Mesh/13 Feb 2018

From Sudo Room
Revision as of 20:36, 14 February 2018 by Gcg (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

People's Open Network Weekly Meeting - 13 February 2018 7:30pm-9pm


  • Introductions and Meeting Roles (10 mins)
  • Check in on previous action items (5 mins)
  • Updates & Reportbacks (20 mins)
  • Sundays (5 mins)
  • Proposal: Alameda Open Network Proposal (15 mins)
  • Proposal: Semi-synchronous Chat Options (10 mins)
  • Discussion: Preparation for the stategic plan meeting (10 mins)
  • Discussion: PON Home Node and Exit Node Management (20 mins)
  • Action Items (5 mins)
  • Breakout Groups


Introduce yourself: Name; Preferred Pronoun; Icebreaker Response; BRIEF Announcements

  • Icebreaker question: Favorite candy
  • mai - she/they, favorite candy kit-kat, no announcements
  • jorrit - he/him, flashback to when he was 8, winegums? and Coke.
  • bengo - he him, they/them, special edition Y2K sweethearts
  • JJ - he/him - candied ginger
  • benny - he/they, likes sour patch watermelon gummie
  • sierk - chocolate cookies
  • Robb - he/him/they - candied ginger, SKOR bar
  • Grant - he /they - mentos -
  • jenny - she/her, not a candy person, dark chocolate, any annoucments, nope
  • Ben - he/they, dark chocolate, go to meet with Jeff tomorrow at 1pm (s of campus)
  • jehan - he/him - jelly beans, wants to help with tunnel

Meeting Roles

  • Facilitator/s: Mai
  • Stacktaker: Robb
  • Notetaker/s: Benny, Grant, Jenny
  • Next meeting's facilitator(s):

Check in on previous action items

Check previous meeting minutes at and past action items below:

  • create meeting agenda template (eg ) and archive nov-feb notes - jenny
  • re-ping lolspace & respond to julia (AJ+ contact) - jenny
    • pinged devie a couple of days ago to see if they would be interested, requested to check back in March, location is close to village, sped up timeline to be part of al jazeera feature
    • got positive reponse from AJ+, hoping to hear back from lolspace
    • jorrit - thanks, for listening to comments from last weeks meeting despite not being here
    • no reply from Devi yet (lolspace) - forwarded Julia's response to sudomesh signal group (folks seemed positive)
  • nyc mesh meetup - jehan
    • the meetup was in a bar, back patio, apparently their meetup schedule is not that regular, they have some but they are for particular purposes, this was a more general public meetup. The event started small, but grew and then turned into a panel discussion with the members of NYCmesh, they have a lot of inbound interest, a lot more than they can deal with. They don't look at individuals, but try to hook up entire building by talking to landlords. They are using a firmware, old custom build of QMP? SO they have trouble updating it, but it does work. They don't pay for bandwidth, they have a gigabit, average 50mbps with peak of 300mbps. They have ~1/3 or ~1/5 chance that people could get internet access. Using primarily Ubiquiti AirGrids

Also hungout with some people in brooklyn who's internet access is through NYCmesh. Bandwidth just from one node in Manhattan. 5-8 inbounds a week -> 5-8 inbounds a day after press.

    • jorrit - what wold you copy that they do there?
    • jehan - could be wrong about assessment, but having a high up node that can be connected to easily seems to have been very helpful to them. everyone who mails in, they reply, can you see the verizon building from your apartment? okay we can do it. [we have a harder process]
    • mai - do they have any mesh service?
    • jehan - there's a guy running a wiki from his house
    • previous meeting w/ nycmesh:

Updates & Reportbacks


Priority bugs we need help with are located at: - For a high-level overview of recent Github activity, see the Trello:

  • see helpwanted link
  • bug #8 - debugging exit node,
    • jorrit - seems to have figured out how to set up an exit node (in Germany!) poked around the actual exit node, saw some stuff that could be fixed
    • jorrit - feel like I'm looking at a house burning down
    • jorrit - will talk more about operations later
      • to fix bug #1, need to fix bug #8. made a quick script for creating exit node, want to explain to 3-4 other people. wants to do a breakout.
    • grant - I really like "setting up an exitnode" as a sunday skill share. props! wondering--if a node is already up and never gets reset, how can you get root access once it's on their private internet?
    • jorrit - need to reproduce it
    • jenny - yar(sp?) says jorrit *does* have root access to the mesh exit node
    • Bengo - are the trello links supposed to be closed?
    • jenny - invite only yeah, pasting link.
    • jorrit and Bengo [exit node explanation]
    • jorrit - want to dispute netflix idea that netflix is blocking peoples open traffic
    • grant - idea: other bugs, #2 and #13, can be fixed by documenting makenode
    • robb - recalls someone mentioning netflix doesn't allow vpn traffic
    • jorrit - need to test
    • jehan - no im positive that's it
    • jorrit - wants to see proof
    • jehan - could ask netflix, hey this is not a vpn ur confused; work with us

Node Mounts & Outreach

Node Mounts Trello:

  • ben - going to jeff's place tomorrow, if anyone would like to do a weekday mount?
    • sierk - grant and I are gonna do a survey at Ed's place. Also found another location within LOS to Driveway Follies, also talking with Biblioteca near LOLspace/Village

Communications & Events

Comms Trello: // Events Trello:

  • mai - AJ+ already discussed
  • jenny - january 11th, people wanting to start mesh network in LA. who wants to follow up?
  • sierk - i could follow up--asked the person who asked me about it to e-mail, but they never got in touch

Internal Logistics / Finance

Internal Logistics & Bureaucracy Trello:

  • INSURANCE: Will & Jenny have been trying to get a quote for Directors & Officers insurance. Last reply was 1/25 from Tracy @ Brignole & Ghiri: "I have not been successful in obtaining a quote for the Directors & Officers Liability coverage. I have either been declined by markets due to the business operations or a minimum premium of $15,000. I do have a request for the following:
    • Jenny sending CT-1 initial reg form to the Attorney General tmrw. Once approved we get a # and then have to file RRF-1s probably for 2014-2017.
    • Business Tax exemption - jnny can't remember if she did it form mesh when she did it for omni & sudo. will show up on the radar if/when we start making $$. will x-check
  • FINANCE: $1,187 in bank acct.
    • Mozilla requesting a W-9 - jenny will send tmrw w/ CT-1.
    • Still requesting equipment budget requests! Please add to the notes.
    • Setting up proper accounting this year - ping jnny if you're interested in helping/learning.
    • jenny - needs dollar amount from jorrit for cow/chicken reimbursement
    • bengo - what's better than a spreadsheet
    • jenny - would like to setup a quickbooks online
    • jorrit - what's your plan to help others to help you? Oracle of All Things Complicated
    • jenny - could send out a doodle
    • mai - something we could discuss on Saturday (strategic planning meeting)


Node Mount

We try to arrange a rooftop node mount every Sunday afternoon. Ongoing spreadsheet is located at:

  • Next Sunday's mount:

Office Hours

We host office hours every Sunday from 1-2pm in Sudo Room. To sign up for a shift, fill in your name here:

Proposal: Alameda Open Network

grant talked with Amos White this afternoon

  • he is proposing to head up a large-scale, pratical roll out of a People's Open Network in Alameda and would like to hold workshop(s) very soon (he suggested starting Sat. Mar. 3) on how to go about doing this.

could sync up well with on-going outreach in East Oakland and

  • grant - I wanna know what people think of doing Alameda workshops with amos--practical stuff, getting/setting up antennas
  • jorrit - as you were talking, I was seeing them coming here to get trained (as opposed to us going there). can slowly segue into bigger event.
  • mai - do they have an internet access point? were they hoping we would provide that?
  • grant - i explained that he would have to look into what places in Oakland he had LOS to
  • robb - diane's has a perfect shot at alameda and richmond
  • jenny - still need to power the node
  • mai - is there anything we can do regarding the richmond tower?
  • jenny - the holdup is only that we need to coordinate with Jonah(sp?), probably buy new equipment to put up there, and do a site survey. they have a second tower that might have better LOS to the south. I've been putting in on the agenda, but I wasn't here for a bit.
  • mai - I think this would be a great thing to discuss at the strategic plan discussion. It's a great path forward.
  • jenny - I brought up determining a peering agreement to solidify our partnership--his only concern was that he didn't want to get tied into anything too serious in case they have to back out of the agreement
  • jorrit - we had a long discussion about that, and i think we decided that there would be no (peering) agreement
      • Agreement was to not have any formal agreement
  • jorrit - re: Amos, first thing is to get him familiar with how stuff works. that can happen right away. e.g. by the time he has two nodes up and running, our exit node should be resolved.
  • robb - they should do a BYOI (in alameda)
  • jehan - if they can/want to pay for an uplink, they might be able to get an uplink through digital realty
    • right across from Digital Realty which is part of SFMIX (SF metropolitan internet exchange)
  • jorrit - i want to make sure we don't scare anyone away with our own fears (e.g. bugs we are fixing that they may never have to face)
  • ben - [questions about whether trees are blocking LOS from Diane -> Alameda]
  • mai - agree w/ jorrit that best way forward is to invite Amos to BYOI office hours. It's awesome that he sounds so enthusiastic.
  • grant - I also suggested office hours! (high five). I also liked robb's suggestion to get Alameda people organized. I tried to make it clear that we can't just hop on over and set everything up. But we can provide tools and knowledge so you can hold your own workshop. Can organize people in Alameda who are interested, bring them here, and we can train them about how to have their own workshops.Going to shoot him an e-mail tomorrow. Will reference notes.
  • jehan - is it safe to delete the email server?
  • jenny - that's the contact info for mozilla

Proposal: Semi-synchronous Chat Options

  • Rocket.Chat at (NOT YET SECURE and maybe up/down, but will have demo running Tuesday night)
    • daniel & grant worked on it a bunch on sat, trying to set it up on a raspi
      • now set up on a DO droplet running ubuntu.
  • Let's Chat - had trouble running on sudomesh server
  • Mattermost - haven't tried
  • Matrix - haven't tried
  • sierk - what will it replace?
  • jenny - IRC, info@ emails
  • jehan - how does it compare to matrix?
  • grant - dunno--rocketchat had clear docs.
  • ben - am happy w/ mattermost at my home. it seems pretty similar.
  • x - also recommends Discord, for voicechat
  • Consenso? Do we wanna go ahead with another chat platform?
    • benny: get it minimally solid for us to try out
    • sierk: try it for a few months then we know what features are missing
      • we're-a test it out for awhile "get on there"
      • "if you don't wanna try it maybe you can get in there & let us know" - jehan

Discussion: Preparation for People's Open Strategic Plan Meeting

  • How should we get input from people who can't attend the meeting? In order to also inform what our community would like to see out of PON.
    • Suggestions: Send out a survey before Saturday
    • proposal: mai - draft a form to send out tomorrow afternoon
      • inviting suggestions for what should be on survey, will post to signal group

Discussion: PON Home Node and Exit Node Management

Discuss/share current state of, and ideas regarding: patching home nodes, maintaining exit nodes and monitoring network health

  • jorrit - the reason for bringing this up is to get some ideas from the group on how to move forward on managing the network we love so much
    • there are two bits to it, the managment of the deployed nodes (home and extender) and management of the exit node
    • would like to first share how we do it today, right now we have one active person who has access to ssh and patch nodes
    • some months ago there was a the KRACK attack, that one person jumped up and fixed all the nodes
    • the exit node, there is are maybe one of two who are familiar with, jehan is one, no he isn't, ok there is one. Yar has access, but doesn't know what is running.
    • mai - was the propsal at the last update that you would spread that knowledge?
    • jorrit -that was the technical part, this is the organizational part
    • sierk - very relevant, mentioned something similar to grant, we get to the state that we have a lot to maintian and that people rely on, have a few volunteers who are active maintainers of services and repos. suggest we discuss this on saturday - point people for specific aspects
    • jorrit - for this discussion I wanna talk specifically about a plan of action once the old exit node goes up--do we patch the existing nodes?
    • jehan - we have nothing to lose
    • jorrit - but the one person who has acces to do it.
    • jenny - it's obviously problematic to have only one person with access to do this. it's currently the case because earlier, there was only one person sufficiently paranoid, competent, and gung-ho to do this. now we probably have more people who can have this responsibility. we need to come up with personal security protocols for system administration.
    • jorrit - what I'm hearing from you is that we're okay with remote management of home nodes (if people consent)
    • grant - current state of network is remote patching with opt-out--you need to know that you can opt-out before you can opt-out. this is a bug in previous firmware. i vote for, with new firmware, you need to opt-in to being open to remote fixes/updates.
    • jorrit - what I'm hearing from grant is that you want explicit opt-in.
    • jenny - I also like this.
    • jehan - right now there are people out there w/ open wifi hotspots that are using their home internet connection. we should definitely remote in to fix this.
    • grant - this is related to the "document makenode" issue. I'm documenting makenode right now so it will be easier to make these firmware changes we're talking about.
    • mai - seems like if we have a mailing list of people who have nodes, then we should let them know about the bug (if they're providing open wifi access to their internet) for the sake of transparency. in that same e-mail, we should let them know that we're working on this issue, and we can help them fix it themselves, or we can fix it for them remotely. Reaching out to them is in the nature and spirit of this project. Letting them be empowered to fix it themselves, or choose to let us do it for them. That's my communication suggestion.
    • sierk - Similar to Mai, but more long term--are we going to discuss this elsewhere, because there are some details... it will take some time. new mailing list?
    • mai - definitely should discuss on saturday
    • ben - very little to lose by giving other people agency about the bug. more of a problem if people don't know and something happens.
    • jenny - only issue i see is that we do not have a mailing list of node operators.
    • jorrit - sounds like we have consensus about what to communicate but not how. I stil want to get consensus on whether we should proactively remotely patch things if we can.
    • grant - bug number 1? yes we should remote patch. but we should discuss whether to remote patch all other bugs.
    • mai - are we going to say that we're going to fix it proactively?
    • jorrit - i want to separate (1) that we are going to proactively patch and (2) that we are going to send an e-mail to operators
    • grant - we did this in september--marc got some, didn't get others. in the past i have sent an e-mail to the list.
  • Possible Plan: Get exit node up, fix the code, update firmware on existing nodes,, then:
    • update firmware with opt-in process for remote maintenance
    • establish protocol for those with remote access

Should we remote-patch Bug #1?

  • Mai: Does this entail we let them know beforehand?
    • Jorrit: No, we'd just patch it. Send an email.
  • Aye: Grant, Jenny ??? we are evidently unable to move onto decision-making-moment :P :)
  • Nay:
  • Abstain:
    • ben - how long does a mount take?
    • robb - it's usually like 6 hours. could probably do it in 3.

related topic: makenode documentation/deprecation

(announcement): Anyone wanna help document makenode? see - also

Action Items

  • Send CT-1 and W-9 - Jenny
  • Follow up on D&O Insurance - Jenny, ?
  • Reply to Julia / AJ+ - Jenny
  • Move emails to new Migadu server - Jenny/Jehan
  • Get secure and persistent on sudomesh server - grant
  • Draft group consensus and send to Amos White - grant
  • Send survey to mailing list ahead of strategic meeting - mai
  • Weekday node mount with Jeff 12pm Wednesday - Ben, mai, grant

Breakout Groups

At the end of the meeting, break out into small groups to hack on or discuss specific topics.

  • Exit Node Debug: Jorrit, mai, grant, bengo, benny

Last Meeting Notes

End of Meeting