From: Sameer <sameer@soda.berkeley.edu>
Date: Mon, 21 Feb 94 22:56:19 PST
To: cypherpunks@toad.com
Subject: A crypto-oriented inet connected household
Message-ID: <m0pYpz3-000157C@infinity.hip.berkeley.edu>
MIME-Version: 1.0
Content-Type: text


-----BEGIN PGP SIGNED MESSAGE-----






	This is something that I've decided to do, and we have to get
working on things, quickly.

	Check it out-- if you're interesting joining this thing, mail
me, and we'll talk. (sameer@soda.berkeley.edu)

	The idea at this point is to get a place in June, and then get
our net link maybe in the fall or winter sometime. (We'll likely have
a SLIP link starting in June.)
	If you're not a hacker/tech/geek/etc. -- don't worry. Read
through this and if you think there's something you can contribute, we
can talk.

	The name "NEXUS-Berkeley" *really* needs some improvement. We
haven't thought of a good replacement yet.

(This is still a draft)

1) Introduction
	What's the NEXUS-Berkeley Project all about? It's a lot.

	There are goals, and there are dreams.

	Most of all though, we're optimistic. We know we can do it.

	So, then.. what are we doing?

	We're hardwiring the Gaia Consciousness.

	In other words. A bunch of people who want high-quality net
access to their home are getting together, pooling funds, so that they
can do it. That's us, the NEXUS-Berkeley folk.
	Our plan right now is pretty simple. We're going to find a
place in Berkeley to move into. Thus we all have to work on how a
bunch of net-fiends like ourselves can live together.
	Then we're going to get a dedicated line to the internet, and
connect it to out home network of computers. I'd like to see a
terminal in the kitchen, the bathroom, a few in the living room, and
personal privately owned machines in each individual's bedroom. All
this hardware will be wired together into a local network, which is
then connected to the internet, via a dedicated line of some sort.
	Then we're going to start moving towards self-sufficiency--
providing internet services to the local community to bring in income.
Minimally, the net-link should be paid for by our income. Ideally,
we'll have enough income to pay the rent.
	The first step is shell accounts, sold cheaply. We can't
compete with the giants, netcom, crl, etc. We will provide a service
along with the shell accounts that won't be found at the giants-- a
community-oriented personalized approach. We want to build a community
around this thing. We'll have a local BBS, monthly meetings, internet
tutorials, and maybe even parties, where the attendants can take a
break from dancing and log into vrave.
	We don't want just tech-folk, though. We want BALANCE. We need
writers, artists, teachers, techies.. If you have a talent, it's
likely that the NEXUS could put it to good use. We need teachers to
run the tutorials. Artists to make flyers and decorate the place.
Writers to write press releases and other written work. Sociologists
to study us and keep us from killing each other.

2) Technical issues
 A) The link

	We have a bunch of options for the link. For startup we can
probably only afford at 56k line. There's a few choices in that
regard. We can go with The Little Garden, a cooperative, whom Sameer
would very much like to deal with, because of the cooperative nature
of it all.
	There's Internex-- they provide ISDN, which is high-bandwidth,
but right now ISDN runs .01/minute. If ISDN ever goes to untimed
service it would be excellent, but I'm not counting on it.
	There's sharing 5 SLIP lines with another site, which has the
benefit of cost, and a max-throughput of 56k, but it's sharing it with
another site, which has its obvious drawbacks.
	  If we want to resell the link 56k can only last so long. T1
is an inevitable situation. We will probably end up with an overbooked
T1 [The Little Garden] and move up from there (time and money will
tell) The provider is also quite up in the air. We can't be positive
who we will want to deal with.  Demand exceeds supply, so finding a
provider who is civil and non-anal retentive about usage could be a
chore. (Sprintlink seems like a good option.)

 B) The computers and hardware
	We'll need a main computer (or cluster, depending on how much
CPU-power we need) which is owned by the group, some terminals, (bad
ones for the kitchen & bathroom, but good ones for the living room) a
router, and a CSU/DSU pair. Individual resident members should use
privately-owned computers for their rooms.
	I'd like to have it set up so that there's a main NFS-mounted
filesystem with most stuff, email, news, general files, but each
individual's private computer is very isolated and paranoid about
security, where they can store sensitive things such as PGP private
keys. CFS can also be installed on the personal workstations, so that
private computers need not have that much local drive space, while
preserving privacy.
	  The main choices of processors seem to be quite numerous. We
have pretty much ruled out intel processors at this point. Sun and DEC
seem to hold the most promise for servers.  We have to learn a lot
more about machines before deciding what will suit our needs. We can
get something used.
	  What kind of ethernet are we going to use? How fast will it
be? What vendor?

 C) Machine Speed/Upgrades
	We will constantly be wondering exatcly where our income will
be best spent. Trying to make sure all of the hardware we have is up
to the task is going to be a royal pain. Each upgrade brings new
problems and exposes new bottlenecks.
	  A vitally important factor is machine speed. People hate
waiting 30 sec to a min for their mail or news to index or for a
program to start. This gets people complaining fast.
	Batch processing: For incoming mail and news, we can deal with
slower machines here, but we will need power news and mail get backed
up at another site and we get hit with a deluge all in one day (this
happens more than you may think)
	We should probably allow users to use 5 megs of diskspace with
maybe a one dollar per meg per month charge for diskspace used over
five megs.  We need need to know this before we know how much HD space
to buy.

 D) Dialups 
	This is a HUGE pain in the ass. The system Aron currently on
staff for (eskimo.com) has a basic policy of 90% free. Meaning 90% of
the time you get through on the first try. This has worked out to 13
users per dialup. We will likely find something similar. Optimally I
would like to see around 10 per dialup, but resources will again be a
factor.

 E) Miscellaneous Services
	Games: Something where the user has a graphic client at home
and uses our machines to communicate with other users in the same
general area or around the world I know DOOM can be played on
networks.  It would be interesting to talk with id software and see
what we can hack together.


3) Social issues
 A) Residents
	We have to decide how many resident members we want. They all
have to be pretty compatible in living style, as well. It seems that
5-10 would work out pretty well, with seven as the ideal.  We also
need *balance*. We need techs, artists, writers, teachers, and more
sorts to round out the local community. A household full of
tech/hacker/geek-types won't be well-rounded enough for my tastes.
(Well all be net-fiends, though, of course.)
	  Residents will be those that can make a healthy commitment
of time and money to the nexus. Time and a desire to work are
*clearly* more important than the size of your monetary
contribution. We will however need a decent amount of capital to get
the ball rolling. All types of people need apply. We have enough
people to run the tech end of things even now (though we would like a
few more).
 	  We intend to try living together for a month or 3 to iron
out the kinks before sinking our hard won cash into a ton of
equipment. There are a ton of horrible nasty issues that have to be
worked out before we can even THINK of doing this seriously.

 B) Local community
	We can set up a BBS to serve everyone in the community, to
help foster the community-oriented aspects. Sameer thinks that the
local discussions groups should be accesible ONLY via the BBS
(hopefully we can find a good free BBS system for UNIX) because
there's an entirely different feeling when accessing something via a
usenet newsreader than a BBS interface.
	We want people logging into the BBS (if they want to login
directly to a shell that is cool too) as the default. Something that
is intuitive and user friendly. Something that encourages
communication between our customers. Encourages them to make their own
spaces within the BBS (various conferences or groups-- it should be
trivial for a user to start his or her own local group). We should
encourage people use the machines for thier projects and interests --
hopefully things we haven't even conceived.
  It is important to get the users as involved in the bbs as
possible. Aron has noticed that people will put up with a lot of crap
if you tell them *why* it is happening and *what* is being done to fix
it. Encourge folks to build a place where they will feel comfortable.
	Once again the community aspect of the bbs is key, it is the
only thing we will be able to offer folks that crl and netcom
can't. Although I do not want to do this at the expense of
performance, community is something that we can only work to foster;
if we pull it off will not cost us anything in the way of money.
	  Although 95% of all muds are complete shit there are a
select few that are very much worth playing.  They also do a ton for a
community vibe (at least some of them do) This is another thing that
can start small (depending on resources) and grow to hugeness.

  C) Nexus Fissioning
	  These are basically the issues of nexi fissoning, or ouside
parties with the recources to do so "glomming on". Most of this can be
addressed at a later date.

  D) Meetings
	User meetings are key. We nned to be able to meet these folks
face to face and show them that we are in fact real folks. It makes
people feel better about the service and enhances the community
vibe. Aron would suggest monthly meetings. Sunday afternoons are
usually the best time. A place like a pizza joint usually has no
problem with this since they get little buisness on sunday
afternoons. Also that is the time most people will be able to make a
meeting. Before Aron left Seattle eskimo was getting between 20-30
folks a month sometimes as many as 50. He's sure it has swelled since
eskimo's subscriber base has exploded.

4) Economic issues
 A) Incorporation
	We should get an EIN, and a bank account, and all that good
stuff, but the question is whether or not we should incorporate as a
non-profit?
	  It has been mentioned that the "safest" way for us to go
about organising as a buisness is to organise as a corperation. This
protects us from being sued as individuals and has tons of other nice
benefits-- limited liability. There are also numerous reasons we would
like a corporation for tax reasons, because by moving personal
expenses to corporate expenses we can get bigger tax writeoffs.

 B) Incomes for residents
	At the outset everyone will need day jobs.

 C) Self-sufficiency
	Once the nexus becomes self-sufficient (I'm optimistic)
dealing with new arrivals/departures will get VERY interesting.

 D) Shell accounts for the community
	We can provide dialup access to the net, and our edge on the
big providers will be the niche market I've been talking about. The
community-oriented approach will give us our market. I'd like it to be
sorta like the well but less expensive, and less elitist. Monthly
meetings for users, occasional parties, and the like will build a
strong commitment between the nexus and the dialup members.
	A good pricing structure could be $15/month, $30/quarter, and
$90/year.

  F) Anonymous crypto services


5) Projects
	Projects will be what gives us noteriety in the E and RL
communities. They will also be a major source of income (depending on
the nature of the project obviously). Hopefully, they will also be a
hell of a lot of fun. We need any and all ideas for more project
areas.
	People who joined the NEXUS-Berkeley will have to decide for
themselves how they can contribute using their own particular skills,
in their own project. Thus before anyone joins the NEXUS-Berkeley an
individual should write up an informal "Project Proposal" such that
their commitment is made firm and we know how they'll be able to help.
	We will be writing up a "Request for Projects" and posting it
among various newsgroups, so that interested people may join us if
they so desire.

  A) Programming

	Being a mostly E oriented group (although not totally E, but
that will be the focal point and that which brings us together, a
facilitator if you will) we will most likely have tons of programming
projects going. Some of which will relate directly to the BBS, others
which may have uses to a variety of people. Writing free software for
net distribution and the like. The majority of the things below are
just ideas I am throwing at you. Feel free to pick them up and run or
add your own.


  B) AI
	This is an interest of Aron's and he would like to work on
various things in this area.

  C) Graphics

	Eye candy anyone? SGI or Amiga kinda stuff here, Dunno how we
can turn it into money though I *know* there are many ways. We could
possibly offer the basic scanning and manipulation services in order
to help pay for the workstations and other assorted equipment.

  D) Cryptography
	Sameer is very interested in implementing and developing
cryptographic and anonymity tools through the Nexus for fun &
profit. There is a good deal that we can do, including pay for use
anonymous remailing services, anonymous IP forwarding (IP laundry) TCP
in UDP tunnelling so that nomads can have TCP/IP connectivity
whereever they have a telnet dialup.
	Each of these schemes can bring in a good sum of money, if
arranged properly

  E) Operating Systems
	Have you ever been 100% satisfied with the OS you were using?
Although the manpower (ok personpower) for such a massive undertaking
would not come solely from the Nexus.
	I do not see why it could not be a focal point for
development. There is a ton of public domain code to start with and
build on. I would favor a microkernel UNIX variant. We could offer
free accounts to other folks who are willing to do coding (up to a
reasonable level). We could offer a full blown killer OS for the price
of Windows or System 7. Aron fully realizes this is one of his "lofty"
goals, but he would love to see it happen.

  F) Zines
	Chris Blosser has expressed his desire to publish a 'zine for
his contribution to the project. It would be a wonderful form of
community building, including merely public relations all the way to
contributions from subscribers to the various services which we will
sell.

  G) Music
	Making money in that buiness?  If we had the space (correctly
zoned of course) we could put on shows of various types, showcase
slightly obscure bands, and make some money in the process. There is
quite a bit of shit to consider insurance and crap.
	We could incorporate the net into our music making-- some
intercity/international collaborative music-making would be an
excellent project. This could be incorporated into collaborative
artwork as well, similar to things such as the OTIS Project.
	

  H) Raves & Parties
	As for a raves and rave like parties, these are other things
we can do (both free and for profit though if anyone wants to throw a
$20 event we will kill them) We have a few folks who are knowledgeable
about what to do here.
	It would be really great to put on parties at which we have
terminals available, either logged into vrave or connected with other
Nexus-homes across the world each of them holding parties at the same
time, so that international parties can be arranged. Sameer has
mentioned this idea to many people and they are very interested-- this
seems a great way of getting people interested in becoming members or
simply attending our parties.  
	Of course there are concerns about securing the equipment and
noise in the neigborhood. And numbers of people.  This is a great idea
for community building.
	Do we even let folks know where the house is? All these
party ideas are great, but do we really want a ton of people we don't
know, knowing that we have 100 grand worth of equipment in our house?
Security is a big issue here, as well as the fact that we may not have
enough room to pull shit like this off. This is definetly something
that needs to be hashed out.

  I) Testers 

	We will probably end up with a wealth of people who want to
beta test our projects. We can offer incentives to these people as
well (although early use of the program should be enough).

6) The house
 A) Location
	Sameer wants it to be close to the UC Campus. Sameer attends
the UC, so I'd like to live close by. There's more people in the area,
which is a plus, as well. Choosing a place close to the PacBell's
Berkeley Cental Office would give us improved line quality and a
cheaper leased line.
	  The location is to be somewhere in Berkeley, while there is
also a small possibility that the house could be in San
Francisco. There are a variety of factors involved, some of which will
not be resolved for a while.  There is also a more miniscule
possibility that the nexi will fission off before even getting started
with locations in each city. (If we end up with say 20 *seriously*
interested people this would be the most likely scenario.)

 B) Size
	The house should be large enough to house 5-10 people (around
7 being considered optimum) When it comes to floors and number of
rooms, that is still up in the air this depends mostly on how people
want to live, some can handle living in rooms with one or 2 other
people while other people *need* thier own space.
	Ideally, the lower level would be with a kitchen, living room,
and bathroom, and the upper levels would house the private spaces of
all the residents, so that we can host parties while preserving the
security of each individual resident's space.

 C) Cost
	??? I don't know much much a 6 BR place near campus costs. I
read one ad that was a 6 BR place, allegedly "close to UC" for
$2400/month.






-----BEGIN PGP SIGNATURE-----
Version: 2.3a

iQCVAgUBLWmcq3i7eNFdXppdAQHS0AP9EvGLjvpZf/p2uSaywniS2GW/gMiP7P1P
UKc9Dt5Sh8bhxFlW/pc2UsR7x4Xx2/qdkYHMbSP/KjbBJTa7viXtqLIP63vrgScz
9TaY6suKaelp7Hzg+S9zxbkzxry1LwXFpmGkxGze8prOG4FdEXCPQALCfaQI/E9U
xHEPWrALQZU=
=PqFj
-----END PGP SIGNATURE-----