Xref: utzoo news.admin:15618 comp.org.eff.talk:2946
Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!think.com!spool.mu.edu!munnari.oz.au!manuel!cmf851
From: cmf851@anu.oz.au (Albert Langer)
Newsgroups: news.admin,comp.org.eff.talk
Subject: Re: What is Usenet? LONG!
Message-ID: <1991Jun30.100732.11768@newshost.anu.edu.au>
Date: 30 Jun 91 10:07:32 GMT
Article-I.D.: newshost.1991Jun30.100732.11768
References: <1991Jun9.225229.25374@alembic.acs.com> <1991Jun20.212602.14470@newshost.anu.edu.au> <1991Jun23.050938.29045@wolves.uucp>
Sender: news@newshost.anu.edu.au
Organization: Computer Services Centre, Australian National University, Canberra, Australia.
Lines: 290

In article <1991Jun23.050938.29045@wolves.uucp> ggw@wolves.uucp 
(Gregory G. Woodbury) writes:

>Albert's latest has sparked some hard thinking about my view of the net,
>and this article represents my initial responses to changes in that
>view.  Since I wrote it as the changes were occurring, there is some
>rambling in places.  I hope that it can provide a springboard to the
>discussion that Albert hopes for.   

[and later on]

>Evolution is in effect and Albert's comments have effected some change
>in my perceptions of Usenet. I still disagree with some of the positions
>he holds, but I understand him more fully.

Well, I can't ask for much more than that (at least not all at once :-)
Your thoughtful comments are much appreciated (not just because of the
change in your view and perceptions but especially the requested change in
tone).

I'm still waiting to see if others are interested in a wider discussion
but meanwhile I will just comment on some of your points.

>	Oh please, at a certain point it becomes obvious that a
>particular point of view is not being shared by those-who-count-in-
>the-net (be they the mythical citizens or the sysadmins).  When it
>becomes obvious that some group is going to be emplaced despite all
>sorts of procedural errors and serious concern over the content of the
>group in the main hierarchies, all one can do is sigh and wait for a day
>to come when it is appropriate to say "I told you so" (if appropriate).

I don't want to start that debate again, but would just say the "certain
point" should be when the vote results come in with two-thirds in favour
and at least 100 more in favour than against despite various arguments
strongly put by those voting no. Even more so when participation in 
both the debate and the vote is several times greater than usual.

I am mentioning this because such debates do more than merely waste
bandwidth. They focus attention on day to day trivialities for which
procedures are ALREADY in place to deal with routinely, and thus
PREVENT serious discussion on major issues like strategic planning
for where the net is heading (both technically and "socially").

It is difficult to have the kind of discussion you now want, in an
atmosphere dominated by continuous storms in teacups that only
encourage people who might be interested in serious planning to
avoid the groups where it should be taking place.

>	Usenet is certainly a prototype of some kind of electronic
>community.  When it started, the hard core computer types were the
>creators of a new medium for themselves to use and play with each other.
>As the academic networks grew, and the Internet became more and more
>used, more and more users (who are NOT hard core computer types) have
>entered into the net and forced it to adapt to their presence.

Yes, one of the problems is that "hard core computer types" are not a
very random cross-section and certainly not one especially noted for
skills in establishing social institutions and structures for inter-
personal relations - indeed there is an undesirably high percentage
of "computer autistics".

On the other hand, a common technical culture probably made it easier
to get by than with a more heterogeneous group.

I want to see this technology, or more advanced versions of it, available
to and used by most of the literate population - or at least that
increasingly large section of it that will have access to computers
either at home or at work. That would amount to establishing another form
of mass media, rivalling the print media, but far more open, flexible and
above all interactive. Achieving that involves solving major technical
problems, but also signifcant organizational/social problems as regards
how the flood of contributions can be structured so that people can
find what interests them and others can draw their attention to what
they might be interested in.

The crude structures evolved so far may have been adequate for the original
community but they cannot last much longer. Usenet will either recognize
that it is a separate entity from its sites and establish structures that
can adapt to future requirements or it will be left behind in a technological
backwater and die.

>	Usenet is currently in a content situation where there is a
>rapid growth of special interest groups, all vying for market share, and
>still thinking that the effective way is to get onto the broadcast bands
>and be supported by the public wheal.  Other groups are using similar
>technologies (e.g. video/usenet) in directed ways (e.g. cable vs
>broadcast / news vs mail) to find their markets.

I think the future is likely to involve highly selective and customized
news feeds (with intelligent adaptive customization). But at the same
time the most efficient way to distribute news over a wide geographic
area is likely to be some kind of broadcast - i.e. pumping it all out to
everywhere so you can select locally (e.g. by satellite).

>	Then there is the FORM of Usenet.  Sites in discreet situations, run
>by folk caught in a mode of private service providers.  New situations
>are emerging and new administrative paradigms are needed to allow
>effective management.  However, the inertia of the given situation does
>not allow the rapid development of techniques for quasi-public
>information management - it is still being looked at as a set of
>discreet sites.

Yep. I think the net is still dominated by academic and research sites
for which "news" is only a relatively minor part of their functions,
with only a minority of public access sites that view themselves as
"service providers". But either way, inertia results in the focus on
discreet sites with inadequate organization for planning the development
of Usenet itself, as an entity distinct from its sites.

It just so happens that most user terminals are currently connected to
a multi-user site. Already there are single-user sites and one day
that might be the norm - whether they feed each other in a similar way to
the present relaying of news, or whether they use satellite downlinks.
(Both options seem feasible to me. ISDN in Australia is already cheap
enough to relay 1 MB of news in 2 minutes for the cost of a local phone call
and both traffic and rental charges can be expected to decline. On the
other hand satellite receiver cards for 24 hour 9600 bps SCPC are already
cheaper than high speed modems or ISDN cards, cheap window shade fresnel
zones made from alfoil have been demonstrated as effective antennas for
Direct Broadcast TV satellites in Europe and renting a satellite channel
is dramatically cheaper than either ISDN rental or the leased lines 
currently used for the Internet or the costs of long distance PSTN calls.)

We should be focussing on how users will want their news structured,
not on "sites".

>	Yes and no.  The models of site administration are still
>holdovers from the Middle Ages! (or is it that they are just middle
>aged? :-)  A few large university sites may have a strong enough user
>base to qualify as a democratic city-state, but most sites are still the
>feudal demense of the administrators and the system owners.

I think that's just transitional from the days when computers, and
communications lines, were incredibly expensive so access to one was
a "privilege". Attitudes change slower than reality, but they will
HAVE to adapt or their users will setup their own sites. It's the
same reactionary bullshit as the MIS departments that tried to
stop PCs and it will be just as unsuccessful. 

>	Here is the rub.  If all the sysadmins walked out and went on
>holiday for (lets say) three days, Usenet (and the Internet) would
>collapse into chaos.  The amount of work necessary to keep a site
>running (Usenet facilities only) varies from minimal to full-time.  It
>is in NO CASE zero.  Until the majority of sites can have usenet run
>completely automagically (with zero human effort), the sysadmin is going
>to be a necessary piece of the pie.

As far as I can make out the Internet does not depend on site sysadmins
but has its own paid staff with sites as customers like any other
telecommunications corporation. Usenet however is a much more interesting
phenomena, operating the kind of store and forward mail and news service
that would "normally" require an expensive PTT operation, but using no
paid staff and instead relying on the voluntary services of sysadmins
or newsadmins to a large scale but still informal cooperative.

Certainly that makes sysadmins or newsadmins an essential part of the
setup (and gives them the decisive say on such things as how the
net itself functions technically, establishment of feeds and so on),
but it doesn't give them any more say than other users on how the information
content is structured. After all it is the people who write messages
that provide the information content on an unpaid volunteer basis (cf 
newsagency journalists and some SIG providers on commercial computer
conferencing services) - and they aren't all admins.

Although I agree it will be DIFFICULT to have Usenet run automagically
(or with just NICs, not local admins at "sites") and it certainly is
not happening now, I think it's going to be ESSENTIAL for reaching out
really far beyond the current academic/research environment. The numbers
of admins that would otherwise be needed just aren't available.

Telephone operators were once essential with telephones but that COULD
NOT continue with mass telephone penetration so they HAD to be replaced.
The plans for replacing admins are already published (X400 and X500).
The disparagement of those plans by admins (and software developers of
the old school) are about as well founded and convincing as similar
cries from other people with special skills that are being made obsolete,
or who have designed technology that is becoming obsolete (but of course
equally popular among their colleagues).

"It's too big. It will never fly" :-)

>	This site runs pretty smoothly, but I still spend several hours
>a day keeping it going.  My upstream (currently) is a major node that
>has several people devoting a fair amount of their ("spare" :-) time to
>Usenet.  If these folk decided to let Usenet "run itself" for any length
>of time, a major portion of the net would collapse.  If only the top 50
>(or so) sites of Brian Reid's "influence list" were to drop out, the net
>would collapse.
>
>	To be sure, something else (with a strong resemblence to Usenet)
>would arise in a few days time, but the effect of the temporary collapse
>would set in motion such changes that Usenet' (prime) would not be the
>same as Usenet.

It amazes me how little attention has been given to automating the chores,
just to reduce them substantially, even though it would take a major
effort to eliminate them completely. While I agree about the effort
required to keep things running smoothly I disagree that there are
still 50 sites without which the net would collapse.

I suspect the net would merely partition (briefly) into a couple of
hundred or so isolated sub-nets with an average of 100 to 200 sites
in each and those would quickly reconnect with each other without
any major upheaval.

Just looking at the maps of connections makes it look as though the
net is not all that redundantly connected. But don't forget those
"paths" or "links" are merely configuration file entries, not physical
equipment for communication channels that cannot be flexibly redeployed.

The actual communication channels are provided by telecommunications
companies (both PSTN and Internet) and do not depend on any Usenet
sites. Every Internet site would be able to establish a link to any
other Internet site immediately and ditto for every PSTN site
(perhaps having to go further and at greater expense than the previous
links, perhaps not).

>[...]  The development of the independent city-states is a
>necessary step in the history of democratic development.  And be
>careful, it was the feudal barons, upset with the usurpation of their
>feifdoms by King John that sparked the signing of The Magna Carta and
>the establishment of Parliament.  Only later, after the Universities had
>nucleated the development of urban centers, were the barons forced to
>add the House of Commons and create a true democratic situation.

We could have an interesting discussion by pursuing these analogies
further, but I fear it has already gone too far for purposes of
news.admin and will leave that alone.

>	The money will still come from somewhere, and the "gold rule"
>will still apply.  The situation will only get worse as "public" users
>enter the net.fray.  The burden of public sites may make the system
>break down in such a way that a population split becomes necessary.

Well, I agree the "gold rule" will still apply in the sense that
even when a new mass media is established, the rich and powerful
will have a predominant influence through being able to hire
writers and journalists whose "product" is more interesting and
therefore more influential than many "raw" contributions. But I
don't agree that the wealth required to own the physical means
of transmission (whether "sites" or satellite channels) will be
an important factor. Even now with broadcast media like newspapers
and TV stations it isn't the printing and broadcasting equipment 
that is decisive but the ability to staff large organizations
providing the "product".

But why speak of the situation getting "worse"? That would only
happen if NECESSARY CHANGES are not made to adapt to the DESIRABLE
consequences of the "public" joining in.
 
>Take a look at the history of "FidoNET" to see what might happen.

Please summarize and/or tell me where to look.

>	Finally, take a look at the human species - one species
>(biological technology) and disjoint populations (segregation by nearly
>any basis you care to mention :-(
>
>	Usenet will evolve in a similar manner.  A generic technology
>and disjoint populations.  That fissioning is already in progress.
>
>	Usenet will go on, but which "Usenet" will be the "real Usenet"?

Again the pessimism! When you see a glass of water that is obviously
half-full, do you describe it as "half-empty"? :-)

The development of human society has seen the uniting of different
populations as well as splitting within that.

Usenet at present is uniting different technologies (e.g. FidoNet as
well as the old Usenet) and different populations (from different
countries and with different interests) and will also have to
differentiate in its handling of that.

Whether Usenet itself adapts, or splits, or dies one can be fairly
certain that the "Matrix" of email links between people will only
grow wider and not become compartmentalized into non-communicating
subgroups.

>[The line eater is a boojum snark! ]           <standard disclaimers apply>

They threatened its life with a railway share...

P.S. Warning (or hopeful sign :-) - I felt obliged to respond in view
of previous comments but may not be able to keep this up due to other
commitments. I hope others do.

--
Opinions disclaimed (Authoritative answer from opinion server)
Header reply address wrong. Use cmf851@csc2.anu.edu.au