D), MS-DOS Fidonet systems, commercial chat
systems (America Online), and "family oriented" systems (Prodigy).
If you were not running on a Unix machine or if you didn't have one
handy there were things about the net which were puzzling
or maddening, much as if you were reading a BITNET list and you
don't have a CMS system handy.
>12. Usenet is not an ASCII network.
There were reasonably standard ways to type Japanese, Russian, Swedish,
Finnish, Icelandic, and Vietnamese that used the ASCII character set to
encode your national character set.  The fundamental assumption of
most netnews software was that you're dealing with something that looks
a lot like US ASCII, but if you were willing to work within those bounds
and be clever it was quite possible to use ASCII to discuss things in
any language.
>13. Usenet is not software.
Usenet software had gotten much better over time to cope with the ever
increasing aggregate flow of netnews and (in some cases) the extreme
volume that newsgroups generated.  If you had been reading news then with
the same news software that was running 10 years previous, you'd never have
been able to keep up.  Your system would have choked and died and spent all
of its time either processing incoming news or expiring old news.  Without
software and constant improvements to same, Usenet would not have been.
There was no "standard" Usenet software, but there were standards for
what Usenet articles looked like, and what sites were expected to do with
them.  It was possible to write a fairly simple minded news system
directly from the standards documents and be reasonably sure that it
will work with other systems, though thorough testing was necessary if
it was going to be used in the real world.  You did not assume that
all systems were tested before they have been deployed.
Usenet was in part about people.  There were people who were "on the
net", who read rec.humor.funny every so often, who knew the same jokes
you did, who told you stories about funny or stupid things they'd
seen.  Usenet was the set of people who knew what Usenet was.
Usenet was a bunch of bits, lots of bits, millions of bits each day
full of nonsense, argument, reasonable technical discussion, scholarly
analysis, and naughty pictures.
Usenet (or netnews) was about newsgroups (or groups).  Not bboards,
not LISTSERV, not areas, not conferences, not mailing lists, they're
groups.  If someone called them something else they were not looking
at things from a Usenet perspective.  That's not to say that they were
"incorrect" -- who is to say what is the right way of viewing the
past? -- just that it was not the Net Way.  In particular, if they
read Usenet news all mixed in with their important every day mail
(like reminders of who to go to lunch with Thursday) they were not
seeing netnews the way most people saw netnews.  Some newsgroups
were also (or "really") Fidonet echoes (alt.bbs.allsysop), BITNET
LISTSERV groups (bit.listserv.pacs-l), or even both at once!
(misc.handicap).  So there were some violent culture clashes
when someone referred to you favorite as a "board".
Newsgroups had names.  These names were both very arbitrary and very
meaningful.  People fought for months or years about what to name
a newsgroup.  If a newsgroup didn't have a name (even a dumb one like
misc.misc) it wasn't a newsgroup.  In particular newsgroup names had
dots in them, and people abbreviated them by taking the first letters
of the names (so alt.folklore.urban was afu, and soc.culture.china was
There was nothing vague about Usenet.  (Vague, vague, it was filling up
millions of dollars worth of disk drives and you want to call it
vague?  Sheesh!)  It may be hard to pin down what was and wasn't part of
Usenet at the fringes, but netnews tended to grow amoeba-like to
encompass more or less anything in its path, so you can be pretty sure
that if it wasn't Usenet then it will be once it's been in contact with
Usenet for long enough.
There are a lot of systems that were part of Usenet.  Chances were that
you didn't have any clue where all your articles will end up going or
what news reading software will be used to look at them.  Any message
of any appreciable size or with any substantial personal opinion in it
was in violation of some network use policy or local ordinance
in some state or municipality.
Some people were control freaks.  They wanted to present their opinion of
how things were, who ran what, what was OK and not OK to do, which
things were "good" and which were "bad".  You ran across them every
so often.  They served a useful purpose; there was a lot of chaos
inherent in a largely self-governing system, and people with a strong
sense of purpose and order made things a lot easier.  Just don't
believe everything they said.  In particular, don't believe them when
they sad "don't believe everything they said", because if they posted the
same answers month after month some other people were bound to believe
If you ran a news system you could be a petty tyrant.  You could decide
what groups to carry, who to kick off your system, how to expire old
news so that you kept 60 days worth of misc.petunias but expired almost immediately.  In the long run you would probably
have been happiest if you made these decisions relatively even-handedly since
that's the posture least likely to get people to notice that you
actually did have control.
Your right to exercise control over netnews usually ended at your
neighbor's spool directory.  Pleading, cajoling, appealing to good
nature, or paying your news feed generally yielded a better
response than flames on the net.
One of the ways to exert control over the workings of the net was to
take the time to put together a relatively accurate set of answers to
some frequently asked questions and post it every month.  If you did
this right, the article was stored for months on sites around the
world, and you'd be able to tell people "idiot, don't ask this
question until you've read the FAQ, especially answer #42".
The periodic postings included several lists of newsgroups, along with
comments as to what the contents of the groups were supposed to be.
Anyone who had the time and energy could have put together a list like this,
and if they had posted it for several months running they would get some
measure of net.recognition for themselves as being the "official"
keeper of the "official" list.  But don't delude yourself into
thinking that anything on the net was official in any real way; the
lists served to perpetuate common myths about who was talking about what
where, but that was no guarantee that things actually worked out that
In the real old days, when it cost real money to make long distance
phone calls to send netnews around the world, some people were
able to get their management to look the other way when they
racked up multi-thousand dollar phone bills.  These people were
called the "backbone cabal", and they had a disproportionate
influence on news traffic because, after all, they were managing
to get someone else to pay for it.
More recently, communications costs were (for many sites) buried in with
a general "internet service".  If you wanted to have a disproportionate
influence on news traffic, you needed to be able to beg, borrow, buy or
steal access to great big disk drives (so that you could keep a full
feed) and lots of memory (so that you could feed a lot of sites at once).
There was a vigorous, competetive cash market for news feeds; you
could get a newsfeed from a local provider via modem or via Internet
in all 50 states of the USA, more than 50 countries, and via
satellite in most of North America.  The notion that any one system
was a "pre-eminent site" was past; communications costs had
gotten low enough, and traffic high enough, that if any one node
were to have gotten wiped out completely it would have still been possible for
everyone to be back on the net within weeks.
You were better off starting up a mailing list.
If you *had to* start a newsgroup, you were best off starting a mailing
list anyway - even an informal one - to plan the newsgroup.  Get
a half dozen people to all agree on the basic goals, topics of
conversation, etc.  Figure that you have about two months to agree
that there's something worth talking about, get a hundred other people
to see your way, and run the vote.
There were time-honored rituals for newsgroup creation, designed
mostly to minimize the amount of work that news administrators
(the people who have managed to corral a bunch of disk space to
store news) had to do; in particular, this involved minimizing
the number of mail messages they had to read every day.  The
process involved handing off responsibility to a group of people
well-steeped in ritual (the Usenet Volunteer Votetakers) who
ran through the process for you.
I'm not sure what camels have to do with anything.  The only real
camel that had anything to do with Usenet is Larry Wall and Randal
Schwartz's "Programming perl", aka the "Camel Book", published by
O'Reilley.  Larry wrote "rn", one of the second generation of news
readers that let you ignore some news that you didn't want to read.
The process of getting rid of unread news got to be a complex enough
decision process that he wrote a programming language (perl) to help
him write a newsreader to replace "rn".
He never finished the new newsreader, though that's not at all
surprising.  "perl" is a pretty useful language, though.  If you
can understand "perl" you'll have a much greater appreciation for
the ability of news admins to get rid of things they didn't want to
There are easily $12M worth of computers that I could have pointed to that
were responsible for the transportation of netnews around the world,
plus another $12M per year in communications bills spent to keep
news flowing.  Much was made of the risk that miscreants would
do something horrendous that will mean The Death Of The Net As We
Knew It.  It seems unlikely, however, that this collective enterprise
would be endangered by any one user's actions, no matter how bold
they might be about trying to propogate their message against the
collective will of the rest of the net trying to keep them in check.
Was was surprising was how the success of the net became indistinguishable
from its failure.
If you are unhappy, what are you doing reading netnews?  Take a
break.  Stretch.  Walk outside in the sunshine or the snow.  Relax
your brain, watch some TV for a while, listen to the radio.  If
you need to communicate with someone else, give them a phone call,
or see them in person.
It's good to not spend too much time all in the same place with
a fixed focus - rest your eyes everyone once in a while by
looking around at something else.
Don't worry about missing anything, it'll all get re-posted
if it's any good.
Hours can slip by, people can come and go, and you'll be locked
in Cyberspace.  Remember to do your work!
-- Brendan Kehoe
Part of the apprenticeship for a network guru was knowing enough
other people and attending enough conferences to find out where
things were hidden.  This worked just fine when the Internet was
a small network.
-- Ed Krol
The second newsreader philosophy believes that you want to read
only 10 percent of the articles in any given group....  This
philosophy is far more realistic.
-- Adam Engst
... Usenet, als das Usenet noch Usenet war, und kein nicht-klickbares
Anhaengsel des WWW ...
-- Gert Doering
Copyright 1996 Edward Vielmetti.  All rights reserved.