[Noisebridge-discuss] Mesh Thursdays @ Noisebridge?

[Jonathan Lassoff]

On Sat, Feb 9, 2013 at 12:29 AM, Alcides Gutierrez <alcides888 at gmail.com> wrote:
> Thanks for the quick reply, Jonathan.
>
> To be a little clearer on my position: I'm working on a mesh project for my
> particular neighborhood (Excelsior/Mission Terrace) that is not reliant upon
> the Internet, but perhaps has gateways here and there. My vision is a
> quasi-darknet with community-building services such as streaming media
> (educational/cc-license/etc), bulletin boards, voip (if manageable),
> local-business sites, and maybe even social networking.

Gotcha. That would be awesome!
I think some mix of internet and intranet services is what'll make
free (libre) networks like this worth the average user's time. I
realize that a big part of getting a private network scaled up and
getting more content available is going to be finding and recruiting
technologically-savvy users that can colocate hardware in their homes
or offices.

> From what I understand, scaling up directly can be difficult, but if working
> in network clusters, traffic might be easier to manage. I'm not definitely
> sure, but I'm intent to find out.

I love talking about this. My main focus and background is in network
engineering and ham radio, so theorizing about topologies and
protocols is my bread and butter.

Scaling up most topologies built with wired and/or fixed networks in
mind is often doomed to fail with wireless. Managing spectrum
efficiently is really important in keeping density up. It only takes a
small collection of misbehaving or uncoordinated radios to cause
enough interference to cause the goodput to drop significantly.

I think the future of doing this well is in low-power radios with
directional antennas pointing at near-neighbors, and having them
coordinate to avoid interfering with other neighbors.

This CBN project and many others are having pretty good luck with WiFi
gear from Ubiquiti networks that speaks their AirMax protocol. It
works around the "hidden node" problem that commodity WiFi gear has,
and makes for decent small-scale point-to-point and
point-to-multipoint topologies.

> I plan to document my journey as much as possible with videos and other
> media. I'd like to make this as easy for other neighborhoods to replicate
> (and innovate) as possible and hopefully we can make connections between our
> neighborhoods.

If you want to dig into something right away, I'd look into getting a
few Ubiquti Nanostations and some interested neighbors that all have
line-of-sight to one another.

> I do, however, want to exchange experiences and advice (I'd probably receive
> more than I could offer), to increase interest in mesh in general and also
> encourage more energy to be invested in mesh deployment and field-testing.
>
> Johnathan, I might wait before I request time if you're busy. I'm still in
> an early-development phase. I'd like to have a little more experience to
> have more fruitful conversations in the future.

I'm not super-busy, but I work out of town on most Thursdays, so I'd
come straight from work and take a different path.
I could meet up on 2/21 or 3/14.

--j