IRCaBot 2.1.0
GPLv3 © acetone, 2021-2022
#i2p-dev
/2025/12/15
@eyedeekay
&eche|on
&zzz
+Irc2PGuest61619
+R4SAS
+RN
+RN_
+StormyCloud
+T3s|4
+cumlord
+dr|z3d
+eche|off
+hagen
+orignal
+postman
+qend-irc2p
+snex
Arch
Birdy
Daddy
Danny
FreefallHeavens
Irc2PGuest20240
Irc2PGuest21366
Irc2PGuest32328
Irc2PGuest57061
Irc2PGuest62416
Irc2PGuest68429
Irc2PGuest7705
Irc2PGuest96466
Onn4l7h
Onn4|7h
Over
Sisyphus
Sleepy
SlippyJoe
Stormycloud_
T3s|4_
Teeed
aargh2
ac9f_
acetone_
anontor
b3t4f4c3___
dr4wd3
duanin2
duck
gellegery
leopold
makoto
n1
nilbog
not_bob_afk
nyaa2pguy
onon_
poriori_
profetikla
r00tobo
rapidash
shiver_
solidx66_
thetia
tr
u5657
uop23ip
urist
user1
vivid_reader56
w8rabbit
x74a6
zelgomer
zyx
zzz folks, there's some questions about the mechanics and the quality of the translations on the new beta site, we need some feedback
zzz according to stormy, all the old transifex translations have been tossed in favor of some sort of machine translation
zzz if you're fluent in any of the translated languages, please take a look around, link in /topic, and give StormyCloud your thoughts
zzz the number of translations is also greatly reduced, if something important is missing, let us know
zzz also, we now have all the specs and proposals translated, they weren't before, but I don't know if that's a good idea or not
zzz correction, in the docs/ subtree, only proposals have been translated so far, not specs/ or non-spec docs/. So look at proposals to get the general idea of the quality
zzz eyedeekay, in your go-i2p i2cp server side, you should be reporting the API version 0.9.xx, not the Java router release version 2.10.0. I believe you have it right in your go-i2cp client code.
eyedeekay OK. I have a ton of other problem with the go-i2cp client code unfortunately, working through them
zzz ok this one is a server side issue, both sides should send the 0.9.xx version string
eyedeekay Ack I'll check the server side one in in just a sec
zzz nits, but all your server side docs also refer to "I2CP version 2.10.0" which isn't right either
eyedeekay Did those too
zzz advertise the API version from the I2CP spec version chart based on what you support
eyedeekay fd346d07203b7d6bae2f305d2ca101cd88faafd3
eyedeekay I will work my way down the chart and select an appropriate value today if 0.9.67 doesn't totally fit, I hadn't actually done a checklist against that yet, only validated against i2psnark
zzz not sure what you did but you have more to do. protocol.go line 74-76; server.go comment line 1320; protocol_test.go lines 324-332; disconnect_test line 53 (maybe?)
eyedeekay Oh oops my sed wasn't good enough, thanks I'll be more careful
eyedeekay Heh, and I thought those tests would be pointless...
zzz curious how long it took you to do ratchet. was it hard? never heard a peep out of you about it ))
eyedeekay It was only about a week and a half but I've been working like 14 hours a day since late October, was too tired to talk ))
zzz zowie thats fast
zzz might want to add PQ ratchet to your roadmap
eyedeekay I was really pulling out all the stops
eyedeekay Sure. Right now you know my immediate goal is to have embedded ready to present in a couple weeks but in January I'll want to get right back into the difficult stuff
zzz it took us months. you're now officially going faster than we are
eyedeekay Well I can't keep this up much longer. My hope is that I'll get to the point where I have more spec-implemented than spec-to-implement which feels like it's happening soon
eyedeekay And at that point I'll probably shift to a more reasonable pace and go after some of the probably-considerable details I've missed this month
zzz it seems like this is rapidly transforming from 'side project to learn i2p' into 'real project targeting ???' which is going to trigger some discussions or meetings we haven't had yet
eyedeekay Yeah. I've learned a lot, and it's in pretty decent shape for what it is, I've put effort into keeping it organized even though I've been working faster than I probably should
eyedeekay Ultimately my goal isn't really to be a whole freestanding router though, at least not right now, what I want is turnkey application development libraries that handle the whole router lifecycle end-to-end
eyedeekay Also I have some grand designs of a go ecosystem for application development that makes I2P integration essentially transparent
zzz all the docs you're doing make it look like a serious project and if that's the current goal now, there's talk todo about where it lives and what it's labeled and how much resources it gets
zzz some of that is probably intra-stormy discussion and some is probabably here
eyedeekay Totally. I know what my goals are, happy to be transparent about them, this is evolving into a serious project, I'm taking it seriously and intend to be dogfooding it by February at the latest
zzz trying to find some post on our criteria for 'official' software, perhaps something that came out of the i2p browser debacle...
eyedeekay The design is embedded focused, anybody can take it and make an I2P router by embedding it in an application and handling user input, as such it's focused on presenting API's not applications, and the actual library code is spread out across much of the go-i2p namespace partly for reusability purposes and partly to delineate underlying functions from router functions from application-level functions
zzz here's one on 'endorsement of related software' zzz.i2p/topics/2020
eyedeekay All right I'll see what I hit on that checklist so far
zzz may not be completely applicable but a decent checklist
zzz you may wish to share it with altonen also; perhaps we would want to "endorse" his project also, in parallel with yours, at some point?
eyedeekay Agreed I'll bring it to his attention
zzz when I said "where it lives" I mean this, and a lot of your other stuff, could become "official I2P projects" or some new thing called "official StormyCloud projects", for example
eyedeekay I don't have a strong opinion on which org should be supporting it/it should be supporting yet but at some point I'll want to do something like that
zzz that will also require you to draw the line somewhere through your 100 projects that run the spectrum from solid to abandonware ))
eyedeekay Hey I'm down to like 275 from over 1000 before now ))
eyedeekay But yeah I'll keep cutting down