March 14, 2019
<oneiric_> time to meet is almost here
<oneiric_> thanks to everyone for attending/lurking
<oneiric_> item 0: Greetings
<oneiric_> hiyo all
<oneiric_> welcome Corklander, looks like it may just be us today again
<oneiric_> item 1: Current project status / what's been done
<oneiric_> have spent the last few weeks refactoring and implementing generic wrappers for ecies
<oneiric_> went down a rabbit hole that took way longer thatn expected
<oneiric_> in the middle of switching crypto++ for libsodium
<oneiric_> last remaining piece is the AES impl, which will be replaced with tiny-aes
<kinghat> hi. lurking.
<oneiric_> reddsa took a backburner spot, and i will refocus on it after necessary refactors
<oneiric_> hi kinghat
<oneiric_> should be finished with refactors in next few days, and will MR the changes on gitlab
<oneiric_> will leave it up for ~1 week for review
<oneiric_> other than that, been working on other housekeeping refactors, and attending i2p dev meetings
<oneiric_> any questions/comments?
<kinghat> there were i2p dev meetings? were they public?
<crim-> will this have an FFS/ICO ?
<kinghat> whoops. not i2p xmr related dev meetings.
<oneiric_> yep, every week for #ls2 meetings discussing new leaseset2/netdb specs
<oneiric_> lol the greatest ICO crim-
<oneiric_> with all the premines
<crim-> wheres the old project?
<oneiric_> which old project?
<crim-> there was a bid FFS on kvri?
<crim-> is this replacing that?
<oneiric_> nop, not from me
<oneiric_> i'm still up in the air on an ffs
<crim-> okay, continue
<crim-> well, the last one run with the money
<crim-> or so I've heard
<crim-> people might be afraid for another FFS
<crim-> atleast me
<oneiric_> idk about all that. my last ffs got a little borked with overpromising/miscommunication, which has hopefully been resolved
<oneiric_> if you or anyone would like to discuss it, i'm open
<kinghat> im just here for the tini2p 😄
<Corklander> Same. Other projects are for other meetings, IMO.
<oneiric_> its my understanding that i fairly addressed regarding milestone completion, maybe you think differently crim-?
<oneiric_> regardless, i want people to feel they can approach me openly about funding questions
<Corklander> Ah. +1
<oneiric_> not here to swindle anyone, and i think my track record proves that
<oneiric_> any other questions about item 1?
<kinghat> my brain has already forgotten about the many different privacy network implementations that are in dev atm since the kovri mess.
<oneiric_> true, i2pd is still actively developed, along with java i2p, and i2p-zero packaging
<oneiric_> and there is ire of course :) the rust impl
<oneiric_> there may be more
<oneiric_> so onto item 2: Short-term road map
<oneiric_> so will continue with refactors + reddsa/encls2 this week
<oneiric_> hopefully should be ready to test within the next 1-2 weeks
<oneiric_> after that will shift focus back to ecies and i2np for router-to-router communication
<oneiric_> hopefully the generic wrappers i've written will be flexible enough for whatever crypto primitives are settled on
<oneiric_> leaseset2 and proposal 123 afaiu are still the main focus for i2p project, though focus maybe be shifting to 144
<oneiric_> don't quote me on that
<oneiric_> hopefully should have router-to-router communication within the next month or so
<oneiric_> at least ready for local testnet / vms
<oneiric_> also need to work on some docs / other meta housekeeping, and separating impl into ipp files
<oneiric_> any questions/comments on short-term plans?
<Corklander> For the testnet plans, do you have specific objectives?
<oneiric_> yeah, i'd like to build something people can spin up locally with no outside networking
<Corklander> As in: 1) must complete 'hello world' round trip. 2) must have memory leaks below 100MB/day, etc?
<oneiric_> tbh, haven't thought that far into it. but can come up with some metrics for automated testnet runs
<oneiric_> CI is still on my todo, so thanks for the reminder
<Corklander> I'm asking because I'm not sure where to separate unit and integration tests for a network-layer project.
<oneiric_> currently, i have separated the independently testable stuff into unit_tests
<oneiric_> stuff that requires networking goes in net_tests
<Corklander> Ah, right. Well that makes sense to my brain now. :)
<oneiric_> so far, haven't seen the need for a separate integration_test, but will if the need arises
<oneiric_> had a slow-test dir for a bit, but it was only for elgamal, and only when prime search was turned on
<oneiric_> now there's no elgamal at all so !!!!! ::)
<oneiric_> also zzz brought up that it wasn't obvious the library is header-only, so people may think nothing has been done
<oneiric_> all of the impl is currently in .h files
<oneiric_> i will be splitting impl into .ipp files, but the library is still intended to be header-only
<oneiric_> will write a reference router in .cc, but library user's are encouraged to just include the headers without needing a separate bin for the router
<oneiric_> if no more on 2, item 3: I2P proposal implementation (123, 144)
<oneiric_> not much has changed, str4d added a more fleshed-out reddsa spec
<oneiric_> there is currently some discussion over exactly how to do aplha generation and key blinding
<oneiric_> there will be another meeting in #ls2 on i2prc this upcoming monday, iirc
<oneiric_> 144 is the ecies-x25519-ratchet proposal
<oneiric_> any questions on tini2p impl plans or questions in general re: the new spec proposals?
<oneiric_> cool deal, looks like we'll be ending early
<oneiric_> last item: Confirm next meeting time
<oneiric_> same time in two weeks?
<kinghat> sure. gotta jet. see ya!
<oneiric_> l8er kinghat, thanks for lurkin
<oneiric_> ok so 2019-03-28 at 18:00 UTC
<oneiric_> meeting over, thanks all
<wowario> thanks oneiric_
<oneiric_> thanks for the input Corklander
<oneiric_> np wowario
<Corklander> Good work, btw
<oneiric_> thanks man
Post tags : Dev Diaries, Kovri I2P Router, Cryptography