Sudo Room

Mesh/24 January 2015

Revision as of 22:13, 7 April 2015 by Tunabananas (talk | contribs) (added 1/24 mtg notes)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Attendees

Marc, Max-b, Alex, April, jake

Definitions

What differentiates sudomesh from similar projects?

People's open being part of a global network. Sudomesh the organization is in charge of the software. This is a bit different from Commotion.

It's not a mesh in a traditional sense, of like a single omnidirectional antennas. We're talking about community networks. Other installations around the US do not integrate the notion of providing high levels of service. We want to provide a high speed, reliable connection.

We are looking at a slightly differnt kind of network topology. We'll need some number of backbone links, likely with base stations that individuals can connect to. Different than the sort of "mesh" cloud of omni-directional links that "mesh" is often referred.

People's open also requires people who wants to peer on the network to sign a license agreement that explicitly denotes principles of network neutrality.

What is peoplesopen.net vs. sudo mesh?

PO is the network itself. It's a much bigger notion of devices running between each other. Sudomesh would be the group of people who are working on the devleopment of PO and are mounting things. Sudomesh handles the local iteration.

Peering agreements?

Levels of participation, what?

  • You can provide roof space, get connectivity from the high-speed uplink, and share that by creating a hotspot. By default this puts you on the network
  • Share your existing ISP connection using our firmware and access the network

Learning from Commotion

Mega thx to Preston Rhea for sharing his experiences with us. 

   Vast majority of people who understood how to do networking were either at OTI or already involved with projects around the world.

   New concepts were required to communicate to folks: had to convince people that this wasn't just "free internet".

   Detroit Digital Stewards project - "Commotion Construction Kit" https://commotionwireless.net/docs/cck/

   21 week program 2 days a week. A lot of community outreach work.

   In Redhook, they had a partnership with Brooklyn Fiber (kind of similar to monkey brains)

   Important to "co-involve" people

   What failed in other projects?

   Not having strong enough community engagement 

   Other project links:

   http://mountpleasantcwn.net/map-of-mtpcwn/ (DC)

   http://redhookwifi.org/red-hook-wifi-map (Redhook map)

   http://cassco.co/ (Detroit)

   Major takeaway: we have to build and foster our social networks to sustain our wireless network. Biggest failure of projects Preston has worked on was the difficulity of actually serving the needs of the communities in range.

Tunnels vs Tunneldigger

Are there advantages to not automatically connecting an open ssid to the internet through tunneldigger?

Our form of tunneldigger adds engineering complication. Also exit nodes are a point of failure.

Also currently tunnel establishment is a weak point. Can we make it bullet proof?

The alternative is for mesh points themselves to advertise internet tunnels. People could then connect to those tunnels (either the picostation or people's own devices)

Can we build both and then test them against eachother?

Let's build a STRICT mesh firmware (without tunneldigger) that just does mesh routing and basic tunnel to our exit.

Uplink Options

What questions do we need to have answered when exploring uplink options?

What is the price of transit?

  • 100megabits
  • Gigabit
  • Should be <$2 a megabit
  • Flat monthly fee? Hard cap (get cut off)? Soft cap (charge you more or slow down connection)?
  • Are we allowed to mount an antenna on the roof?
  • How do we get bandwidth to our first relay/the omni?
  • How much will it be for them to run fiber to us?
  • Would we have access to the facility?

Here's the list of uplink options:

LMI

  • They have a colo at 49 and Telegraph. That's 1,100 ft. from the Omni. One customer is getting 80 at $100 a month. But we could negotiate a different offer.
  • They are very nice.
  • They said they could even run fiber?? ~$2,000-3K

Internet Archive

  • Jenny is talking to Ralf.
  • April heard from Brewster that they are happy to link with us and are happy to brainstorm ways of going about it.
  • They are reopening their efforts, apparently
  • We have a node in kensington that probably has LOS. Getting it south to Berkeley/Oakland might be a little trickier...

Cogent

Level 3

  • Left a message
  • 365
  • Left 2 messages
  • UC Berkeley
    • Ah ok yeah so this MAYBE an option (a couple mesh members work for a berkeley WISP right now). He doesn't really have offices in Downtown Berkeley, his office is currently in West Berkeley, near 9th and Durant. He does have antennas on the building and an uplink, and if we had LOS there, we could probably get some bandwidth, but yeah it would need to be VPN'ed to an exit server.
    • There is another office in Downtown Berkeley that has pretty much the same setup. We'd likely need to provide hardware and we wouldn't necessarily have physical access to the buildings though.

Decision

  • Get connected to Internet Archive's Richmond connection, if possible.
  • Reach from place in Kensington, to garden roof node in Berkeley, to Omni
  • Set up a meeting with Ralf to see if we can repositin their antennas to look at us?
  • Someone come with April or Jenny to talk to them soon.
  • Update after meeeting with them: they are also looking for high points in the Kensington area, and are open to working with us, but asked for more information about our project and suggested we might be able to help them with antenna locations. fair enough :]
  • Get connection (if price suits) from LMI
  • April talked to LMI on the phone. They are hella close to us; we should also get a 100megabit connection from them.
  • Someone besides April who knows what they are talking about should call them this week to iron out the details.
  • Phone number: (510) 843-6389

Map Logic

We need a topographical map of the Bay Area that's more interoperable and better than Google Earth, pls. One option: from Omni, try to head directly east to the hills. We really need a building to mount upon in the Oakland hills. Do you know someone?

  • A map of broadband access in Oakland would be nice
  • A map of libraries also would rule

Logistics

Website Edits

  • peoplesopen.net
  • Front page
  • What if we made it more personable and made an explicit social and political statement at the beginning? I think that'd may be more of a hook.
  • Watch the video below for more information on mesh networking, and stay tuned for updates from the People's Open Network.
  • About
  • How to help
  • Wishlist
  • Developer
  • Documentation
  • Map
  • Wiki
  • sudomesh wiki

Some inspiration:

Babel

Lots of progress.

project management

let's try to use some software tools, max recommends Fulcrum

Remaining items undiscussed

  • Let's update our to-do lists to help people better plug into the project!
  • Let's create a spreadsheet of people working on people's open
  • Let's create an interest intake form for people to fill out on the website
  • Let's edit the website!!!!!!!!! It's not where it needs to be.
Retrieved from "https://sudoroom.org/mediawiki/index.php?title=Mesh/24_January_2015&oldid=9332"