開發者會議記錄 DevMeeting 20180909

    • 由於與表定分叉更新的時間已非常接近(約於20181018),因此下次的版號將直上V0.13.0.0而非原規劃的v0.12.4.0

    • 最新的PoW已更動為CryptoNightv2,需要一些會C++的開發者幫忙審查

    • Kovri 的程式碼已正式從GitHub移至Gitlab,而主要的Monero 程式碼尚未決定是否要移動

      • 很可能以Gitlab為主而Github作為鏡像備份

      • 也可能Monero就直接用Gitlab的軟體自行架站

      • 防彈協定(BulletProofs)與CryptoNightv2程式碼的合併並於testnet上運行預期會在9月10-11號開始

      • 下次開會時間為2018年9月16號

以上翻譯來自於紀錄原文: https://monerobase.com/article/monero_devmeeting_2018-09-09

完整會議記錄:

<rehrar> Hello everyone. Dev meeting time.
<rehrar> 1. Greetings
<rehrar> anybody out there today?
<vtnerd> yes
<rbrunner> Hello!
<oneiric_> hi
<jtgrassie> hey
<rehrar> 2. Brief review of what's been completed since the previous meeting
<rehrar> What's been happening?
<rehrar> lel
<rbrunner> The blockchain grew :)
<rehrar> Xeagu increases the blockchain size with his mass every week now
<rehrar> slow week then? :)
<jtgrassie> Kovri alpha yet?
<rehrar> Kovri has had an alpha release since just before Defcon
<rbrunner> Well, what did not happen is 0.12.4.0, and FluffyPony confirmed yesterday that it definitely won't happen anymore
<rbrunner> Next ist 0.13.0.0
<jtgrassie> Makes sense
<rehrar> ok, so all of that stuff is just being rolled into 0.13.0.0?
<rbrunner> Looks like it
<rehrar> Was there anything critical there?
<rehrar> and did he happen to give any indication on the status of 0.13 when he spoke?
<rbrunner> No, just a short confirmation what we already suspected over in moneru-gui
<rbrunner> We have a hardfork date of October 18
<rehrar> so I hear
<jtgrassie> Latest PoW change still seems under discussion
<rbrunner> Well, a review would be nice, but hard to get by the necessary people with the knowledge ...
<jtgrassie> I'd certainly welcome a vtnerd going over it
<endogenic> not necessarily
<endogenic> there are toonnnnss of c++ and hw devs in the world
<endogenic> we need better outreach to devs specifically
<endogenic> things like dev meetups for c++ people
<endogenic> i'll start going and making announcements
<jtgrassie> agreed
<endogenic> "good opportunity to have your work recognized and make a big impact"
<rehrar> "Change the world: Monero"
<jtgrassie> we don't really know if a change is even needed for next HF
<pigeons> let people building asics spend/waste a little more money first
<endogenic> or "dont change the world: it was better when money was backed by something" lol
<jtgrassie> I was refering to change of PoW, not the world - which yes needs changin!
<rehrar> Alright, well, anything else happen this past week?
<rehrar> Maybe not to be discussed in too much detail, but anonimal started the effort of moving off of Github
<rehrar> kovri is now on gitlab
<vtnerd> jtgrassie : yes I plan to scrub over it again, and have again somewhat now
<rehrar> I know there was discussion of doing that with Monero in general
<rbrunner> Which did not get very far, didn't it?
<vtnerd> the sqrt stuff is the one iffy part where my eyes gloss over a bit
<pigeons> moneromooo is setting up a gitlab instance
<+moneromooo> Well, I was going to, but I didn't know anonimal did too. I thought he moved it to gitlab.com.
<+moneromooo> Unless that's what he actually did.
<rehrar> I think he is on gitlab.com, yes
<pigeons> Yes I think that's what he did (move to gitlab.com)
<+moneromooo> OK, then I'll do it.
<jtgrassie> vtnerd: cool. What I liked about your last April changes is they were clear and simple. This PR is big.
<rehrar> Monero doesn't necessarily have to be in the same place as kovri though?
<+moneromooo> No. If he wants to move kovri there, he can. If not, not.
<+moneromooo> It'd probably be less confusing to have it all in the same place though.
<jtgrassie> I'd agree
<rehrar> Alright. Whatever you guys decide.
<@ArticMine> The question in my mind is how many developers and contributers to Monero support Microsoft
<pigeons> what do you mean by "support"?
<oneiric_> hopefully few
<jtgrassie> I'd guess very few
<rbrunner> With "support" you mean "would vote to stick with Microsoft"?
<@ArticMine> I mean would want to continue using a Microsoft platform
<@ArticMine> Yes
<rehrar> I don't say this often
<rehrar> but Microsoft is not bae
<@ArticMine> So I suspect anonimal's move is just the tip of the iceberg.
<oneiric_> I like that anonimal ripped off the bandage, and moved to gitlab
<oneiric_> gets the ball rolling
<pigeons> there does seem to be general rough anecodatal "consensus" among the more active contributors I see that they would prefer to move away
<oneiric_> even if not final solution
<rbrunner> Thankfully, technically, it does not matter much
<rbrunner> Just linking to earlier things will be hard of course ...
<pigeons> there are decent import tools in gitlab recently improved
<@ArticMine> pigeons that is my read also
<jtgrassie> I like the idea of GitLab as the primary repository and GitHub a mirror
<endogenic> i thought it was already decided to do that
<jtgrassie> That means anything that cannot be imported still has the legacy links
<jtgrassie> endogenic: yes that was also my understanding
<rbrunner> May I say that we "decide" in confusing ways sometimes ...
<rehrar> decide here means that everyone agrees it's a good idea, and nobody gets it done
<rehrar> Monero has cool word definitions like that
<@fluffypony> I think the main thing is self-hosted GitLab vs gitlab.com
<+moneromooo> \o/
<@fluffypony> and we're largely leaning towards self-hosted
<@ArticMine> Yes
<rbrunner> Makes sense, IMHO, if move then move right
<rbrunner> Right being self-hosted
<jtgrassie> but then someone has to maintain it and get called an ameteur when it goes down ;)
<rehrar> I'm guessing we don't have an ETA on this though?
<+moneromooo> None.
<rehrar> purrrrfect
<+moneromooo> Since pony's around, I want to talk about when to fork testnet whenever convenient.
<oneiric_> is there a way to fund the server with FFS and have a group of trusted people with admin creds?
<rehrar> take it away
<oneiric_> my bad
<rbrunner> take away what?
<@fluffypony> moneromooo: is there anything preventing us from doing it tomororw?
<rehrar> Oh, I thought moneromooo was going to ask some questions from pony
<@fluffypony> *tomorrow
<rehrar> that's what I meant take it away (to moneromooo)
<@fluffypony> rehrar: he already did
<rehrar> oh
<+moneromooo> fluffypony: (1) merging the bulletproofs, and (2) maybe merging CNv2.
<@fluffypony> we're getting dangerously close to when we should release 0.13
<+moneromooo> (2) being the annoying part here, since we can wait forever with no guarantee we'll get a review :/
<jtgrassie> 2 seems premature
<rbrunner> Why only "maybe" for CNv2? Test the damn thing :)
<@fluffypony> jtgrassie: we've got a fork coming up in 5 weeks, I don't think it's premature
<+moneromooo> Because if it changes, we'll need to reorg testnet again. maybe not too much of a bother I guess.
<oneiric_> looking forward to OhGodAGirl's review
<+moneromooo> So I'd be fine merging 1 and 2 for now, and reorg on monday or tuesday if that's alright ?
<@fluffypony> let's do Monday, we can always re-reorg if we need to
<jtgrassie> fluffypony: fair enough, but we don't even know if it's needed really.
<+moneromooo> And btw for those who don't know, it'd be *very* nice to have everyone running the release by the 11th, due to https://github.com/monero-project/monero/issues/4287
<+moneromooo> OK, monday then. I'll fixup the testnet reorg kludges.
<@fluffypony> ok cool
<jtgrassie> and by needed I just mean the CNv2 change
<iDunk> Just a reminder that the testnet will be rolled back to a height where nethash is ~10 kH/s.
<jtgrassie> this CNv2 PR needs much more review given the amount of changes
<+moneromooo> Oh, yes, forgot this one ^_^
<rbrunner> jtgrassie: It's only testnet :)
<rbrunner> And if it gets used maybe bugs will surface earlier
<rehrar> moneromooo when you say it'd be nice to have everyone running the release, do you mean devy people or everyone everyone?
<jtgrassie> true, I'm just paranoid with large changes!
<+moneromooo> Everyone.
<+moneromooo> (if they care about DNSSEC)
<rehrar> alright, I'll see about coordinating something with the Community and Outreach workgroups
<rehrar> I think it'd be beneficial to do it anyways so we get as few "support" questions as possible
<rehrar> Alright, anything else?
<pigeons> luigi1111: fluffypony maybe look into merging https://github.com/monero-project/monero/pull/4342 which fixes windows build
<rbrunner> yeah, windows build working would be nice of course
<iDunk> And 4352 fixes linux builds.
<rbrunner> Anything that does not need fixing? :)
<+moneromooo> Bulletproofs hopefully.
<rbrunner> Is this the time and forum to make a final decision about the name of the release?
<rehrar> Alright, well, pending other topics of discussion, I think we can decide on next meeting date and adjourn.
<endogenic> do we want to review the bulletproofs changes since the audit?
<endogenic> i know ops are confirmed as constant time etc
<+moneromooo> Well actually they might not be constant time, interestingly.
<oneiric_> when will quarkslab report be released?
<+moneromooo> Oh, kinda forgot about that one.
<+moneromooo> After 0.13 I guess, they found a DoS which was supposed to go into 0.12.4.0.
<oneiric_> Ok, thanks moneromooo. When is 0.13 planned?
<+moneromooo> See above.
<rbrunner> About that release name...? Beryllium Bullet, Beryllium Bullet Cluster, or something else? Or not the time and place to decide this?
<endogenic> so.... bulletproofs...
<oneiric_> saw it being discussed, but no date. unless it was the "monday or tuesday"?
<rehrar> rbrunner: it was being discussed in the Monero forums, let me grab the link
<+moneromooo> That's the testnet reorg.
<rehrar> https://forum.getmonero.org/6/ideas/90722/release-name-for-oct-0-13-0-hard-fork
<rbrunner> rehrar: Yes, thanks, saw it there as well, but that's again a question about deciding ... right?
<oneiric_> so past oct 18 for a report on bulletproofs?
<rbrunner> Would be nice to have the name, to prepare docs, ReadMes, the installer, etc.
<endogenic> maybe we can discuss release name after we confirm if bulletproofs needs to be reviewed again
<endogenic> if not let's just put it live on mainnet now. yolo
<rehrar> endogenic: because you want Berylium Bulletproofs?
<rbrunner> The review would be for the changes made since the reviews, right?
<endogenic> nah it's for the name
<endogenic> just kidding, yes
<oneiric_> moneromooo: if quarkslab found a DoS in bulletproofs, why does that hold the release of paper until after they're implemented?
<+moneromooo> It was not in BPs.
<rehrar> ah, so we waiting for get the next release out.
<oneiric_> understood. Is there anyway to remove that section, and release an amended version of the paper?
<rehrar> So if we're not goign to decide the name now, and you want to be a part of the discussion, it'd probably be best to post on that forum. I think there was a reddit thread too. Otherwise keep an eagle eye on the IRC channels for when it will be discussed
<+moneromooo> stoffu: I've added 5 days' worth of blocks to the testnet bodge in the BP PR if you want to OK again ^_^
<endogenic> rbrunner: my reply was to you fwiw
<rbrunner> Yes :)
<+moneromooo> oneiric_: maybe, notmy call though.
<oneiric_> no worries, thanks moneromooo
<endogenic> can we ask them?
<endogenic> i guess sarang is the point of contact
<endogenic> afk
<oneiric_> ok, I can ask during tomorrow's MRL meeting
<rehrar> ye
<rehrar> alright guys, I guess that's it.
<rehrar> September 23rd sound ok? Or as we approach release, should we have it every week?
<endogenic> not it
<endogenic> i will ask again about bp change review at mrl mtg tmw as well but same thing will happen
<endogenic> now afk
<oneiric_> thanks endogenic
<rehrar> I will tentatively set a meeting up for next week, and it can be canceled then. So the 16th. Unless there are any objections.
* Tom_Cruise ([email protected]) Quit (Ping timeout: 252 seconds)
<rehrar> It can be super quick, just touching base on release stuff
<rbrunner> With a meta entry so people don't forget ...
<rehrar> indeed so!
<rbrunner> Nice :)
<rehrar> alright, that's it then. Thanks for coming everyone

發表迴響