monero-site/_posts/2017-11-19-overview-and-logs-for-the-dev-meeting-held-on-2017-11-19.md

18 KiB

layout title summary tags author
post Overview and Logs for the Dev Meeting Held on 2017-11-19 Discussion of open PRs and issues, Bulletproofs, Monero Research Lab, Monero integrations, dedicated Monero hardware wallet, multisig, and miscellaneous
dev diaries
core
crypto
dEBRUYNE / fluffypony

Overview

An overview can be found on MoneroBase.

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.