monero-site/_posts/2019-05-23-logs-for-the-tini2p-dev-meeting-held-on-2019-05-23.md

52 lines
2.8 KiB
Markdown
Raw Normal View History

2019-06-03 07:43:14 +00:00
---
layout: post
title: Overview and Logs for the tini2p Dev Meeting Held on 2019-05-23
summary: Current project status, Roadmap, Meta issues, and miscellaneous
tags: [dev diaries, i2p, crypto]
author: el00ruobuob / oneiric
---
# Logs
**tini2p** 0: Greetings
hello to the void
1: What's been done
I2NP messages + processing
NetDB initial implementation
Global housekeeping refactors
Endian util removing Boost::Endian dependency
endian util was merged here: tini2p/tini2p!7
I2NP + NetDB + housekeeping is in this open merge request: tini2p/tini2p!8
Tunnel messages + processing still need implementation, along with some of the other networking parts of NetDB (that require Tunnels)
**tini2p** will discuss more about Tunnels in the next meeting item
global housekeeping was for style + correctness, cleaning up some early spaghetti code
some housekeeping remains, so it's definitely still WIP
2: What's to come
ECIES updates following #ls2 discussion and prop. 144 updates
Experimental Tunnel impl under ECIES
Writing a proposal for Tunnel changes under ECIES, following goals of prop. 144
Use standalone ASIO to replace Boost::ASIO dependency
my initial experimental ECIES implementation diverges a bit from the current spec proposal
will be updating the code to match the most recent revisions, and discussions from #ls2 meetings
some mismatch going forward is to be expected, since there are a number of TBD sections in 144
**tini2p** the hope is to get an MVP impl working, and come together on the necessary changes for the spec
similarly, Tunnel changes under ECIES were left out of prop. 144, and I will be implementing an MVP following the spec goals of 144
after getting something working, will write up a companion spec proposal for Tunnels under ECIES
one difficulty/interesting design challenge will be I2NP fragmenting + encryption for fixed-length tunnel messages
during this next run, I will also be replacing Boost::ASIO with standalone ASIO, completely removing Boost dependencies
**tini2p** no Boost will hopefully reduce overall code size
it will also ironically increase platform portability, since the project won't be tied to any particular Boost version (since no Boost)
for tini2p's separate client library, HTTP/WebSockets will be an interesting problem to solve
hopefully by the time the client library is being implemented, Beast will also have a standalone option
the first two items (ECIES + Tunnels) will likely take a large amount of time, probably spanning the next few weeks (if not more)
**tini2p** the goal is still to have a working (able to talk to other routers via Tunnels) by alpha release (2019-07-10)
`#ls2` I2P meetings have been very productive the last few weeks, and will hopefully continue to be moving forward
short meeting today
3: Next meeting time
2019-06-06 18:00 UTC
meeting closed