Monero Dev Meeting Note Highlights – 2018-02-11
Logs
351 lines
rehrarArticMine luigi1111 luigi1111w fluffypony smooth hyc moneromooo anonimal vtnerd pigeons endogenic ErCiccione
vtnerdpresent
ArticMineHi
msvb-mobHello.
ErCiccionehi folks
vtnerdjtgrassie
sgpHi
rehrarAgenda: https://github.com/monero-project/meta/issues/174
rehrarJaquee medusa dsc
rehrarkenshi84 MoroccanMalinois anyone else?
MoroccanMalinoishi
rehrarWe've already kind of start on 1. Greetings
rehrarIf you're still hanging out in the peanut gallery, throw a hello on by to us.
rehrarHeck, if you don't plan on participating, but just want to watch, we'd still appreciate a hello. :)
VespcoHello
endogenichello
endogenicthis is dog
rehrarAs you guys probably see, it's not 17 UTC. We're trying out an hour earlier by request of last meeting. We can decide today whether it was a good idea or not.
pigeonshi
rehrarsuraeNoether sarang
rehrar2. Brief review of what's been completed since the previous meeting
rehrarAnyone have anything to report?
moneromoooDB crash fix. Misc minor fixes found by coverity. Pool max size. Fluffy blocks on by default. I think that's the main ones.
moneromoooAnd I just PR'd my "store prunable and unprunable data separately" patch.
moneromoooOh, and the –testnet-xxx options are gone now. No more duplication.
moneromoooNone of that is merged yet, but nothing got merged in two weeks I think.
rehrarThanks moneromooo! You are a shining star. Take a bow.
rehrarAnyone else have anything to report? If not we can move on.
moneromoooNot all by me.
vtnerdI guess I will report for jtgrassie, who looked at the echo off permanently issue with the wallet
rehrara bow for your presentation ;)
vtnerdhe has a fix that is similar to moneromooo 's patch, but does trigger immediate shutdown on ctrl-c still
luigi1111Does that need a resync?
vtnerdthere are still underlying issues in the signal handler, and if I can't figure out a way to unbreak that code easily we may have to go with the somewhat-hack approach until that gets worked out
iDunkluigi1111: It will convert the db from v1 to v2, AFAIU.
moneromooono resync
luigi1111Perfect
vtnerdand I'm still interested in wallet scanning perf (as usual), and figure out how to do atomic swapping … if anyone has thoughts on the latter let me know privately for now
vtnerdor what this is the whats been completed section, whoops
endogenici have a question for later rehrar
rehrarok
rehrarAlright, we ready to push forward?
sarangHullo, sorry for delay, was filing taxes :/
rehrarAny dev stuff to report for the last two weeks?
rehrarsarang \^
sarangIt's just about the most fun you can have while still wearing pants
sarangYes, a few quick thing
sarangThe BP paper was massively updated to reflect a lot of the optimizations we've been including
sarangI'm running final checks against that paper (and made a few corrections to the updated paper that have been sent to the authors)
moneromoooOh good point, I forgot about more BP stuff ^_^
sarangI had also worked up a BP technical note, but the paper update makes that obsolete
endogenicmooo same
sarangHowever, I think we've made all the optimizations that we want
sarangOn the audit front, OSTIF groups are slow to provide SoW but apparently that's happening early this week
sarangBenedikt Buenz (BP author) wants to review as well
sarangThoughts on how much to offer Benedikt?
sarangHaving both him and an audit group review will be good: he'll confirm our math, and the audit group will check our final code for holes
rehrarMake him an offer he can't refuse
saranglol
sgpI don't really know what the right amount is
endogenicno pony heads though pls
sarangI'm thinking a rate of 800 USD per day would be competitive if it only takes him ~5 days or so
sarangHe's the most qualified person to do this
endogenicthen that's cheap
endogenicwe're lucky
ErCiccionethat's less than i expected
sarangYeah, I'd go higher if no objections
sarangI asked what his rate is, and am waiting to hear back
ArticMineIt is on the low end
endogenicwell ErCiccione do keep in mind we benefit from addtl reviewers
endogenicso frugality is useful there
rehrar20 XMR bam
sarangSo I'll do a combined FFS once we've confirmed the OSTIF rate and Benedikt's rate
endogenicthanks sarang :)
sarangSepararely, I've been working with suraeNoether on some multisig developments from BPASE
sarangcurriculum development for 3-week summer course
sarangWill be speaking at a Portland crypto conference about privacy (funded by the organizers, no FFS needed)
sarangand the paper on PHANTOM (related to SPECTRE) just came out, so I'm on top of that
sarangis done talking now
rehrarwoohoo, sounds awesome
sarangquestions welcome
endogenicawesome
sarangread the new BP paper if it interests you
rehraralright let's jump into the big one: 3. March hardfork items + code freeze
rehrarSo… code freeze has been "close" for a couple months now
rehrarwhen is it? :P
moneromoooWhen we pushed all we need,
moneromoooWe need pony back, so we can't discuss this if we don't know that.
moneromoooOne thing we need to discuss before we do though:
moneromoooWe discussed changing PoW slightly, and periodically, in order to further our goal of decentralization - to deter ASIC creation some more.
moneromoooSo we're planning to change PoW before the fork, if the community agrees.
rehrarbefore March?
moneromoooYes. We're still planning on a march fork.
pigeonsi would expect changing the PoW before this fork to be at least slightly controversial
rehrarThis seems a bit sudden and soon
moneromoooEmphasis on slightly btw. It's still mostly Cryptonight.
rehrarah, makes more sense
sgpI would like to hear a lot more about the changes and what they mean
rehrar\^
rehrarcan you explain moneromooo?
endogenicwhat deters ASICs?
ErCiccionesgp: +1
iDunkIs this about the parameters tweaks othe suggested ?
pigeonsendogenic: having the work and investment they put in invalidated
moneromoooJust a few more simple ops in the inner loop. They mean… different hashes.
endogenicpigeons: how to detect their work?
ArticMineIt keeps ASIC developers off balance
endogenicor what is it :p
moneromoooendogenic: it doesn't need to. I have patches for most miners ready.
moneromoooThose (software) miners just update. Any ASIC can't.
endogeniceven if it's modular?
moneromoooWhat is a modular ASIC ?
endogenicit -> asic
msvb-mobpigeons moneromooo: We in the hardware team are onboard with PoW changes but cannot recall all hardware wallets worldwide to swap chips.
endogenicmodular hardware
moneromoooYou have hardware mining ?
pigeonsmsvb-mob: I wouldn't expect hardware wallets to need to mine or validate PoW
msvb-mobNot to make the debate a can of worms, but if anybody has a strategy idea for us hardware folks, let's please consider the probloem offline.
pigeonsjust sign i would think
rehrarmsvb-mob, why does a hardware wallet need to do anything with PoW?
msvb-mobpigeons: Okay, but changing the Cryptonote PoW wouldn't that mean changing transaction operations as well?
moneromoooNo.
pigeonstransactions don't use PoW
pigeonsbundling them in blocks do
msvb-mobpigeons rehrar moneromooo: Okay, that's the answer I wanted to hear, thanks.
rehraryou sign the tx and broadcast them to the network, then the miners (separate entities) mine and put them in blocks
rehrarso you're good, unaffected :)
rehrarwell then, it seems we don't have a lot to discuss on point 3 without the horsey
rehrarseeing as how Core Team has more or less delegated to him to handle releases and stuff
rehrarthough I hope he may show up soon, as….it's getting close to March
rehrarAny final thoughts on 3?
endogenicwe could just ask him lol
endogenicthat usually works
moneromoooI did. It didn't.
endogenic"we" :p
luigi1111He should be home by now
moneromoooMight still think 5 GMT though.
endogenicfluffypony:
rehrarI let him know yesterday about the time change
moneromoooDid he reply ? He's been mostly AFK for days.
rehrarand he responded. Got the message.
rehrar"ok cool"
rehrarwe can break that down and analyze that sentence if we want
rehrarAlright, maybe he'll show up and we can come back to this
rehrarmeantime, let's move on to 4. Code + ticket discussion / Q & A
endogenicooh me me
rehrargo endogenic
endogenicno sorry my mistake
endogenicticket discussion first?
endogenici would step on that otherwise
rehraryeah, if you don't mind. Next is 'any other meeting items'
rehrarand you can go crazy there :)
endogenicwell i thought it was q&a but yeah
rehrarah, I see. I thought it was unrelated.
rehrarshrug for lack of other people piping up with discussion, go ahead
endogenichah ok
endogenicso i want to beef up our code review capabilities
endogenicthe reason is that i found some code got merged without it really having been verified
endogenicwhich in some cases is necessary
endogenici mean it's not in everyone's interest to verify every single special case code branch
endogenicexcept
endogenicwhen it becomes a problem for the codebase
endogenicproblem is
endogenicwe cant just say "who is the best reviewer… vtnerd… you are now responsible for reviewing all code"
endogenicfirst no one man can take responsibility for that
endogenicfor everyone
endogenicsecond what if something happens .. bus effect etc
endogenicso i wanted to ask. can we figure out a system for this? like sarang's work but we must understand imho he has to be involved to vet the vetter
endogenicor it would be the blind leading the blind
endogenicin a manner of speaking
moneromoooYou just need people to review, don't you. So review, or find people to review.
endogenicno
moneromoooTo the first part ?
rehrarIt's tough on an open source project. People do many stuff in their free time and aren't getting paid. Perhaps it would be beneficial to get an FFS proposal by some volunteer coders (two or three) to review all PRs?
endogenicbasically. but to both… it is not able to be the solution
ErCiccionei think the problem here is the same the GUI has. We need more people…
endogenicwe have people
endogenicthey exist
endogenicdo they have the incentives?
rehrarthis is true
endogenicrehrar sort of in that direction
endogenicbut there needs to be a mediator who is technically qualified imo
endogenicand there are other problems
endogenicvtnerd just pinged me privately with one
endogenicimo we have to sort of standardize the effort
endogenicworking group?
endogenicthat's one part
vtnerdwhoa, standardize? not something I would make claims to. I've never seen an effective way to do that
endogenicthat's not what i mean
rehrarwhat we also need to realize and people and incentives, is that throwing money at people doesn't always make them want to do something
rehrarIf they work full time, they may not want to come home and be obligated to continue working (as opposed to contributing when they want)
endogenicwe ourselves already have the right incentives
endogenicbut disinterested parties may just gloss over lines of code
endogenicwhich look innocent
endogenicwhen i say standardize i mean we should take the fact that such problems exist and make them visible to the community
vtnerdahh ok, interesting.
rehrarWell, there is a Skepticism Sunday going on right now
endogenicno
endogenicwe want a working group with a document of problems that happen such as what i mentioned above
rehrarAh, I see.
rehrarWould you be willing to head this group then?
endogenici cant rn
rehrarWell, you and I can talk later after the meeting, and I'll see what I can do.
endogenicvtnerd might be
endogeniche is an expert
rehrarI'll be down to give whatever help I can to whoever heads up this group.
endogenicand he*
endogenici appreciate that :)
rehrarI think everyone agrees that more thorough review of the code can only be a good thing, it's just a matter of people, time, and actually getting it done
rehrarAlright, msvb-mob wanted some time for a few hardware things if that's ok with everyone.
msvb-mobYes please, but just short.
rehrarfloor is yours
msvb-mobI'll be advertising planning work for the Las Vegas events of August once in a while, anyone interested can look at:
msvb-mobm2049r[m] made a more or less incredible contribution to the greater Monero community by a first release (today) of a Monerujo-hw firmware.
rehrarwoohoo!
msvb-mobWe want to do the most we can with the firmware (on our own self grown hardware) and a few questions are arising about how far we can take a testnet approach to new users.
msvb-mobNew hardware users, that is. Since we can't recommend using the wallet for real funds.
msvb-mobSo if it would be okay to recommend trading testnet funds for a free t-shirt at events, for example.
msvb-mobThe last question on my agenda, is if anybody knows of the chance to insert a hardware wallet request to the group making videos (Savandra.)
msvb-mobThat's all for me.
msvb-mobfluffypony focused discussion on a vulnerability mitigation group last meeting, is that a topic we want to follow up on?
ArticMineYes
msvb-mobArticMine: Yes to which question?
ArticMineTo the vulnerability mitigation group
rehrarAs far as I know, savandra's last video is focusing on the community and that ends his FFS. We can ask him if he'd be willing to do another FFS to make another video for Hardware.
rehraralright, any last minute meeting items?
rehraralright, next meeting time
moneromoooNo talk about the vulnerability mitigation group ?
rehraroh, sorry. I may have misread all that
rehrarI thought it was a simple question/answer
rehrarTo follow up on that topic, the Vulnerabliity Mitigation Group is headed by sgp and is on mattermost
moneromoooWhat does this do ?
rehrarsgp, you around?
moneromoooTry to devise workarounds till a fix is in, etc ?
sgpYeah, but are you sure this is the same? Ours is the "malware response workgroup"
rehrarI think….this is the same, yes.
rehrarBut….maybe not? :P
ArticMineIt is a different issue
rehrarthen disregard me completely
rehrarwho is qualified to speak on that?
msvb-mobsgp: The 'malware response workgroup' is what I meant when contorting the words above.
sgpThe scope of "my" workgroup is to help those whose computers have been exploited for mining or ransomware
msvb-mobsgp: Sounds like most primary things about the workgroup have been decided then, good work.
endogenicsgp: i've been curious about the personal network security consequences of operating a monero node too. would be nice to talk with you about that sometime
rehrarAlright. We're over the designated time. So let's confirm next day:
rehrarFebruary 25th. Did this new time work for people?
rehrarwe were short dEBRUYNE and gingeropolous today. :'(
rehrarbut it may be for other reasons
rehrarWe can keep 16 UTC for next meeting unless there are great objections.
rehrarIf not, meeting over. :) Thanks everyone for coming.
ArticMineI would suggest also keeping it through the time changes
msvb-mobrehrar: Most excellent moderation, thank you.
sgpNot really meeting-worthy, but I would like to add the Monero Coffee Chats to the getmonero website
msvb-mobrehrar: If possible, would you please fork and pull request your source logos to the /documents/graphics/logo/ directory?
msvb-mobrehrar: Logo sources, as long as their formats are Opensource we don't need bitmaps.
rehrarOk, I'll get to that as soon as I can
sgpAre people against me uploading compressed versions of the coffee chats to the site? Or are people concerned they will be too large
rehraran hour is kind of long and would make having a fork on your local machine a big pain
rehrarespecially as there will be more and more of them
rehrarPerhaps links would be better?
msvb-mobsgp: Same consideration we have with large promotional photos of PCBs, which don't belong in revision control for obvious reasons.
vtnerdarticmine : canada does that crap too (arbitary timezone changes)?
vtnerdahh they do too, I hate timezones altogether actually, as stated previously
sgp@rehrar does the website allow me to embed YouTube videos, or would I have to do a simple link?
ArticMineDifferent countries do it at different times in the year . Even within a country there a all sorts of differences
ArticMineSo for something where there are people all over the world it is best to stick with one time at UTC
rehrarI don't see why not an embedded video
rehrarexcept that people with javascript shut down won't be able to see it
rehrarI think we should keep the website free from the necessity to turn on Javascript for anything, so I would advocate for a link, but that's just me.
Sources and notes
- Meeting log: Overview and Logs for the Dev Meeting Held on 2018-02-11, getmonero.org. The official post carried the raw log and pointed readers to Hello Monero for the overview.
- Original Hello Monero URL slug recovered from the Wayback Machine index; the original page content did not survive, so this reconstruction carries the log as archived by the Monero Project.