16:00:29 <meskio[mds]> #startmeeting tor anti-censorship meeting
16:00:29 <MeetBot> Meeting started Thu Oct  1 16:00:29 2026 UTC.  The chair is meskio[mds]. Information about MeetBot at https://wiki.debian.org/MeetBot.
16:00:29 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
16:00:34 <meskio[mds]> hello everybody!!
16:00:50 <meskio[mds]> here is our meeting pad: https://pad.riseup.net/p/r.9574e996bb9c0266213d38b91b56c469
16:00:50 <meskio[mds]> ask me in private to give you the link of the pad to be able to edit it if you don't have it
16:00:50 <meskio[mds]> I'll wait few minutes for everybody to add you've been working on and put items on the agenda
16:00:54 <onyinyang[mds]> hihi :D
16:01:10 <meskio[mds]> I guess I can remove from discussion the topics from last week
16:01:24 <meskio[mds]> or do we need to talk more this week about them?
16:04:56 <pryty26[mds]> Hi
16:06:06 <meskio[mds]> I guess this is a quiet week
16:06:21 <meskio[mds]> some of us have been AFK and others are AFK now...
16:06:37 <pryty26[mds]> Oh, 4i_un doesn't seem to come.
16:06:48 <pryty26[mds]> meskio: yep
16:07:11 <pryty26[mds]> I just continues my rust-pt things...
16:08:11 <meskio[mds]> nice, do you already have an example of a dummy transport?
16:08:38 <pryty26[mds]> Yeah
16:08:44 <pryty26[mds]> Lemme show it
16:09:02 <pryty26[mds]> (Ofc it's not very ready, I'm finishing the ExtOrPort part and the socks part
16:09:46 <pryty26[mds]> Is it ok to just copy codes in here, since it seems that everyone is using Matrix. So markdown would be compatible to everyone's device?
16:10:21 <meskio[mds]> if the code is public just give a link
16:10:57 <pryty26[mds]> (Oh unfortunatly I have not pushed it yet ) （；´д｀）ゞ
16:11:18 <meskio[mds]> ok, don't worry, take your time
16:13:00 <meskio[mds]> you can show us another time when is ready
16:13:10 <robgjansen[mds]> yeah I would rather see a more complete version later than an in-progress version now
16:13:46 <pryty26[mds]> Okay
16:13:54 <robgjansen[mds]> I have started reviewing the rust-pt lib. I will send feedback later today, in dm to pryty26
16:14:13 <robgjansen[mds]> (unless there is a preference for some other sharing method)
16:14:48 <pryty26[mds]> Well, every method is almost fine to me :)
16:15:08 <robgjansen[mds]> The high level so far is that the code is reasonable. It seems more complete than the quick and dirty version I wrote in Proteus. But let's see :)
16:15:36 <meskio[mds]> nice to hear, thank you for the review
16:16:29 <pryty26[mds]> Oh wait,
16:16:45 <robgjansen[mds]> No prob. Also just to make this clear, I am just a community volunteer so my feedback is not "authoritative" in any way. I'm just some guy who can share opinions that might be wrong ;)
16:16:47 <pryty26[mds]> I just pushed it, lol ( Because I just have to add a function to the example )
16:16:49 <pryty26[mds]> [ ](https://gitlab.torproject.org/pryty26/rust-pt/-/tree/main/transports/aspt?ref_type=heads)
16:16:50 <pryty26[mds]> Bruuh sign in?
16:16:53 <pryty26[mds]> https://gitlab.torproject.org/pryty26/rust-pt/-/tree/main/transports/aspt?ref_type=heads
16:18:18 <meskio[mds]> I think rob and pryty26 can discuss this on a DM
16:18:24 <meskio[mds]> anything more on this topic?
16:19:51 <pryty26[mds]> meskio[mds]: Ah,
16:19:56 <pryty26[mds]> I just talked
16:20:01 <pryty26[mds]> To the arti team
16:20:21 <pryty26[mds]> In tor dev channel ( Idk whether you have seen that :)
16:20:25 <pryty26[mds]> They said that:
16:20:54 <pryty26[mds]> ahf (IRC)
16:20:55 <pryty26[mds]> if what you're doing works with c tor and arti without changes to either there, then i think you've done something cool there
16:20:55 <pryty26[mds]> the more tight coupling between rust and pt's isn't something i think you can do withut very big api conversations with the arti team AND the anti-censorship team, and neither are ready for that right now
16:21:45 <meskio[mds]> I agree with those statements
16:22:30 <robgjansen[mds]> I don't see how rust-pt lib needs to discuss with arti at all. It is a library for parsing PT configuration from the env. I don't think the intention is to implement a new PT here
16:22:36 <pryty26[mds]> Because Idk whether you remember, last time, someone mentioned that Arti may like to manage the API design
16:23:08 <meskio[mds]> focus on make it work with the existing PT spec, arti is not ready to integrate PTs as libraries
16:23:08 <robgjansen[mds]> Only actual PT devs would need to worry about Arti api.
16:23:09 <pryty26[mds]> robgjansen:  Yeah, but AFAIR
16:23:31 <pryty26[mds]> Someone said that  Arti may like to 'manage' or at least have some opinion regarding the API design
16:24:12 <robgjansen[mds]> Yes, ok, I see. Perhaps arti doesn't want to use the current PT v1 spec.
16:24:24 <pryty26[mds]> That's why I raised that :)
16:24:55 <pryty26[mds]> Well, TBH, I have no clue whether they want to use the spec
16:24:59 <meskio[mds]> there is a future where we'll move away from PT v1 spec into some form of PT libraries into arti
16:25:04 <meskio[mds]> but that future is not there yet
16:25:19 <pryty26[mds]> I do not remember what is the exact code in tor-ptmgr
16:25:19 <meskio[mds]> and we'll need to work together between arti and anti-censorship to design that API
16:25:28 <meskio[mds]> you can ignore that future and focus on PT v1
16:25:49 <meskio[mds]> and this is the reality of PTs for a while
16:25:50 <pryty26[mds]> Ok
16:26:00 <pryty26[mds]> And as last meeting I asked
16:26:17 <pryty26[mds]> may I ask what is the 'good API design'
16:26:54 <pryty26[mds]> AFAIK, meskio  have said something that rust-pt should make sure that API suits their opinion
16:27:07 <meskio[mds]> that is not a trivial question to answer, but I will trust rob might have good opinions there as he has already worked on some PTs
16:27:53 <meskio[mds]> so I see rob's review very useful here
16:29:02 <pryty26[mds]> Oh great.
16:29:02 <robgjansen[mds]> Well in terms of a working strategy here: I think the current rust-pt should focus entirely on the V1 spec, and target existing PTs like Proteus, WATER, etc. The the purpose of this lib, we can basically ignore the future arti-pt api because that doesn't exist yet.
16:29:19 <meskio[mds]> +1
16:29:20 <robgjansen[mds]> s/The/For/
16:29:26 <pryty26[mds]> Thx
16:29:45 <robgjansen[mds]> And if someday a new arti-pt spec is produced, we can decide if we need a new crate or if we want to extend this one to support it
16:29:56 <pryty26[mds]> Also, I may like to ask some questions if there is a PT developer who has written golang PTs and Rust PTs
16:29:58 <robgjansen[mds]> * is produced and stabilized, we
16:30:02 <pryty26[mds]> Or at least one of them
16:30:52 <pryty26[mds]> Because I believe they would have great experience that I would like to know
16:32:00 <pryty26[mds]> Also, if there is a new rust PT developer, I would like to let them know rust-pt
16:32:44 <meskio[mds]> you already got a couple of PT developers looking at your code: rob and 4i_un
16:32:55 <pryty26[mds]> Yep
16:33:34 <pryty26[mds]> robgjansen:  So would you be free if I would want to ask some questions
16:33:36 <robgjansen[mds]> Yeah that's more than most people can say, so I don't think you need to worry :)
16:33:36 <pryty26[mds]> :)
16:34:26 <robgjansen[mds]> You have been asking questions :) But we can discuss more in DM when i send feedback.
16:34:48 <pryty26[mds]> Ok thx
16:34:49 <pryty26[mds]> I have nothing more
16:34:51 <pryty26[mds]> regarding that
16:35:04 <meskio[mds]> :)
16:35:10 <pryty26[mds]> (•ω•`)
16:35:20 <meskio[mds]> we have an interesting link about the russian fingerprinting of snowflake: https://github.com/net4people/bbs/issues/603#issuecomment-5817192874
16:35:23 <meskio[mds]> looks like is working now
16:35:49 <meskio[mds]> and if there is nothing else to discuss I guess we are done with this meeting
16:36:00 <meskio[mds]> I'll wait 1min to close it to see if something else pops up
16:36:31 <pryty26[mds]> I have to go to AFK
16:36:37 <pryty26[mds]> Thank you guys
16:37:05 <meskio[mds]> #endmeeting