Overview and Logs for the Dev Meeting Held on 2019-06-30

来自: el00ruobuob / rehrar

Logs

<rehrar> hey guys
<rehrar> it's time for a meeting
<rehrar> as always, we'll try not to drag and make this longer then it needs to be.
<rehrar> 1. Greetings
<sarang> Hi
<rbrunner> Hi!
<vtnerd__> Hi
<kinghat> shalom
<rehrar> Alright. 2. What's been completed since last meeting.
<rehrar> Anyone have an update on stuff?
<rehrar> dsc_ selsta dEBRUYNE for GUI people?
<rehrar> moneromooo TheCharlatan CLI?
<dsc_> thecharlatan working on reproducible builds for GUI
<rehrar> fluffypony luigi1111 smooth for Core Team?
<dsc_> I'm working on better tails integration
<dEBRUYNE> GUI v0.14.1.0 is around the corner
<rbrunner> Ah, that corner?
<dsc_> I've updated https://autonode.xmr.pm/ to show more remote nodes
<moneromooo> I did some more work on share-rpc, seeing someone's started reviewing.
<moneromooo> It would be nice if more people reviewed, and if someone did something with it
<sarang> moneromooo: do you consider the CLSAG implementation branch suitable for PR/review (not to merge yet, just for review)
<moneromooo> Yes.
<moneromooo> It doens't have your latest changes though.
<sarang> Right, and I plan to update 5707 (and its CLSAG equivalent) again soon
<sarang> those are pretty minor overall
<moneromooo> Then it might not be be ready.
<luigi1111> I don't have an update
<dsc_> thanks for the update luigi1111
<rehrar> luigi1111: you can update us on the brand of soda the core team drinks as the relax in the lounge
<dEBRUYNE> With respect to a timeline for CLSAG, is October too optimistic? We must also take into account third party wallet providers, which will have to make changes too
<rehrar> thanks everyone for the updates
<rehrar> dEBRUYNE: this is going to be unpopular to say, as it's been discussed before, but if we're "right on the edge" with CLSAG, then why not just push the fork back a month?
<sarang> dEBRUYNE: No final word from potential reviewers yet… I reached out again
<sarang> So I do not have a timeline for CLSAG review
<rehrar> 25% savings in ring sigs is kind of cool and nice to have.
<dEBRUYNE> rehrar: So push the October fork back to November?
<rehrar> if it'd give us the breathing room that would make enough people more comfortable with the timing, then yes
<sarang> Why not wait until spring?
<rehrar> Ah, wait no. we want RandomX in ASAP, huh?
<vtnerd__> This is 25% savings just for the ring sigs portion, not the entire tx, right? I don't see the reason for the push
<rehrar> so pushing it back maybe isn't great
<sarang> vtnerd__: a 2-2 txn shrinks by 25% overall
<sarang> not just sigs
<sarang> (ring sigs themselves shrink by approx 50%)
<rehrar> oof
<rbrunner> My gut feeling tells me that RandomX and those CLSAG thing together will be a bit too much …
<vtnerd__> Hmm need to read the paper then, didn't realize it dropped that much
<sarang> you save 320 bytes per spent input
<sarang> and about 20% on sig verification time
<vtnerd__> They are separate things entirely
<vtnerd__> They are either ready independently or not
<sarang> And to clarify, there's working code already that anyone is free to review
<sarang> (verification code will be tweaked a bit still)
<vtnerd__> Yeah the issue was a math review though … ?
<sarang> Yes, and a formal code review
<sarang> But getting early internal review would be useful
<rehrar> even though waiting on bulletproofs was a good idea, it was still painful to have that six months of big big transactions
<vtnerd__> Oh you wanted both for this too? Ok
<rehrar> that we forever carry
<sarang> vtnerd__: it's important enough that I think both math/code review are important
<sarang> esp. since MLSAG never got a formal audit
<vtnerd__> I mean its either that or risk an entire blow up of the coin
<vtnerd__> In response to rehrar
<dEBRUYNE> sarang: I'd be fine with spring, that at least gives third wallet providers plenty of time to work on it
<luigi1111> I think spring yes
<luigi1111> for clsag
<rehrar> yes, but my suggestion was pushign back a month for audits
<rehrar> which lessens chance of coin blow up
<sarang> Well, I can let everyone know when I hear back from our potential reviewers
<sarang> OSTIF are also putting out feelers
<dEBRUYNE> <rehrar> yes, but my suggestion was pushign back a month for audits <= Can you elaborate?
<dsc_> Has anyone heard from fluffy regarding GUI release?
<dsc_> Has he arrived home yet?
<rehrar> if someone could see into the future, we can both see the outcome of the audits as well as save money by not paying for them
<rbrunner> If we wait with CLSAG until spring, will there be time to build something nice on top of it until then in addition?
<sarang> Also: earlier internal code review may reveal bugs that we can fix before sending code off to the reviewers
<sarang> rbrunner: ?
<rehrar> dEBRUYNE: it's just pushing this fall fork back one month to give some breathing room to get some CLSAG audits in
<rbrunner> Like those famous atomic swaps, or whatever
<rehrar> but as I said I see reasons not to do that. It was more in response to vtnerd__
<dEBRUYNE> dsc_: https://www.reddit.com/r/Monero/comments/c6y542/any_news_on_the_gui_release/esc02my/
<sarang> DLSAG's key image issue makes it unsuitable for implementation just yet, IMO
<sarang> CLSAG is basically ready without such (known) problems
<dEBRUYNE> rehrar: Seems like a bit of a slippery slope
<dEBRUYNE> Seems safer and more prudent to wait six months
<dEBRUYNE> In the grand scheme of things six months isn't that much anyway
<rehrar> I don't disagree, dEBRUYNE
<rehrar> proving alternative viewpoints
<rehrar> although how long six months is in tech and blockchain relatively is much bigger
<sarang> It's pretty moot at this point anyway
<sarang> Until we hear about audits
<rbrunner> Well, I think that "blockchain relative time" has slowed considerably lately
<rehrar> true. So maybe move on.
<sarang> It's a moo point. Like a cow's opinion
<rehrar> rbrunner: not with fireice and ryo nipping at our heels.
<sarang> Doesn't matter
<rbrunner> Lol
<rehrar> Pretty soon we'll be doing coordinated, organized attacks out of fear
<rehrar> alright, moving on
<rehrar> 3. Code/Ticket discussion
<rbrunner> I have used the Go RPC bindings: https://github.com/monero-ecosystem/go-monero-rpc-client
<rbrunner> My 2 PR's to improve on that were just merged today.
<rbrunner> Can confirm that the bloody thing works.
<dEBRUYNE> rbrunner: Have you spent any work (or chatted with the team about it) on OB lately?
<rbrunner> Yes, and the ball is definitely in their court now: https://github.com/OpenBazaar/openbazaar-go/issues/1638
<moneromooo> That sounds like a lot of nice work there.
<rehrar> much applause!
<rehrar> Is that something Revuo worthy, you think?
<rehrar> Go RPC stuffs?
<rbrunner> Ah, hmmm, maybe not yet the OpenBazaar stuff, quite early yet.
<rbrunner> RPC is of course ok to mention, maybe other people will use it
<rehrar> Alright, if there's nothing else we can move on
<dEBRUYNE> Nice work rbrunner
<rbrunner> Thanks!
<rehrar> I'm assuming in "Old Business" we don't need to continue discussion on Payment IDs?
<dsc_> very nice rbrunner
<moneromooo> Only if there are new arguments.
<rehrar> going once
<rehrar> going twicee
<dEBRUYNE> rbrunner: "and no support for moderation" <= Is that even feasible at this point?
<rehrar> hearing none, I think we can move on from PIDs
<rbrunner> Doubtful, unfortunately.
<rehrar> Ok, any other meeting items?
<rehrar> Just as kind of announcement. People going to Vegas for Defcon, some of us are going a couple days early (starting monday the 5th) to hang out and chill. All are invited and welcome.
<dEBRUYNE> rbrunner: I guess they could apply a similar process as bisq for moderation
<dEBRUYNE> <rehrar> I'm assuming in "Old Business" we don't need to continue discussion on Payment IDs? <= I think the rough consensus was that we should start with a full software removal first
<dEBRUYNE> (please correct me if I am wrong)
<rbrunner> Don't remember exactly, isn't Bisq multisig-based as well?
<dEBRUYNE> Partly (for the BTC part)
<rehrar> dEBRUYNE: every core team member that spoke up (in the issue) was very against that. Smooth, ArticMine, binaryFate
<moneromooo> I'll keep the parsing code with an opt-in flag in monerod I think.
<rbrunner> to hang out and chill <- is that even allowed for Monero people?
<dEBRUYNE> rehrar: No?
<rehrar> ohhhh wait
<rehrar> never mind, I'm speaking of PID removal in general
<rehrar> not long
<dEBRUYNE> They were against parsing tx_extra / temporary ban
<rehrar> I got confused cuz you said "software removal 'first' "
<rehrar> and thought you were hinting at the reversal later
<rehrar> my bad
<dEBRUYNE> Hmm no
<dEBRUYNE> As far as I know, no one was against a full software removal
<rehrar> cool
<dEBRUYNE> There were people opposed against (i) temporary banning the tx_extra field, (ii) permanently banning the tx_extra field, (iii) parsing the tx_extra field to disallow long payment IDs
<rehrar> then we gucci
<sarang> Define "full software removal" for clarity, plz?
<sarang> As in, no GUI option to add one for outgoing?
<rehrar> as I understand, Isthmus says he has some very interesting research going on about "treasures" found in th tx_extra field
<rehrar> I'd like to see that research when it happens, and we can discuss the pros and cons afterwards
<rbrunner> Huh? Really?
<moneromooo> In fact, I already made that code. I have shitloads of stuff that's not being PRed just because it conflicts due to merging being stalled now
<rbrunner> That does not sound too good …
<dEBRUYNE> moneromooo: Do you have a list for luigi that he can merge?
<moneromooo> Yes.
<dEBRUYNE> Could you post it here too?
<moneromooo> Sure. One moment…
<dEBRUYNE> sarang: removal from both CLI and GUI
<sarang> With no option to enable via flags?
<moneromooo> 5647/5666, 5650/5651, 5663/5664, 5668/5669, 5675/5676, 5678/5684 (there's more)
<dEBRUYNE> That was my idea kind of
<moneromooo> and 5690/5694, 5681/5708.
<sarang> Note that this could cause some exchanges/services to recommend specific wallets (that do continue to support) to their users
<sarang> which could be good or bad, depending on the wallets
<dEBRUYNE> That's a potential risk, yes
<dEBRUYNE> I deem it more likely that they will simply switch though
<moneromooo> I could keep the –enable-paymend-id-bad-for-privacy, and instead of enabling them, it prints "lolno".
<dEBRUYNE> moneromooo: Will send luigi the list in PM as a reminder
<moneromooo> Oh I did
<rehrar> dEBRUYNE sarang this is all one big experiment. I'd say let's try it this way and see how the exchanges react
<moneromooo> I just don;t want to annoy him too much.
<rehrar> if they don't do as we hope, then we learn from that next time we have to make a decision like this
<rehrar> but this is all hypothetical at this time. Let's see what happens.
<rbrunner> I really doubt that the exchanges have time on their hands to work against the community, but who knows
<rehrar> well, we do have LiveCoin
<moneromooo> If they recommend other wallets, I won't have any regret in breaking those in next fork ^_^
<rehrar> either way
<rehrar> 4. Any last meeting items?
<sarang> 5707 speeds up MLSAG, and will be sped up a bit more
<moneromooo> Does anyoine want to review share-rpc ? Or did I mention that already…
<sarang> review will be welcome
<dEBRUYNE> rehrar: Should we perhaps discuss v0.14.1.1?
<dEBRUYNE> <rehrar> dEBRUYNE sarang this is all one big experiment. I'd say let's try it this way and see how the exchanges react <= Yes
<dEBRUYNE> To be fair, it is mostly the big ones that are remaining that we need to get on board
<rehrar> moneromooo: I'll put it in the REvuo volunteer opportunities (for all the good that will do)
<moneromooo> AFAIK it's waiting for a freebsd patch from TheCharlatan.
<dEBRUYNE> Bittrex, Bitfinex, and Binance
<rehrar> moneromooo: can you PM me a link to it
<vtnerd__> moneromooo: I will look at it this week
<moneromooo> https://github.com/monero-project/monero/pull/5357
<moneromooo> Thanks
<rehrar> thank you
<rehrar> dEBRUYNE: what do you want to discuss regarding 0.14.1.1?
<rehrar> the floor is yours
<dEBRUYNE> Timeline kind of, what do the devs prefer?
<dEBRUYNE> We can move a bit faster now that we have deterministic builds
<moneromooo> As soon as the bsd patch is in.
<moneromooo> (and the patches above)
<rehrar> I may be getting confused because of the numbers, but didn't pony do builds already?
<rehrar> or this is new? what's going into it?
<moneromooo> The BSD patch and the patch list above.
<rehrar> ah, k
<dEBRUYNE> moneromooo: All right. Does this need a release v0.14 equivalent? https://github.com/monero-project/monero/pull/5705
<dEBRUYNE> rehrar: new release
<moneromooo> I dunno, ask TheCharlatan about this one.
<dEBRUYNE> All right
<rehrar> ok, anything else to discuss about this release?
<moneromooo> The GUI I suppose ^_^
<rehrar> by all means, moneromooo. Take it away.
<moneromooo> I don't have anything to say about it.
<dEBRUYNE> GUI will follow CLI for v0.14.1.1
<dEBRUYNE> First have to wait for pony to finish the v0.14.1.0 builds though
<moneromooo> There are missing builds for .0 ? I didn't even realize…
<rehrar> pony is traveling again
<rehrar> he does that
<dEBRUYNE> Yes
<dEBRUYNE> Also we kind of had to retag, which has delayed the release a bit
<moneromooo> Gonna run out of places to go soon.
<rehrar> he's running from the community methinks.
<vtnerd__> Me? Lol I'm in a car on lte so I dropped service once and hit leave once stupidly
<dEBRUYNE> rehrar: He provided an update yesterday
<dEBRUYNE> Anyway, vtnerd__ I wanted to ask you if you had already started some work on dandelion++ ?
<rehrar> dEBRUYNE: link?
<dEBRUYNE> https://www.reddit.com/r/Monero/comments/c6y542/any_news_on_the_gui_release/esc02my/
<rehrar> alright, if there's nothign else, I think we can call it here.
<rehrar> discussion can obviously continue after the fact
<vtnerd__> I started to look into it. At this point my ffs pt 2 is hopefully going to make that transition easier (but it won't be a complete d++ implementation)
<rehrar> Two weeks from now is the 14th of July. Same time.


Post tags : Dev Diaries, Monero Core, Cryptography