[Freedombox-discuss] CCC Meeting Notes

[Isaac Wilder]

Hello Sascha, and thank your for your comments.
Responses inline.

On 08/16/2011 01:13 PM, Sascha Meinrath wrote:
> Hi Isaac,
>
> Just read your notes from CCC and wanted to comment on the
> wireless components. For the FreedomNodes, I would hope that the
> plan is to use 802.11a/b/g/n radios and not tie them solely to
> 5GHz.
Well, the b/g chips that are integrated into the dreamplug could be
used to communicate with client machines, and auxiliary radios would
be used for contact with other nodes. We were envisioning the
FreedomNodes as consisting of two parts, potentially connected by some
sort of PAN technology. The 'compute' part would be inside the home,
and the 'transponder' would be mounted outdoors. (This would also
allow for the use of solar energy, at least for the transponders).
> The 5GHz frequency, while often less congested, has a harder time
> propagating through architecture, trees, etc. than 2.4GHz.
Agreed. I don't have enough experience in the field to know whether
it's worth the trade-off. We are also trying to remain cognizant of
cost restraints, and have had better luck finding radio chipsets at a
reasonable price point that support only a single band.
> Since most consumer-grade equipment runs on 2.4GHz, being able to
> bridge between 2.4GHz and 5GHz would maximize the options for
> ad-hoc meshing.
I see what you're saying. I agree that it would be best. I only wonder
if we won't have to make some tough choices. I'm not familiar with the
process of procuring component hardware in bulk, and I don't know what
sort of savings we could expect were we to fabricate our own hardware.
Perhaps you would be willing to put me into contact with someone who
could help me find answers to these questions? (That goes for anybody
else that know hardware supply chains, or knows someone who knows
hardware supply chains. Please let me know)
>
> In terms of the Freedom Tower, the 3650-3700MHz band only allows
> higher-powered use for "lite licensed" devices -- which means
> registering the device's location with government authorities.
Perhaps I misunderstood, but I was under the impression that a single
license allowed an operator to run an unlimited number of sites
without reporting their location, as long as they are outside of the
exclusion zones. We do need something for the long-range links. 3650
seemed the best choice to us. Perhaps there are viable technologies
that we overlooked?
> 3650-3700MHz is also not harmonized globally, so that could present
> a problem for sales of these devices in many locations around the
> world.
I am not familiar with international regulatory climates, but I do
assume that the stack will have to modified for compliance. Then
again, there are places where even wi-fi is strictly off limits. In
the end, I doubt that available spectrum resources will be sufficient
to enable wide-scale adoption of the fractal-mesh architecture. I
believe that we will have either to lobby for more open spectrum, or,
in all likelihood, engage in civil disobedience.

Old arguments about efficient allocation seems to go out the window in
light of advances in software defined and cognitive radio technology.
It is the people's spectrum, and it should be used to their benefit.
That's my take, anyways.
>
> Overall, I love the idea of several different types of FreedomBox
> equipment; however, I also want to ensure maximum interoperability
> and extensibility of the technology.
I'm glad, and I appreciate your feedback. You are right in your
assertion that flexibility is key. Let's work to make it happen.
>
> --Sascha Meinrath Director, Open Technology Initiative New America
> Foundation
-Isaac Wilder
The Free Network Foundation

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/freedombox-discuss/attachments/20110816/1d1d8f1c/attachment-0001.html>