Stargate, costs, and alternative services

[John Quarterman]

This article is intended to address a few mistaken assumptions
that have been posted recently.  I am a member of the USENIX
Board and of the Board committee which is working on the Stargate
project, though my opinions do not necessarily reflect those
of the board as a whole.


>Lauren ... there is no reason in (*&^(*&^ that the basic outline of
>a Comercial Stargate should remain behind closed doors at this point ...
>you have been playing with the basic issues and estimates for nearly
>more than a year, and approaching two. You should be able to summarize the
>proposed organization, it's staffing, and budget within 50%. And you
>darn well should be able to give some statement about the subscriber
>base demographics and the assumptions behind it.

This seems to be the most basic misunderstanding:  that Lauren,
the Board, or someone has a detailed business plan for Stargate
and is hiding it from the public.  Sorry, that's not so.
Lauren is an excellent technical consultant and has done
some amazing work both with the equipment and as a liaison
with outside organizations that have to be involved.
However, his main interests are not in the business end.

One of the main reasons the USENIX Board has taken a direct interest in
the project is in order to produce a business plan.  But that did
not happen a year ago, or two years ago:  it happened one month ago.
We do not yet know how much the service will cost to the subscriber,
how much investment capital will be required, how many staff will be
needed, etc.  And we can't know until the amount of certain key
expenses are settled.  Negotiations are in progress to answer these
questions.

As Lauren noted at the USENET BOF in Atlanta, it is hoped that,
if a business plan can be prepared in time, a testing period
will commence in January 1986 and last for several months.
During that period some hosts (exact number as yet undetermined)
would be able to sign up for a small scale test of Stargate.
This test would serve as a market survey and allow setting
subscriber costs appropriately as well as establishing demand.
Meanwhile, we are taking into account the statistics already
collected by others about USENET.

Incidentally, USENIX does not plan to run Stargate indefinitely.
The hope is to have a separate organization take over after the
testing period (or even during).  We do not yet know whether
that organization will be non-profit or for-profit.  It is
obviously preferable for it to be the former in order to
keep subscriber costs down.  However, it may be necessary
to make it for-profit in order to raise capital to start it.
We hope not.  But we don't know yet.


Many people seem to believe that Stargate is intended to replace
USENET.  This is not true:  Stargate is an alternative service.
The only way it would be likely to replace the existing network
were if most USENET participants decided to drop out and become
Stargate subscribers instead.  This is conceivable, given a good
enough Stargate service, but is hardly likely to happen immediately.


No one involved with the project has proposed limiting redistribution
of news received by Stargate, to my knowledge.  Personally, I don't
believe it could be done, even if it were desirable.


Stargate is not the only possible competitor to USENET.  There are
others already and anyone is free to start another at any time.  This
does not mean that a competitor has to be exactly like COMPUSERVE,
USENET, or any other existing network in content.  Though it should
be noted that the existence of COMPUSERVE has not destroyed USENET.

Nor is Stargate the only network that USENIX is willing to consider
supporting.  In response to the desires of the USENIX membership,
a Board member posted a request for proposals to USENET
several months ago.  It asked for proposals for improvements of mail or
news service similar to the services currently provided by the UUCP
mail and USENET news networks.  I will repost it.

Please be aware that more than a summary of proposed services
or technical means of achieving them will be needed:
a business plan for how to make it work financially is crucial.
For instance, a frequently-overlooked item is staff to do billing.
-- 
John Quarterman, UUCP:  {gatech,harvard,ihnp4,pyramid,seismo}!ut-sally!im4u!jsq
ARPA Internet and CSNET:  jsq at im4u.UTEXAS.EDU, jsq at sally.UTEXAS.EDU