Overview and Logs for the Dev Meeting Held on 2017-11-19

parent 4ef230df
---
layout: post
title: Overview and Logs for the Dev Meeting Held on 2017-11-19
summary: Discussion of open PRs and issues, Bulletproofs, Monero Research Lab, Monero integrations, dedicated Monero hardware wallet, multisig, and miscellaneous
tags: [dev diaries, core, crypto]
author: dEBRUYNE / fluffypony
---
# Overview
An overview can be found on [MoneroBase](https://monerobase.com/wiki/DevMeeting_2017-11-19).
# Logs
**\<rehrar>** Hey everyone! I don't know if fluffypony will be here today, but let's see about getting this thing started.
**\<rehrar>** Hello to everyone idling in the peanut gallery
**\<msvb-lab>** Hello.
**\<rehrar>** 2. Brief review of what's been completed since the previous meeting
**\<rehrar>** Can we get a mini-update from dev, gui, hardware, and mrl?
**\<sarang>** I can speak for MRL
**\<sarang>** We've been thoroughly investigated bulletproofs, a new form of CT range proofs
**\<hyc>** is moneromooo here?
**\<moneromooo>** Yes
**\<endogenic>** have one thing to bring up
**\<endogenic>** when there's time
**\<sarang>** They should function as essentially a drop-in replacement that is 10% the size of the original proofs (for one output; even smaller for multiple outputs)
**\<sarang>** with about the same verification time; possibly a teensy bit faster
**\<sarang>** Kudos to andytoshi for a lot of great help on those, and others
**\<hyc>** it's like a TARDIS! the bigger the input the smaller the output! /s
**\<rehrar>** So it's looking really promising that it'd be good for us?
**\<sarang>** yes
**\<sarang>** We'll start translating the Java test code over to C/C++ for more detailed work
**\<suraeNoether>** it's a win-win: smaller-spaced range-proofs, with the same verification time
**\<sarang>** There were some doubts about ver time, but we literally just had a convo over in MRL that pointed out a simple caching that helps us
**\<rehrar>** MRL, folks. Let's give 'em a hand.
**\<sarang>** It was pointed out, though, that there's very little randomness hanging around, which could affect storing stuff "in the randomness"
**\<sarang>** That may be a deepe discussion for later
**\<suraeNoether>** yeah, using these range proofs makes encrypting messages with range proofs far less do-able
**\<sarang>** \*deeper
**\<gingeropolous>** eh who cares
**\<msvb-lab>** MRL is excellent, nice.
**\<suraeNoether>** andytoshi is really the one who nailed it, btw
**\<sarang>** aye
**\<suraeNoether>** aaaaand all his co-authors
**\<rehrar>** Can we send them a fruit basket?
**\<endogenic>** +1
**\<suraeNoether>** which has included, for a second time, Jonathan Bootle
**\<sarang>** rehrar set up an FFS for it
**\<sarang>** 0.2 XMR
**\<suraeNoether>** and for the N^th time, greg maxwell
**\<rehrar>** 100 XMR requested
**\<rehrar>** biggest fruit basket ever
**\<sarang>** Anyhoo, that's the big news from MRL right now, I suppose
**\<rehrar>** awesome. Thanks so much.
**\<suraeNoether>** multisig is looking like i will be ready to hand it off to sarang by the end of the month, as hoped.
**\<suraeNoether>** and thats it for now
**\<rehrar>** Anyone else have an update since the last meeting two weeks ago?
**\<ArticMine>** This is huge from a scaling perspective
**\<hyc>** absolutely
**\<hyc>** dev has done a tone of PR merges latelt
**\<hyc>** lately
**\<rehrar>** can you summarize them?
**\<hyc>** there's still been build issues with readline dependencies but I think they're pretty much sorted by now
**\<hyc>** we've isolated readline from GUI dependencies, etc.
**\<hyc>** mooo can probably summarize better
**\<endogenic>** vtnerd says "oh good"
**\<endogenic>** "woo"
**\<moneromooo>** Hmm. I can go look at github:
**\<moneromooo>** (mostly small things anyway)
**\<rehrar>** FP confirmed he can't make it today. He sends his apologies.
**\<moneromooo>** disguising password length in prompt I guess is still a small thing, but user visible
**\<moneromooo>** less unnecessary locking in some DB readers, which should speed up some RPC
**\<moneromooo>** Fixes for BSD build
**\<moneromooo>** The wallet does not fetch unneeded hashes anymore (ie, starts direct from 1.4e6)
**\<moneromooo>** CORS (for those who know what it is)
**\<moneromooo>** A fix for show\_transfers missing more than one incoming input in a tx
**\<moneromooo>** And I think the rest is either prior to last meeting, or small/internal stuff.
**\<hyc>** yeah but it comprises a couple dozen PRs between then and now
**\<rehrar>** Alright, awesome. Thanks moneromooo.
**\<rehrar>** hardware, I know you give updates in community, but I think most of us that know are excited about the rapid progress you're making
**\<rehrar>** care to share something small about what's been going down?
**\<msvb-lab>** Hardware; we've settled our initial workflows and have built a twenty strong team. Journals are taking notice (see press) and the project is moving along.
**\<msvb-lab>** https://taiga.getmonero.org/project/michael-rfc-hwallet-1-implementation/wiki/press/
**\<msvb-lab>** I'll mark our forum proposal milestone 1 as concluded today probably.
**\<msvb-lab>** We now have an open call to any interested in a hardware wallet, become part of the project and possibly test a prototype. See:
**\<msvb-lab>** https://taiga.getmonero.org/project/michael-rfc-hwallet-1-implementation/wiki/preview/
**\<msvb-lab>** I'll demonstrate our two current working protoypes at this week's first ever Monero Munich meeting.
**\<msvb-lab>** That's all, any questions?
**\<sgp>** What do you expect those who receive the prototypes to test?
**\<msvb-lab>** Very weak expectations, hopefully each finds their own way to contribute and strengthen the project.
**\<msvb-lab>** There's a test plan page, where people can make comments while testing.
**\<msvb-lab>** But that is just one of many choices, each individual can make on their own.
**\<msvb-lab>** Pretty lame answer no?
**\<rehrar>** Alright, thanks msvb-lab!
**\<msvb-lab>** Thanks for the opportunity.
**\<moneromooo>** Sounds like pretty nice stuff is going on there, congrats :)
**\<hyc>** (I counted 72 PRs merged since last dev meeting)
**\<msvb-lab>** hyc: Who make most of the merges, fluffy?
**\<rehrar>** So, we've got Code + ticket discussion / Q & A time available for the devs to discuss dev-y things. But all the mergy stuff may have to wait because of FP absence, but if there's other business to take care of, devs you have the floor
**\<hyc>** yes he does all the merges
**\<hyc>** we didn't get a report from GUI
**\<rehrar>** oh, right! so sorry.
**\<rehrar>** Anybody working on GUI have stuff to talk about.
**\<unknownids>** ping Jaquee ?
**\<endogenic>** so i've been chatting with mooo, Jaquee, and vtnerd recently about integrating MyMonero with the native mainline libwallet
**\<Jaquee>** sorry. i'm on the run
**\<endogenic>** signs point to integrating with wallet\_api
**\<endogenic>** I may end up tackling a small portion of factoring wallet2
**\<rehrar>** ok, no worries Jaquee.
**\<moneromooo>** As long as not too many gratuitous changes ^\_^
**\<endogenic>** 100% mooo
**\<hyc>** all variable and class names being filtered thru pig-latin
**\<gingeropolous>** dsc, in here :)
**\<dsc>** :)
**\<msvb-lab>** Is our GUI crew considered only monero-gui, or are mymonero.com and Moneruj (Android) included as well?
**\<dsc>** has it started?
**\<endogenic>** lol hyc
**\<rehrar>** I think GUI crew is people who work on the monero-gui repo
**\<rehrar>** mymonero and Monerujo are their own thing
**\<rehrar>** dsc has what started?
**\<dsc>** the meeting :D
**\<rehrar>** oh, yeah. We're about half hour in. :)
**\<hyc>** oh - that's worth mentioning too - the repo is now properly named monero-gui, no longer monero-core
**\<msvb-lab>** rehrar: So Monerujo communicate on a different channel, right?
**\<rbrunner>** That rename is a milestone :)
**\<rehrar>** Monerujo utilizes the Mattermost where they have a private channel for the devs and core team, and a public channel for questions. They've also got a Taiga to track progress.
**\<dsc>** aight, so I'm working on a black theme for `monero-gui` (qt5) .. learning QML in the process. Trying to get some more UX developers involved and generally just trying to document things as I go
**\<rehrar>** I can rep Monerujo though, as I'm heavily involved with them
**\<rehrar>** they've got some exciting stuff coming
**\<rehrar>** yes, can you link your thread dsc?
**\<dsc>** sure
**\<rehrar>** The one you made a few days ago.
**\<dsc>** https://www.reddit.com/r/Monero/comments/7dvwq0/contributing_to_monero_as_an_ux_developer_short/
**\<rehrar>** in the comments there are the various proposals for redesigning the GUI. Are there any thoughts on any of those?
**\<rehrar>** well, first dsc you wanted to say something, go ahead. :)
**\<dsc>** No I was finished :P
**\<dsc>** but to answer your question: There are 3 UI proposals of which I picked the one that was most trivial to implement
**\<dsc>** seeing as I'm just one guy with zero QML experience, hehe.
**\<dEBRUYNE>** rehrar: Basically we're first kind of going for the dark theme, which will have a light mode as well
**\<dEBRUYNE>** Because it's most trivial to implement
**\<dEBRUYNE>** (only requires minor changes to the current design)
**\<rehrar>** https://github.com/monero-project/monero-core/issues/419
**\<rehrar>** https://github.com/monero-project/monero-core/issues/869
**\<rehrar>** https://github.com/monero-project/monero-core/issues/949
**\<dEBRUYNE>** Some screenshots -> https://i.imgur.com/89YFUj7.png
**\<rehrar>** for those interested
**\<rehrar>** Ok. Cool deal!
**\<dEBRUYNE>** To clarify, the dark theme is #949
**\<hyc>** cool
**\<dEBRUYNE>** dsc also made a PR which should make the GUI more responsive
**\<dsc>** cant make any predictions as to when I have some tangible, taking quite some time...!
**\<dsc>** s/some/something/
**\<dEBRUYNE>** Also, we're aiming to include subaddresses and multisig somewhere in the future^tm
**\<dEBRUYNE>** And, once there is a core appstore account, we'll start testing the ios port of the GUI
**\<dEBRUYNE>** There'll be an android port as well, fwiw
**\<msvb-lab>** dEBRUYNE: Is Qt5 easy to port to mobile?
**\<sgp>** Is there a new android APK to test? I last saw one months ago
**\<dEBRUYNE>** sgp: Buildbot produces APKs iirc
**\<dEBRUYNE>** Let me check
**\<pigeons>** No that isn't setup now
**\<dEBRUYNE>** msvb-lab: Jaquee did it, so better to ask him :P
**\<pigeons>** I can't promise next week even, very busy
**\<dsc>** rehrar: To give you an idea, right now I have this: https://i.imgur.com/iLwupdB.png
**\<dsc>** or to give anyone an idea for that matter :)
**\<serhack>** Hello
**\<rehrar>** looks pretty chill imo :D
**\<dsc>** aye aye
**\<dsc>** WIP ofc :)
**\<rehrar>** I personally think there needs to be a bit of a restructuring of some info, but this is a good meantime
**\<dsc>** agreed
**\<hyc>** good stuff
**\<hyc>** would be nice to get a more recent android build up for people to test
**\<rehrar>** Alright. That's it from GUI? No other opinions on dEBRUYNE and dsc stuff either?
**\<hyc>** looks like we can move on to ticket Q/A
**\<rehrar>** ok Code + ticket discussion / Q & A
**\<rehrar>** hyc moneromooo can you two take the lead for this?
**\<moneromooo>** Well, if anyone has questions about them, please ask.
**\<hyc>** I'd leave it to mooo. I'm not aware of any open questions
**\<hyc>** ^^
**\<dsc>** ticket Q/A \!= discussing open PR's, right?
**\<hyc>** sure
**\<hyc>** ask away
**\<dsc>** low hanging fruit PRs (that can probably be merged): #919, #954, #963 on monero-gui
**\<dsc>** not so much a question ^\_^
**\<hyc>** probably ask luigi to merge them if fluffy is unavailable
**\<iDunk>** luigi merges GUI PRs anyway :)
**\<dsc>** alright
**\<rehrar>** I think luigi is out for a few days, so be patient. (:
**\<rehrar>** Alrighty, shall we move on?
**\<rehrar>** FFS Formalization: https://github.com/monero-project/meta/pull/87
**\<serhack>** Is there meeting?
**\<hyc>** serhack started 40 minutes ago
**\<rehrar>** we probably don't have time for a large discussion (if at all), but I just wanted to encourage the devs to take a look at the PR
**\<serhack>** Oh
**\<serhack>** I forgot it
**\<rehrar>** The idea of having a set of formalized guidelines for FFS proposals going forward has been thrown around a lot. It'd be good to get some opinions on how this is laid out and any concerns that may arise from it.
**\<serhack>** I didn't understand well: is the project funded by ffs an Official project?
**\<ArticMine>** Does the current system work?
**\<rehrar>** There's been hiccups ArticMine. fireice and guzzi come to mind.
**\<endogenic>** those seem like complete outliers :P
**\<endogenic>** not exactly legit claims to issues imo
**\<rehrar>** Largely based on de facto expectations that are not actually formalized.
**\<msvb-lab>** How do we learn about the hickups mentioned (in order to understand context?)
**\<endogenic>** but proposals could stand to be more concrete, specific, detailed,
**\<endogenic>** rehrar: formalization can be a bit of a rabbit hole as well
**\<endogenic>** sometimes humans need to be involved
**\<endogenic>** ppl need to recognize the situation before a formal plan can be applied
**\<hyc>** indeed. current US political situation reflects this - institutions are designed assuming good faith in all participants
**\<endogenic>** but if the situation is not able to be recognized by ppl then how can it be applied for the desired result?
**\<hyc>** nobody can take all the time needed to identify every possible abusable loophole
**\<rehrar>** What we have currently is a Core Team centered way of resolving disputes that, to my knowledge, hasn't really been utilized since I've been here
**\<rehrar>** since we've got that right now, maybe we stick with it?
**\<endogenic>** but identifying loopholes and plugging them is not a bad thing
**\<endogenic>** it's just that they need to be done by humans imo
**\<ArticMine>** We may need some tweaking to deal with the hiccups; but as afar as I can see at a fundamental level the system is working
**\<rehrar>** don't know where else the graphic is, so I'll just link it here: https://getkovri.org/contributing.html
**\<rehrar>** bottom of the page
**\<rehrar>** This has not been formally replaced, so it is currently active. We can discuss that at a later time though. Let's move on.
**\<rehrar>** 5. Any additional meeting items
**\<rehrar>** If anyone has anything, speak up. :)
**\<moneromooo>** If anyone wants to test multisig, feel free. PR 2134.
**\<gingeropolous>** is there a set date for the next code freeze?
**\<serhack>** A little update by me about Monero Integrations
**\<rehrar>** go for it serhack
**\<serhack>** So we (I and cryptochangements) are going to release Magento Plugin, In the past few weeks, we worked on a new cool payment box for Monero Integrations.
**\<serhack>**
**\<serhack>** I am currently working on OpenCart too so I think we can release Magento payment box the next week..
**\<serhack>**
**\<serhack>** I have talked to OpenBazaar too and they have told me that Monero + Openbazaar would be nice.. it will be a difficult task but I think I can do it
**\<serhack>** you can go on, rehrar
**\<moneromooo>** Everytime I read OpenBazaar, it makes me want to punch that scammer from a few years back...
**\<rehrar>** alright, thanks man :)
**\<rehrar>** atrides?
**\<moneromooo>** That's the one.
**\<rehrar>** doesn't he still run Dwarfpool?
**\<gingeropolous>** yeah, and ppl still mine on it
**\<rehrar>** We should encourage them not to
**\<rehrar>** Either way: 6. Confirm next meeting date/time
**\<rehrar>** December 3rd? Same time?
**\<gingeropolous>** so no freeze date, eh?
**\<rehrar>** Should we move time because daylight savings?
**\<moneromooo>** About mid december
**\<gingeropolous>** word.
**\<ArticMine>** Please do not move the time
**\<moneromooo>** Or january. Unclear.
**\<rehrar>** ok, time not moved
**\<rehrar>** FP would have a better answer?
**\<hyc>** isn't a release due in January?
**\<hyc>** and we're supposed to freeze a month before
**\<iDunk>** March
**\<gingeropolous>** well dec or jan gives us 3 months
**\<hyc>** ok
**\<rehrar>** There's a new freeze/release schedule isn't there?
**\<hyc>** mebbe someone can open a meta issue and set some dates into it
**\<gingeropolous>** i thought we had 3 months of freeze for more testing / fixing than before ....
**\<moneromooo>** 3 months of freeze is gonna make me punch someone else :P
**\<moneromooo>** It's branch, then release from the brnch aftrer 2-3 months.
**\<moneromooo>** It's already painful when stuff doens't get merged after a month.
**\<ArticMine>** The big one is the bullet proofs
**\<gingeropolous>** right. i thought branch is a freeze... i.e., feature freeze.... i.e., freeze on adding wacky new stuff that requires bandaloots and warples
**\<moneromooo>** That's not likely to be ready by then.
**\<gingeropolous>** but yeah... perhaps that should be documented somewhere
**\<rehrar>** (Not to interrupt, but meeting is over for those who need to go, discussion can of course continue)
**\<rehrar>** Alright. You guys are the best! thanks for being a part of this community, and developing the software we all know and love as Morono
**\<msvb-lab>** rehrar: Thanks, good flow of topics, great moderation.
\ No newline at end of file
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment