I2P dev meeting, May 7, 2019 @ 20:00 UTC

Quick recap

  • Present:

chisana, eyedeekay, meeh, sadie, zlatinb, zzz

Полный журнал IRC

20:00:00 <zzz> 0) Hi
20:00:00 <zzz> 1) 0.9.39 release status remaining items (meeh, mhatta, nextloop, str4d)
20:00:00 <zzz> 2) Fixed i2pbote-android release status (meeh)
20:00:00 <zzz> 3) 0.9.40 dev status (zzz)
20:00:00 <zzz> 4) LS2 status (zzz)
20:00:00 <zzz> 5) Status scrum (zlatinb)
20:00:04 <zzz> 0) Hi
20:00:06 <zzz> hi
20:00:10 <meeh> hi
20:00:17 <zlatinb> hi
20:00:51 <zzz> 1) 0.9.39 release status remaining items (meeh, mhatta, nextloop, str4d)
20:00:55 <eyedeekay> hi
20:01:07 <zzz> this agenda item was from the meeting last month...
20:01:29 <zzz> where we were still missing debian, f-droid, and mavencentral
20:01:40 <zzz> meeh you have an update for us on any of those items?
20:02:33 <meeh> me and str4d will pair this release, meaning he walks over all the steps with me, which includes maven/f-droid. I'm contacting him today about this since we last discussed this a week or two ago.
20:02:43 <zzz> and has anybody heard from mhatta re: debian?
20:02:51 <meeh> that should cover fdroid, maven and googleplay
20:03:15 <zzz> meeh this is about .39 - have f-droid and mavencentral releases for .39 been posted?
20:03:33 <sadie_> hi
20:03:34 <meeh> osx launcher will ofc be updated as well
20:03:52 <meeh> no, but when I pair with str4d I learn how I can fix that
20:03:59 <meeh> so then both 39 and 40 should be fixed 
20:03:59 <zzz> we're talking .39 items from 7 weeks ago that were not complete at the last meeting 5 weeks ago
20:04:17 <meeh> https://search.maven.org/search?q=g:net.i2p
20:04:19 <meeh> maven is latest bt
20:04:24 <meeh> w
20:04:24 <meeh> .39
20:04:48 <zzz> ok. so what's the ETA for "pairing" (don't let him infect you with zcash-speak)
20:05:38 <R4SAS> :D
20:05:39 <zzz> and I thought our f-droid problem was server dependency issues, not the know-how. Were those resolved?
20:05:57 <meeh> my answer on that is when str4d have time for me for a quick walkthrough - which I took as he would find time for really quick when I spoke with him last. I'll update you on what he says today
20:06:41 <zzz> for mavencentral, you have everything you need including the howto, right? you have all the permissions you need?
20:06:58 <meeh> no howto
20:07:19 <zzz> I'm asking because 'when str4d has time' means nothing unfortunately :(
20:07:28 <meeh> we have some maven gradle script files in our repo, which I could *guess* how work and try, which I prefer not to mess up
20:07:46 <meeh> I have hope in him this time
20:08:09 <meeh> time to get rid of nagging, to free more time, - sounds something everyone would accept
20:08:42 <zzz> sure, it's the long plan to remove str4d from all dependencies, we're almost there, keep pushing until it's done
20:09:36 <zzz> ok back to mhatta - anybody heard from him? is it time to start calling him or where do we go from here?
20:09:46 <zzz> zlatinb, any ideas?
20:10:04 <zzz> and anybody heard from nextloop?
20:10:15 <zlatinb> calling him last time didn’t work, although he confirmed I had the right number
20:10:27 <zlatinb> no ideas unfortunately
20:10:39 <zzz> meeh, as I understand it you can also just put in some request to get into the real f-droid, we don't really need nextloop. Do you know how to do that?
20:10:40 <sadie_> can anyone take on mhatta's tasks?
20:11:15 <zzz> anyone who's an official Debian Developer... know any others with time?
20:11:38 <meeh> I'm sorry to say I'm even less known with frdroid than android - I've only (at least in my memory) been the hoster of our private repo
20:11:53 <sadie_> no - but we should start looking 
20:11:57 <meeh> but I can ask str4d while I talk to him, he should know
20:12:20 <zzz> ok meeh remember there's two fdroids, ours and the official fdroid.org, so please get the howto for both. Please try finding nextloop
20:12:27 <zzz> twitter, email, etc
20:13:13 <meeh> yea, our own is moved, and should in theory be working when I try those python commands to push it, - I'll test that out after speaking with str4d
20:13:19 <zzz> anything else on 1) ? We didn'd do a great job on .39
20:13:40 <zzz> so let's fix it quick for .40
20:13:57 <zzz> releases is what we do, when we can't get the releases out, we've failed
20:14:16 <zzz> anything else on 1) ?
20:15:18 <zzz> 2) Fixed i2pbote-android release status (meeh)
20:15:35 <zzz> meeh what's the status on the long-awaited bote android release?
20:15:42 <meeh> currently it still hangs on "connecting"
20:16:14 <zzz> but eyedeekay reports it works?
20:16:36 <zzz> did you try it non-firewalled meeh?
20:16:41 <eyedeekay> it works only if you use the built-in router. This is on a tablet
20:16:44 <meeh> ok, I feel we have talked many times now about the release. please don't take me as rude, mean, or anything besides just "trying to be as clear as possible" - I do not release a broken product. Unless I can send a message from my bote-android to my desktop-android - I'm not releasing
20:17:20 <zzz> this is a status meeting for everybody. Pretend you havent talked to me in a month, and tell everybody where we're at please. Youre not talking to me, you're reporting to the world
20:17:41 <meeh> because uless that, what we release is per definition broken - and it would reflect badly back at us
20:18:00 <meeh> ok sorry - but there it is ^ :)
20:18:27 <zzz> so, what's the outlook or ETA or guess for getting it fixed and released?
20:18:41 <meeh> but yea, I strongly feel that pushing a broken release would reflect badly at i2p - does anyone disagree on that statement?
20:19:11 <zzz> I'm not pushing you to release. I'm asking when it will be released.
20:19:26 <zzz> that's the purpose of this meeting, to report status
20:19:34 <meeh> how long time would you think it would take to make bote connect via tcp socket instead of unix domain?
20:19:46 <zzz> I dont know
20:20:06 <meeh> because at best, which I hope, it's that we need to do.. if not, no clue at all :(
20:20:32 <meeh> because if peers aren't the issue, and we still strugles to connect
20:20:38 <meeh> and based upon my debug sessions
20:20:48 <meeh> I suspect it's the IPC between the apps which make nonce
20:20:56 <meeh> and destroy the com
20:21:06 <zzz> ok. perhaps we can get some more testers from asking here in this channel... maybe cisiey
20:21:15 <meeh> cause the tunnel and leases seems ok in my view
20:21:39 <meeh> which my logical next place to look then is between router <-> bote
20:21:46 <zzz> ok. but if it's trying to build a tunnel for bote, then it's not a problem with the i2cp socket
20:21:57 <meeh> maybe the control socket work, but the data transfer stops
20:22:07 <zzz> you're either talking to the router or you aren't. it's all over the same socket
20:22:17 <zzz> (control and data)
20:22:32 <meeh> maybe google do something we don't know - wouldn't be the first time
20:22:50 <zzz> the good news is you're making progress, we've been stuck plenty of times before and always figured it out
20:22:54 <zzz> thanks for the update
20:22:57 <zzz> anything else on 2) ?
20:23:07 <meeh> so i2cp works like tls, it configures and then goes to pure data transfer?
20:23:34 <zzz> it's muxed control and data, all the time. it doesn't ever 'go to' pure data
20:23:54 <meeh> okay, ascii encoded binary data then or?
20:24:15 <zzz> I can walk you through how i2cp works after the meeting if you'd like
20:24:21 <meeh> anyway - we can take impl. details later
20:24:22 <meeh> yea
20:24:27 <zzz> anything else on 2) ?
20:25:02 <zzz> 3) 0.9.40 dev status (zzz)
20:25:16 <zzz> 0.9.40 dev status is DONE!
20:25:30 <zzz> we went for a short 6 1/2 week cycle and we released today
20:25:32 <chisana> congrats!
20:25:54 <zzz> in-net update is alive. Not sure if the English news has been updated though. The swarm seems a little small?
20:26:05 <zzz> eche|off, have you investigated?
20:26:25 <zzz> it's at least partially updating
20:26:47 <zzz> we're going to watch this one closely because we disabled NTCP 1, which shouldn't have any ill effects, but we'll see
20:27:19 <zzz> so, let's talk release items ETA. The in-net update is out, the website is updated. I'll do our PPA and deb repo tomorrow.
20:27:45 <zzz> meeh what's the eta for the android router release?
20:28:01 <meeh> again, I'll do this time with str4d
20:28:23 <meeh> the whole "str4d part" of the release in one pack
20:28:24 <meeh> :)
20:28:28 <zzz> even for android router? you did it by yourself last time. 
20:28:42 <zzz> I'd just like to depend on him for as little as possible
20:28:42 <meeh> Yes, by hacking around maven
20:28:50 <meeh> but the correct order is to deploy maven first
20:29:05 <meeh> Yes, but the goal is that this is the last time we do that, <ever>
20:29:24 <zzz> ok. we'll see if that comes true, hopefully so
20:30:07 <meeh> for the record, I've already contacted him on DM @ twitter
20:30:25 <zzz> ok let us know when you have an ETA
20:30:26 <meeh> which is where I've been talking with him the past times
20:31:02 <zzz> zlatinb, I guess you and I should send out to all 10 of mhatta's emails to try and get his attention? It's been two months since we've heard from him
20:31:25 <zzz> would hate to have some old release in buster
20:31:27 <zlatinb> sure, give me the emails and I’ll blast something out
20:31:38 <zzz> ok
20:31:59 <zzz> eche|off, eche|on you here?
20:32:39 <zzz> ok I guess that's it for 3) ?
20:33:18 <zzz> 4) LS2 status (zzz)
20:33:42 <zzz> I'll give a quick status on the 3 main proposals: 123, 144, and 150
20:34:13 <zzz> 123 is the ls2 proposal. It's mostly done and we got a lot of the encrypted ls2 code into .40 where we can do more testing
20:34:34 <zzz> the proposal itself hasn't changed much at all in the last month which is good
20:35:05 <zzz> 144 is the new encryption proposal. It's seen zero activity in the last month while higher-priority things happened
20:35:26 <zzz> but I've received a lot of comments on it. expect an updated draft in a week or two
20:35:48 <zzz> I'd like to see some implementations of 144 in 4-6 months but we have a lot of work to do to get there
20:36:05 <zzz> 150 is the new 'garlic farm' proposal based on raft, to manage massive multihoming.
20:36:30 <zzz> it's a new proposal but it's gotten pretty good feedback, less complaints by far than 144!
20:37:04 <zzz> zlatinb, when you think we might have an experimental plugin implementing 150? a couple of months?
20:37:24 <zzz> I guess we need to talk about what we can squeeze into .41 to support it
20:37:38 <zlatinb> depends on how we scope it; many of the listed features are really just nice-to-have’s
20:37:59 <zzz> yeah in my revision yesterday I marked a lot of it as not-first-release
20:38:07 <zlatinb> if we cut everything to the absolute bare minimum maybe less than a few months
20:38:47 <zzz> yeah we haven't had our roadmap meeting for .41 yet, so I'm not sure what's where priority-wise
20:38:49 <zlatinb> it’s a medium-sized project I think
20:39:18 <zzz> agreed
20:39:37 <zzz> anybody have any questions on 4), ls2, or related proposals?
20:39:52 <zzz> as usual, all are welcome at our meetings, Mondays 6:30 PM UTC in IRC #ls2
20:40:27 <chisana> if i can get a "working/MVP" ECIES together, will that help with 144 completion?
20:40:54 <zzz> of course
20:41:37 <zzz> somebody has to be first. we've done integration testing with several other proposals, going back a year to NTCP 2.
20:42:03 <zzz> somebody is done first, then somebody else, then you try to get them to talk, then you try to figure out who's right and who's wrong
20:42:33 <zzz> it's hard and slow work but it's the only way to do it
20:43:15 <chisana> for sure, 144 is where I'll be spending most of my time, so will be sure to come with frequent updates/questions
20:43:25 <zzz> we've learned we're almost never done until we have two impls. that's the only way to 'test' the spec
20:43:41 <zzz> great
20:43:49 <zzz> anybody have any questions on 4), ls2, or related proposals?
20:44:30 <zzz> 5) status scrum (zlatinb)
20:44:36 <zzz> you have the floor zlatinb 
20:45:15 <zlatinb> Hi.  The usual scum in parallel.  Please describe in a few words 1) what you have been doing since the last scrum 2) what you plan on doing next month 3) if you have any blockers or need help.  Say EOT when done
20:45:49 <zlatinb> Me: 1) finishing touches on the advanced filter, work on garlic farm, outproxy tuning with arctic, some jogger tix 2) hopefully just garlic farm 3) no blockers EOT
20:47:03 <zzz> 1) bug fixes; signed windows installer; switch to izpack 5; susidns import feature; implementation of encrypted ls2; .40 review and release; working on android I2CP issues
20:47:38 <meeh> 1) I've been redesigning the osx launcher and upgraded swift code to 4.2 soon 5, cleaned up in the code, migrated more from objc. I've also been doing some improvements for the services I host for i2p inc. outproxy. and some android development and learning
20:47:55 <zzz> 1 cont.) wrote garlic farm proposal, lots of garlic farm architecture stuff; ls2 meetings and discussions; work with sadie on icons etc.
20:48:09 <sadie_> working on updating iconography for .40, will continue this for .41 and also updating logo with the logo created by Ura for console, site. Improvements to set up wizard, preparing for trip to RightsCon and outreach with TX, IFF EOT
20:48:55 <zzz> 2) .40 PPA release; .41 roadmap; config splitting work with eyedeekay, garlic farm; ls2 meetings; bug fixes; maybe automate sybil detection -> blacklisting?
20:49:03 <meeh> 2) for the next month I plan to give our browser some love again, updating version to equal upstream, some i2p integration improvements if it's doable, make the osx native context menu -> snark-share usable, and also look into more the issues with mattermost/irc/our-communication
20:49:06 <eyedeekay> 1) Bote, bug fixes, writing guides 2) hopefully split configuration, bote, and bugfixes 3) no blockers EOR
20:49:10 <eyedeekay> *EOT
20:49:38 <meeh> 3) no hopefully not, only would now be awaiting str4d to suggest a time where we can pair and do the release
20:49:48 <sadie_> outproxy docs /testing 
20:49:50 <meeh> EOT
20:49:56 <zzz> 3) blockers: mhatta for debian/tails stuff
20:50:06 <eyedeekay> Oh also I'd like to spend some time with meeh on the browser as well in the next month.
20:50:17 <zzz> EOT
20:50:51 <zlatinb> Thanks, that completes the scrum.
20:50:59 <zzz> ok thanks zlatinb 
20:51:12 <zzz> a couple quick questions before we wrap up the meeting
20:51:27 <zzz> does a 8 week cycle and a July 2 release date for .41 work? meeh?
20:52:12 <meeh> no imidiate things saying it shouldn't work, I think it's ok
20:52:26 <zzz> ok, that's what we'll pencil in until we hear from eche|off 
20:52:57 <zzz> next, meeh, did you get the osx .40 build out? it's on the d/l page because it's triggered from the version number, but I didn't ask you if you'd done it yet
20:53:16 <zzz> so that's a bug in our process
20:53:24 <meeh> no I was thinking of starting that process now after the meeting
20:53:37 <meeh> didn't get time in front
20:54:06 <zzz> ok thanks. I'll fix our checklist so I don't bump the build on the d/l page until you're done, for next time
20:54:33 <zzz> last question for meeh, do you plan to do a browser release to fix the firefox cert thing? or do you even need to?
20:55:19 <meeh> could you eleberate the "firefox cert thing" please? :p
20:55:21 <zzz> as you said, time for some love for the browser. At some point we should talk about whether the 'lab' project should graduate to something more, or be killed?
20:55:48 <zzz> where the plugins like noscript broke. I'll fill you in later or look on twitter or ars technica
20:55:48 <meeh> yes, why don't we take that as a topic on our next meeting then?
20:55:54 <meeh> regarding RC or death
20:55:58 <zzz> ok, will do
20:56:09 <zzz> anybody have anything else for the meeting?
20:56:27 <meeh> I'm gonna start the osx build now if it's not anything more
20:56:44 * zzz looks for the baffer
20:57:00 <meeh> next osx upgrade would auto-update (woho!)
20:57:18 * zzz *bafffs* the meeting closed