Re: [PLUG] PLUG Wireless/ Co-op?

[Noah silva <nsilva@atari-source.com>]

> 
> maybe a good start at any of this would be mapping members ... would it be
> possible to set up that same system that mapped the ricochet points to map
> members' locations?  (and by members, I am talking about any members who
> are INTERESTED in participating in some form of co-op or venture, project,
> whatever.) 

Assuming that the fixed Ricochet modems still have power (that is, they
havn't been removed, they are hard-wired into the power on utility poles),
that system could be used, and it is fairly accurate, but each person
would have to have, or borrow the ricochet modem (or someone will have to
visit each place).  Depending on the need, it may be easier for people
just to look up their address on a map that has the lat and long
listed.  I don't know if the census map I used to plot the points have
street names and such or not.

I was actually thinking of expanding my [private] ricochet network.  I
have a lot of cheap sparc boxes, and they are very
small/quiet/reliable. So, the plan was:
a.) Install Linux on the sparc box, and set up STARMODE and SSH.
b.) Write some sort of routing system for STARmode (rather than using the
modem's built in methods, since all of mine are "mobile" modems).
c.) Have people deploy them.

So f.e. I have one at my house, my girlfriend has one at her house, people
I know around drexel might each have one.  People that are out of range
from me could be routed by my girlfriend, etc.  (we'll assume these are
fixed stations, and that the routes are set up statically)

Anyone who wanted access could just plug an ethernet plug into the box
that was delivered with the ricochet modem.

Anyone nearby any of the stations could also use a ricochet modem to
connect to it from within about a 2 block radius.

The problem here?

a.) Getting enough equipment (not much of a problem).
b.) Writing the software (not much of a problem).
c.) Find enough people to fill the gap. (possibly a problem).
d.) Latency.

Each hop adds latency.  The speed of the modems is actually higher than
the commercial service used (I can get 192kbps no problem), but if you are
person A, and you have to go through B and C to get to me, that's 3 hops,
and each one adds latency.  On ricochet's commercial network, there are
some places where there are quite a few hops.  This isn't bad for
something like email, but it isn't pretty for applications like telnet.

e.) Bandwidth.

192kpbs might be fine, but if there were a situation where 25 stations
depended on one station to link to the other side of the network, the
modem doing the bridging wouldn't be able to keep up.

The practical problems I had were:
a.) (minor) Using the serial port, the modems can only get to 115.2 kbps.
b.) (not so minor) Using the serial port on the older sparcs, it only
works up to about 33.6.
c.) The sparcs don't have USB.
d.) We can't use people existing windows PCs with USB because STARMODE is
only supported under windows (and I wanted to be sure that the network
nodes would be fairly untouched).
e.) Getting enough people to cooperate.  I had a few interested parties,
but few of them lived close enough to get a direct connection, and I
didn't have people in-between.  People at Drexel have Free ethernet, so it
doesn't benefit them.  Drexel also has 802.11 now (though they didn't even
know to turn on the security at all for like 2 years!).

 -- noah silva
 
> I guess that's it for now. 


______________________________________________________________________
Philadelphia Linux Users Group       -      http://www.phillylinux.org
Announcements-http://lists.phillylinux.org/mail/listinfo/plug-announce
General Discussion  -  http://lists.phillylinux.org/mail/listinfo/plug