# Wireless Challenge

https://wirelesschallenge.mozilla.org/

## Call

| Time | Location |
| ------------------------- | -------- |
| PST 3pm-5pm, Thurs 7th | Mumble |
| NZT 10am-midday, Fri 8th | |

Mumble setup:
- [installer](https://www.mumble.com/)
- ADDRESS: `celehner.com`
- PORT: `64738`
- choose push-to-talk

# agenda

Add to this

## introduction of people

What's your name and/or handle, which project are you working on and _very_ briefly: which aspects of the project are you working on and what excites you about this.

## introduction of projects

One person from each project intros the project, others can chime in but let's keep it brief.

- juul: sudomesh
- People's Open Network
- high bandwidth wifi based network
- 3-4 years old now
- a lot of sometimes involved (20), more core 8-10
- engineering
- community organisers
- to be independent from ISPs, compete with dsl/ cable
- DisasterRadio
- low bandwidth, solar powered network, long range (~1 mile)
- is more young (1-2 years?)
- relatively young project
- subset of people's open network folks (plus an extra person from sudoroom (Alex))
- cheaper equipment & easier installation; faster network growth than mounted gear
- https://github.com/sudomesh/disaster-radio

- scuttlebutt
- gossip protocol / database
- doesn't need a central organiser
- 'eventually consistent'
- you can build anything like social networks using it
- very mesh friendly
- no legal org structure

- mapeo / osm-p2p-db
- offline p2p mapping software & tech stack
- emulates the OpenStreetMap API, but uses an eventually consistent local backend
- sync map data via USB keys, LAN, internet
- https://github.com/digidem/mapeo-desktop
- https://github.com/digidem/osm-p2p-db
- related org: digital democracy
- environmental monitoring >> feed to governments to facilitate change
- http://www.digital-democracy.org/
- 5 folks / partner with ecuador indiginous non-profit name?
- mapeo principal project
- around for 5-10 years
- uses open streetmap
- sync's over any duplex stream

## relevant lived experiences

- mix
- I was in network breaking earthquake in Christchurch. Had to organise a highschool with collapsing radio / digital
- facebook most useful: able to post for 1:many communication (I'm alive; I am here; etc)
- phone networks overload quickly; SMS possible but unreliable
- Q: how did rescue teams communicate & coordinate?
- cell phone is the main communication tool on a person; need something low power that can send location
- ssb still works without a "backbone"

- cel - in remote location
- dominic
- lives on a boat
- designed around "internet as a privilege, not a given"
- apps that function offline /w periodic sync
- went to burning man last week
-

- juul
- inspired by what happened with arab spring
- see censorship as main disaster to address
- in copenhagen was unmotivated to start mesh because there was a new law about logging and invading privacy

- jenny
- helped out with tech crew at standing rock. many communication challenges - keeping wifi tower up in high winds and deep freeze conditions, sabotage of our ubiquiti gear around camp, keeping things powered consistently, DAPL hijacking our walkie talkie channels, etc - DisasterRadio would have been an ideal solution!
- experience around writing grants, coordination, documentation, budgeting, overcoming legal hurdles..

## reasons and ways to collaborate

Why are we doing this? Should we even be doing this? What are possible points of collaboration?

mixmix - have $2M to stimulate mesh + ssb + maps
juul - have grant be a validation and publicity rather than just money

noffle: what ties these projects together?
jorrit:
- locating places (map) where I can sync up my social media / blockchain / ipfs
- as I bike around, I'm syncing up
- mix: data mules!
juul:
- processer speed increase is plateauing
- data storage is not

dominic:
- taking something that can run on (radio), can fall back to wifi

LoRa - modulation scheme of radio, allows you to do transmision which are low power, long range. multi KB speed, multiple kms. LoRa gateways listen and pump into the internet. Used for sensor. CLOSED

LoRaWAN - https://lora-alliance.org
- star topology (nope)
- take the modulation scheme
- what's the part number of the lora chips you're using? RN2903 i think? (see https://github.com/sudomesh/disaster-radio / https://github.com/sudomesh/disaster-radio-nodemcu (older)) chur!

ip address, dhcp, web app can be whatever. Webapp
wifi > lora > over mesh
nodeMCU - uses lua, found memory constraints, found leaks, transitioned to using C. almost up to using chat

have a node server which simulates the chip, so you can dev with it.
Juul has offered to send out complete system.

mix: is the work for making dRadio ssb / mapeo friendly a job for people who write C
juul: we need to make a meshing (protocol). lora has such low bandwidth, you can do that over wifi / bluetooth, etc. Perhaps you only get the messages which you hear when you're only.

dominic: imagining implementing a pretty low level thing. perhaps ssb can make use of this, like realtime ssb messages can be sent. maybe a little bit of re-requesting dropped packets

juul: we could do some replay at low congestion times

jorrit: what are we trying to solve? are we solving for christchurch / standing rock / burning man ? would it be useful to flesh them out as user stories?

juul: I want to explore use cases, I'm wary of us exploring this outselves. We should really be talking with people and finding their actual needs
- Q: who are we making this for?

dominic: I had zero signal in oakland. there are holes in the bay area. +1 jorrit's point that we should be using our disaster gear everyday.
we are going to need to simulate the setup. like a map where there are limitations

juul: business model ...?

jorrit: I think do have some experience in our networks - people right next to us we can contact

jenny: designed use cases and user stories based on ethnographic interviews for a distributed mobile mapping application called Tidepools: http://wiki.tidepools.co/view/Ethnography/Interviews
- has already done some research which can inform our work

Dominic: what contacts do we have in the disastersphere or near there?

Juul: we could go the Tesla model - build fancy rich first.
- sell to disaster preppers in SF

Piet: how much actually needs to be built? can we pitch something theoretically feasible

Juul: first deadline is in Nov - a design, no implementation. 2nd deadline in June, a working prototype. This one gives 400K

Mix: alt market:
- farmers with low connectivity
- shipping ?
- satellites - how low is a low satellite?

- Possible communication channel for LoRa: https://en.wikipedia.org/wiki/Satcom_(satellite)

esp8266...

_Sidenote: NZ is pretty isolated from Silicon Valley thinking, e.g. BUILD BIG, build fancy, spend all the moneys_

juul: community layer, hardware layer, routing layer, application layer (eg, short text messages and mapping are initial ideas)

mixmix:

## challenge stack

- community-based research / story telling (same?)
- jenny, mix, sierk, nanomonkey, jorrit(use cases)
- coordinating writing proposal / design doc
- (dominic), jenny, nanomonkey, grant, sierk, jorrit (review?)
- grant administration / budget
- jenny, grant?, sierk
- outreach (getting folks to use it)
- dominic : development work in asia, friend with refugee-camps background
- noffle : unsure whether the groups we work with would use this; could ask more knowledgable DigiDem folks
- jenny: standing rock / water protector camps (as active sites for potential community use)
- jenny: i worked at standing rock with lisha sterling from geeks without bounds, would be a good org to collaborate with
- documentation
- juul?, jorrit?,
- hardware/prototyping
- robb, grant, nanomonkey, piet?
- firmware
- grant, piet?, nanomonkey, cel?
- routing (modelling/protocol)
- dominic, cel?, jorrit (novice)
- software applications
- ssb integration: dominic, mix, cel
- mapeo/osm-p2p-db integration: noffle
- data-mule / guided sneaker-net: jorrit, sierk
- mix:
- Q how is this different to the ssb integration? @jorrit
- A you need to know where to go to sync up
- noff: would this fit in as an alt routing layer? (ie. datamule OR LoRa?)

## action items
- juul: will write up some docs, with links, and send to the group
- dominic: hack on routing protocol
- jenny: send along sudomesh bylaws/processes for decision-making, conflict resolution, etc
- jenny: ~~set up a mailing list for project on the sudoroom mailman ? ~~
- cel: read disaster-radio repos. see if have esp8266; if so, try building and flashing disaster-radio firmware on it. might request hw from juul. tell #spiderfarm about project
- noffle: reach out to DigiDem folks to see if they're interested in collaboration /w existing indiginous groups
- grant: get a one-link chat app running
- dominic: talk to disasterradio (human) to see if they're cool with us using their alias

## next meeting
- juul will send out our standing group meeting times (sudomesh currently meets mon & tues evenings & sunday afternoons)
- noff: next meeting date/time?
- sync up on the #disasterradio channel on ssb

## interpreting the challenges

Official off-the-grid summary from website:

> How can we leverage wireless technology to keep people connected to each other and to vital information sources in the aftermath of a disaster where Internet access is unavailable or compromised?

> The Off-the-Grid Internet Challenge seeks solutions that can provide connectivity in one of the most challenging situations: the immediate aftermath of a major disaster. When disasters like earthquakes, tsunamis and hurricanes strike, communications networks are often among the first pieces of critical infrastructure to overload or fail, leaving individuals disconnected from one another and from essential services and communications. With a total of $1M in prize money, the Off-the-Grid Internet Challenge seeks solutions that help individuals wirelessly access each other and information services like maps and messaging following a disaster.

Examples from [official doc](https://wirelesschallenge.mozilla.org/_assets/NSF-OffTheGrid.pdf):

> A backpack containing a hard drive, WiFi router, computer and battery, which can provide disaster survivors in the immediate vicinity access to resources like maps and local messaging
>
> A drone-based network that can connect
community members and responders even if roads
are impassable due to earthquakes or flooding

## what we need to win (design)

The following is juul's suggestion. Not anything offical

- a filled out application
- collaborations with people who we intend to use these tools
- a nice crowdfunding-style video introducing the idea
- an easy-to-read writeup that conveys the same as the video but in more detail and links to even more detail
- something like a technical white-paper for each aspect of the projects
- write-up of two to three imagined case studies on how people would use it.
- conceptual graphics / mock ups of hardware and imagined use-case scenarios

juul: Suggestion: Let's BE BOLD! We can always under-deliver or change tacks somewhat in the next round. Let's catch their attention in the first round.

juul: Idea: Organize it in a way that allows the judges to start with a quick overview and make it easy to to drill down into increasing levels of detail

These are the things we are officially allowed to link from the application (one link her item):

- Public website/Documentation
- Code repository (e.g. public github page)
- Design/technical documents or schematics
- Video content (e.g. link to youtube video)

## who can take on which tasks?

juul: i suggest we set ourselves a hard deadline for Monday November 6th to give ourselves time to review and edit each-others stuff.

## how would we split/spend the money?

juul: The receiving org or person has to be U.S. based (or u.s. permanent resident). Suggest we pre-agree in writing how to split funding and pipe it into the sudo mesh non-profit and split it out from there according to agreement. If the paperwork doesn't come through then pipe it through Omni Commons and then into sudo mesh.

## do we meet requirements

juul: looks like it

# reference

## deadlines

- Intent to apply: 15th of October 2017, one second before midnight, pacific time.
- Design application: 15th of November 2017, one second before midnight, pacific time.
- Prototype application: 22nd of June 2018, one second before midnight, pacific time.

## documents

[Application guide + application form](https://assets.mofoprod.net/nsf/NSFWINSapplicationguide.pdf)
[rules](https://wirelesschallenge.mozilla.org/rules-and-regulations/)

## chat with mixmix

<mixmix> Hey Juul, do you know all the people coming tmrw (apart from me and Piet?)
<juul> yep. i think the rest are all from our group

<juul> So we have two projects: One is the People's Open Network, which we've been working on for ~3 years. This is a wifi based long distance last mile (we like to call it "first mile") replacement for shitty IPSs using IP protocol and mesh routing to bring 10 - 1000 mbps to people's houses. The second project is very new and is called DisasterRadio. It is the one suitable for this challenge and is low-bandwidth solar powered, no wires anywhere, easy to deploy, etc. etc.

<juul> Most of the people coming to the meeting are from DisasterRadio AND People's Open Network but a few have not engaged with DisasterRadio yet and are mostly just going to be listening.

<mixmix> makes sense in the USA
<mixmix> awesome, have you played with bluetooth mesh yet ?
<juul> no
<mixmix> AHH, Dominic has a friend called DisasterRadio (I'm sure he's told you)
<juul> I know the (single person) band which was totally the inspiration of the name. Did know he was from nz but not that Dominic knew him. We might change the name, I just picked one to get started though I'm kinda hoping to get disaster.radio but who knows.

## Actions

- Dominic to troubleshoot people ssb problems
- Mix - talk to someone about dev setup
- Nanomonkey - intersted in how we mesh different systems together
- Jenny ?
-