Skip to content

Latest commit

 

History

History
231 lines (154 loc) · 18.2 KB

2018_01_15.md

File metadata and controls

231 lines (154 loc) · 18.2 KB

Software Project Meeting Agenda Jan 15 2018

NOTE: Importance and order of agenda items decided prior to or at the beginning of meeting by participants.

Agenda summary

  • Developer meeting rhythm
    • regular time
    • meeting tech
  • how does dans reflection fit in here?
  • rails project status
  • sprint planning process
  • what are we doing right now?
    • sharding keys
    • scuttlebutt

Minutes:

  1. Developer meeting rhythms

shall we do regular or often

propose Developer meeting

  • datetime: Monday 8pm UTC
  • frequency: weekly
  • where: mumble
  • [name=dan] says: this time might be a bit tricky for me for the next week or two as I need to drop nova to playgroup which starts at exactly this time. So for the first few weeks i'll be a little late - BUT thereafter can help get childcare to cover that time.

actions:

  • Kieran someone check in with Dan ([name=dan] says: lets do this on mumble as a way of ensuring we get it set up and eek out config)
  • Mix communicate Mumble setup
    • if someone non-tech wants to join, Mix can support them to join
    • [name=dan] says: I just set up mumble and spoke with [name=mix] . managed to get in running in less than 5 mins. was a bit of configuring needed though, so lets each connect BEFORE the next meeting so we don't have a configuration time sink...(setup on thinkpad running devuan.org)
  1. Dan's reflections

who's read dan's reflection?

what did people take from it:

  • [name=mix]: I'm not clear on what's next
  • [name=peg]: seems like a shame there's a Rails app, but we're reconsidering (e.g. the single admin system)
  • [name=nikolai]: I think single admin is a massive red herring. Changing it wouldn't take very long {[name=dan] asks, not sure what you mean by this [name=nab]? can you clarify?}
  • [name=peg]: seems like the bit that might need changing is the organisation outside the Rails app (like how people take money out) {[name=dan] says: yes, in part. How money leaves the system needs some thinking. I am of the opinion that we don't deal with fiat at all... but this is not my main concern with a single or multiple admin system. The issue is more to do with the framing of responsibility. I think we need a complete overhaul of that... more comments on this below}
  • name=kieran: why are we all here together, what do we think will be useful for people? How would the Rails app be useful if at all? What's our target market - who or what are we trying to build for? helping refugees is defined problem, but building a crypto exchange could be a totally different thing
  • [name=peg]: there aren't any FOSS exchanges run by coops, or exchanges with flexibility about identity proof, which I think excludes a lot of people {[name=dan] says: you can't run a non-p2p exchange which DOES NOT require ID of some sort. regulators will come for you. p2p tech is releasing sofware, so does not need to deal with this... I would suggest this as the ideal approach}
  • name=kieran: i think the Rails thing would need a bunch of polish to be accessible, but I'm up for it if it's the right thing to do. Building something where you can shard keys would be cool, but could be useful as a group wallet, and that could be useful in something like the political/ greek project
  • [name=nikolai]: I'm trying to get (child) to sleep and failing! I'm quite sad that we've built stuff that we're now considering not using it.
  • name=kieran: I've put a lot of effort in and don't really want to chuck it away. I'd like to have a better understanding of how it might be used by people. It feels like it's a good base and it could go in lots of different ways. Knowing how it's going to be used is v important.
  • mix: Agree! Edcated guess at a domain. Solution looking for a cause. If its for refugees, how do we validate whether its useful? If its for the greek project, lets spend a month trying to work with that team.
  • name=kieran: the political team have found it hard liasing with the general dev group.
  • [name=mix]: We're looking to Dan for direction and he's giving autonomy and voice and clarity about some things and clear invitatios to improvise. I want to try doing crypto in primary schools, is that a go or nogo? Or definnte for refugees and the details of whther we do sharding or not is my domain? What are the bounds of the project?
  • [name=peg]: people who otherwse wouldnt have access to crypto -> focus on the margins. Education element to project.
  • [name=nikolai]: I have trust misplaced in anyon in MMT to give guidance. Project loking for purpose hit nail on the head. Happy to build and not necessarily to be part of decision but constantly moving boundaries is frustrating and not a particularly positive experience. We need clarity!
  • [name=nikolai]: Lack of clarity is the struggle. We've built what we'd decided before but now its important to keep momentum.
  • [name=nikolai]: Personal realms of activism don't feel like necesariy impact on the same as the poltical project.
  • [name=mix]: Agree, not so much the same domain too. If we're not coding but for a month we're liaising with the greek project,
  • [name=nikolai]: we'd like to do something with crypto for people who are undocumented to help. Lets get an idea together and then I'll jump on board. When we've got a concrete idea, I'd contribute hapily
  • [name=peg]: To summarise dan, creating multisignature wallets and using ...

From the political proj:

"What we have been focusing on here in athens lately is how to develop IOU crypto-loaded cold storage wallets – old phones basically - on mobile devices to give to people on the move – refugees and asylum seekers. What do you think about this idea? The pre-loaded IOU funds would be rotated from one person to another as they grow. Ideally this would tie into mmt, or cryptokitties, or whatever we end up calling this project, but also it doesn't have to, we can develop this here ourselves. Building something robust enough that all our funds could circulate in would be great, but its imperative that people manage their own keys, or wallets and therefore their own identities and lives."

  • name=kieran: I think all the tech they need already exists, it's about the configuration. Not sure if it's a software or a social solution.

  • [name=peg]: does this need a large team of devs

  • [name=nab]: I'm not sure this is a skillset I have, it looks like a thing that's supporting people to use existing software

  • [name=mix]: I think we should:

    • EITHER: tell Dan what we want clarity about so we know if we're each a good fit
    • OR: propose to Dan some ideas we think are exciting / strategic

New question (mix)

Hopefully to prompt us to consider what direction we want to work in:

What do you want to build, why? (something exciting / strategic)

  • nikolai: using the rails app that is already built to replace the spreadsheet. Because we already build it and would be exciting to use it. Its a good way to test it using an existing system. I dont want to be involved in money-related software untill its really tested - i couldnt sleep!

  • mix: the spreadsheet use is dwindling, how many users are needed?

  • nikolai: the spradsheet has haemorrhaged users... The sooner the better

  • [name=peg]: I started writing a script to parse CSV data, I wasn't sure how to test if the spreadsheet can be compared to the Rails app to test if they're equivalent

    • nikolai: we double handle transaction on csv + rails for a month to compare

  • nikolia: also keen to follow through on our decision

    • we decided last meeting to actively run a test using the real data.

  • name=kieran: I think there's a tension in that Dan is keen ot not maintain the csv

  • [name=peg]: interested in seeing this software used, and share the responsibility. I notice all crypto exchanges are having a hard time given the value is shifting. There's also how you manage people rushing the bank... this is a responsibility which needs sharing

  • [name=peg]: it's not clear to me to what extend this will be relying on Dan. I think he was imagining not to have to move the actual crypto much (internal banking), I guess that's how exchanges work

  • [name=mix]: any groups you;d want to see this used with

  • [name=peg]: my friends. some of my friends are refugees. anybody really

  • [name=peg]: Dan posted up Bisq (a p2p exchange). Imagine something like that but more or an extending group of friends.

  • name=kieran: thinking about what to build / why. The rails project and it's data structure has been really interesting. I'm concerned about running a live server when dealing with peoples money. I'd like to think we have the skillset to manage this, but not sure we do.

  • name=kieran: I'd like to work on the Rails project, I'm also interested in learning about the possibilities of Scuttlebutt as a way to exchange value, shared wallets etc. No single point of failure

  • name=kieran : these are different projects which have diff learning opportunities

    • rails : security and how to manage centralised app, expand knowledge of rails + ruby
    • scuttlebutt : electron app, working out mechanisms that could exchange value p2p rather than via server

  • [name=peg]: I see a p2p exchange eliminates some of the dangers of running a normal exchange. Might be a different way for people to find one another. A platform itself which doesn't

[name=dan] : Hey folks Dan here. Anyone read the mvp thread on loomIO https://www.loomio.org/d/izHukXb7/-mip-002-updated-mvp-towards-multi-sig-sharded-social-backups-and-a-p2p-exchange-coordination-space# ? Happy to use rails system. Not happy to be the single admin. Fund is now worth millions of others peoples money. It's killing mY juju

[name=mix] : a couple of people, not all yet. My main question was (from a surface understanding) was "this seems like what we could build, I think we need more clarity on why".

--discussing how meeting will go futher...

keiran: we are going down the right road in terms of picking out project ideas we like

mix: i like documents, its hard to have a single doc with all these ideas would be better to combine with loomio thrread where ppl can post comments. using text to communicate is good for code.

mix: we could do more than 1 call a week to understand faster?79* mix: im busy working on scuttlebutt. there is no mobile app at the moment but there is one in development -so for refugees in the short term its not a good option but in the future it could be cos it is not dependent on internet.
mix: needs to leave unfortuanelty cos weve still got a lot to talk about

who's up for another call? e.g. weds 8pm UTC ?

name=kieran I'm able [name=mix] yep [name=peg] ja [name=nab] ~maybe have to check

peg and keiran carried on talking for a bit

  • kieran has come up with another approach to using the spreadsheet data. we talked about putting into place nab propsal

[name=dan]: another approach which might work for more folk is that we use what has been built but start small. all of us here have some crypto. why don't we loan out small amounts to a new round of people. this puts less funds at risk... i am happy to put down funds for this. more "greenfield" BUT, i don't wanna be the single admin, someone else is more than welcome to be :) i guess this might alreay have been suggested?

[name=peg] we were just talking about making things more p2p in terms of the exchanging -actually most exchanges are matching 'buys' with sells but we are not doing this. erm do u wanna move to keybase?

[name=dan] : IMHO rails/ruby is excellent for prototyping. 100% we should use what has been built. I am not keen to use it for the existing users. they never signed up to be tested on. we can however test with us and people we know where they sign up for the experimental nature. I guess we prototype in ruby rails as it is quick. get ideas for the p2p. i think other people will find single admin systems are wrong. the overall idea is a good one. but we need better foundations. once we have an idea of how to do better foundations we can expand into p2p. we might find p2p is the foundations we need. IMHO sharded multisig is a foundational requirement.

[name=dan] I don't think we should be thinking about setting up a business. if we are all honest no one wants to run a startup. we can build code which is used by folks but this is different to running a business. Sorry for spill. internet scarcity...

[name=dan] what I am hearing is that nab wants the work that has been done to be used. I 100% agree. think we should roll out a live prototype with new users and it be fresh. I am also happy to run an offline version in tandem witrh spreadsheet and give feeback that way. How does that sound?

[name=peg] Grand. I think nicolai is concerned about the responsibily of people using it before it is tested and thats why hes pushing for a the spreadsheet to be modelled. Maybe we can find another way to test it???

[name=dan] ^^^ we test smaller :) we start with smaller balances. Still live. But they can be our balances. I am happy to throw some hundreds at it etc Lets start a new spreadsheet which isn't worth a fuck tonne of money :) lets start smaller (like it used to be before crypto went bananas and we were startin gthis out...)

[name=peg] that sounds encouraging i think everyone will feel happy. maybe we can start playing around with the heroku version this week.

[name=dan] Having played the role as admin I think it is worth everyone having a go at this. Be the admin for some people around you. I think we can eventually design in a way where the admin role is replaced by clearly defined relationships. mmt, participant and auditor (or something like this) and control of the wallet is managed through clearly defined relationships through a multi-sig.... in this way we acheive the same result, onboarding new folks into crypto without it costing them but we do it in a way which doesn't emotionally and psychologically/emotionally cost a single group/person... No such tool exists in this form to make this happen... it is also a totally new subjectivity for grassroots folkx to be doing with crypto... so - i don't agree that it's just about teaching people the tools. The tools don't exist. The building blocks do, but most normal folkx aren't software builders - that's where we come in.

Good ideas... looking forwards to start experimenting. [name=peg]

// Pre-meeting notes

[Name=Dan] Process note. My Internet in Suva is nearly unworkable alot of the time. My laptop drops off Internet every 30secs. Means need to try using phone for e wrything. Basically I don't have practical conditions to really be able to lead IMHO. Async contrib to things yes. But not lead. If that makes sense.

Organisational

Loomio Reflection and Review

There has been no real relationship between (Political Project) and (Software Project) https://hackmd.io/s/rJekfy-GM#observation3

  • I would like a pipeline with the political project having some easy (and maybe mid term but no long term yet) defined goals in priority order and for these to be fixed for a period of time so the software project can try to build them. I find it very dishartening, but understandable given our proto nature, to build something that is not used. [name=nikolai]

  • Dan has circulated an email to [name=nikolai] and kieran about Political Project Starting point [name=dan] https://www.loomio.org/d/izHukXb7/-mip-002-updated-mvp-towards-multi-sig-sharded-social-backups-and-a-p2p-exchange-coordination-space

  • Please also note that the reflection piece is in Draft mode, so has a lot of bugs. Is inaccurate of me to have put that there is no relationship between the Political Project and Software Project. The two are related & the software project is working on code which has political ramifications. It would have been more accurate to say that there has been a seperation of processes. [name=dan]

Rails Project

Current Features

  • spreadsheet parsing - what does it need to show in order to prove that we have a working, robust transaction storing system? [name=peg]

  • Equity - Fed up of looking at javascript Math.round...... Some help finishing this would be appreciated as I can't get my head around whats required to get the numbers to work. (kieran)

  • Docker - Where are we at? Still needs sidekiq setup in docker-compose.yml. Then can we merge? name=kieran

  • Heroku Bugs - 2FA needs correcting (outstanding issue in Github) name=kieran UPDATE: [17/11/18] name=kieran this has been resolved and pushed to Heroku.

  • Withdrawls - recent update to PR name=kieran

Proposed Features

  • Complete market history of coins from start date of spreadsheet name=kieran

Project Proposals

  • Methodology / space for creating project proposals name=kieran

  • Dan has suggested MIPs [name=dan] https://www.loomio.org/d/tB0ODXdn/-mip-001-mmt-improvement-proposals

  • Using dyne.org Secrets for keys!? [name=peg]

  • Idea - Electron app for sharding private keys (secrets) and providing access to collective/group wallets through secure scuttlebutt. name=kieran

    • Can this be done? Would it be useful?
    • Scuttlebutt is about to get loomio + doodle, already has meeting booking! [name=mix]

Organisational Experiment

Comments

I propose we use this hackmd page to take notes during meeting to post to loomio afterwards [name=peg]

Good idea! Collaborative minutes! name=kieran