On Sat, 5 Apr 2003, David Young wrote:
> On Sat, Apr 05, 2003 at 09:31:31AM +0100, Julian Bond wrote:
> > First the environment. With Meshnetworks, Locustworld and others we're
> > seeing the emergence of technology that makes it comparatively easy to
> > link WLANs. Initially this will be a whole load of small scale
> > experiments with two or three APs linked at a time. But there will be
> > local pockets where we get a sudden "Mesh Disease" outbreak, and the
> > borders of many pockets touch, creating meshes with 100s of APs. This is
> > the epidemic model in disease transmission where isolated hotspots merge
> > to become a pandemic. At this point, a client device talking to another
> > client device within this meshed area have got numerous alternate
> > potential routes between them. Some are purely within the mesh, some go
> > via gateways in and out of the wider internet. Similarly, a meshed
> > client device talking to a server somewhere on the internet may also
> > have multiple potential routes through the mesh and then through
> > multiple mesh-internet gateways.
> >
> > So all I've done here is describe (probably imperfectly) the current
> > problem. Has anyone got any solutions?
>
> Is this a current problem? This sounds to me like a problem in
> the far-off future. I am aware of only two not-for-profit, ad hoc
> wireless networking projects, the Kingsbridge network and the modest
> network we've built here in Urbana. We have six nodes in two networks
> in Urbana. I think that Kingsbridge has six nodes in one network. At
> six nodes or less, these networks are a long way from the routing
> breakdown you describe.
>
> I suggest that people who aspire to build an ad hoc wireless network do
> not concentrate on the hard problems until they become hard problems.
> Hard problems that have needlessly distracted our tiny group in Urbana
> include channel allocation and scalability. We only found out the real
> problems when we started building things. As usual, the devil was in the
> details; details of 802.11 IBSS network coalescence, 802.11 multicast
> lossage, NetBSD's ARP implementation, and Prism2.x firmware bugs have
> been worse problems for us than the routing, which has ordinarily
> "just worked."
Some of the hard problems bite you in a big hurry as you start adding
nodes and aps... After a while having one broadcast ethernet domain gets
old in a hurry, as the background traffic goes up. Want to add multicast
and ipv6 and all of a sudden the incentive to subnet and turn layer2 ports
into layer 3 ports gets bigger in a hurry. In places where most
high-speed access links are via wireless (cities like lome, or lagos or
accra all in west africa) they are starting to learn this the hard-way.
> Dave
>
>
--
--------------------------------------------------------------------------
Joel Jaeggli Academic User Services [EMAIL PROTECTED]
-- PGP Key Fingerprint: 1DE9 8FCA 51FB 4195 B42A 9C32 A30D 121E --
In Dr. Johnson's famous dictionary patriotism is defined as the last
resort of the scoundrel. With all due respect to an enlightened but
inferior lexicographer I beg to submit that it is the first.
-- Ambrose Bierce, "The Devil's Dictionary"
--
general wireless list, a bawug thing <http://www.bawug.org/>
[un]subscribe: http://lists.bawug.org/mailman/listinfo/wireless
