14:11:20 02f 14:11:21 //// 14:11:23 // BalanceExclusions 14:11:25 // - Enotes that match with a balance exclusion will not be included in a balance calculation. 14:11:27 ///F0c+*ü9ä, 14:13:56 sry spilled some stuff over my laptop 14:16:54 Well, you won big at the monkey with a typewriter competition I guess. 14:37:38 clipboard content could've been worse 16:52:43 Meeting in a bit more than 1 hour 17:59:56 Meeting time. Hello! https://github.com/monero-project/meta/issues/1255 18:00:22 hey 18:00:29 Howdy 18:01:08 *waves* 18:01:29 Alright, with that wave we can start with the reports from last week 18:02:51 I wrote a wrapper so `wallet_keys_unlocker` can be used by the Wallet API ([commit](https://github.com/monero-project/monero/commit/c995b8dd03ab7a3d7e03cf078142144a49ae2bbd)), also based it onto jeffros improved version [#9860](https://github.com/monero-project/monero/pull/9860), was excited when the tests passed, but after testing `make_multisig` manually from the CLI it still do esn't work, so will have to do more debugging. 18:02:58 Would be nice to get some feedback on the commit, it feels a little messy with all the forward declarations and I'm not sure there may be room for improvements 18:03:34 Hi 18:03:58 hi 18:03:59 Uh, that #9860 PR looks pretty wild, with all those issues about something so small ... 18:04:41 Is that not yet merged because a second review is missing? 18:05:46 also re-review was requested after two small changes afaict 18:06:13 me: no significant update from last week (was mostly unavailable for personal reasons unrelated to the mainnet situation, I'm back at 100% now). Patched a bug reported by ofrn (fcmp++ wallet refresh recovering from daemon popping blocks correctly), currently investigating another. Once this final set of bugs is out of the way, I intend to take next steps on alpha stressnet (settin g a start date, writing a post with instructions on how to join, etc.) 18:07:04 Sounds like the real fun can soon start then. allpha net will be a pretty milestone I think. 18:07:10 Me: I mostly worked on consensus based hash changes, unrelated to Qubic stuff (e.g. https://github.com/monero-project/monero/pull/10038) 18:07:49 What is an "intermediate PoW hash"? 18:07:59 I also want to propose a new Carrot math audit at the next MRL to address some small things: https://gist.github.com/jeffro256/12f4fcc001058dd1f3fd7e81d6476deb 18:08:41 It enabled tevador's idea to be able to partially check PoW without needed to use RandomX, but a much smaller, lighter function Blake2B 18:09:02 SNeedlewoods: looking at that commit now 18:09:06 got my fcmp testnet up to ~18mb blocks, mining stopped for now while jberman investigates an issue 18:09:10 Ah, I see. That's floating around for quite a while already 18:09:26 https://github.com/monero-project/monero/issues/8827 18:10:16 Yes, exactly 18:10:29 Will link that in the PR, thanks 18:11:31 So this Carrot follow-up audit will preferrably go again to CypherStack, to profit from pre-knowledge? 18:11:49 Yes, it *should* be relatively quick I think 18:12:19 Haven't solicited quotes yet, just wanted to see if anyone would be opposed to it 18:13:11 Proposed follow-up audit lgtm 18:14:27 Yeah, checking again can't be wrong. If a problem sneaked in, however unlikely that is, we must know. 18:15:00 also to note that my fcmp testnet is using jbermans 128in pr, and i feel the public /alpha test/stress net should as well 18:15:09 Agreed 18:15:42 For the record, and for visibility, here a link to kayabNerve's CCS proposal to harden Monero against certain attacks, the "Finality Layer Book": https://repo.getmonero.org/monero-project/ccs-proposals/-/merge_requests/604 18:15:59 Don't be shy, make your voice heard 18:16:11 I'll share a comment on that proposal soon 18:16:20 likely today 18:17:15 The ccs isnt to do any implementation, and shouldnt be misinterpreted as monero is going yo use a tfl, but solely the book, explaining/documenting the proposed solution 18:17:39 It's unfortunate that we probably will have *two* monster projects running side-by-side, at least for some time, but you can't choose that. 18:17:57 So the ccs isnt "to harden monero against certain attacks", but to write a book about it 18:19:31 You are right, too easy to misunderstand the way I put it. But anyway, the *idea* is that this book will lead to a way to harden Monero - if we do anything at all in this direction of course. 18:20:13 I guess we have many quite controversial discussions ahead on the way to that. 18:22:09 Ok, do we have to discuss something today beyond these reports? 18:23:22 For mere length of meeting, MRL wins hands down for quite a while :) We might have more to coordinate once the alpha net starts. 18:23:59 last MRL meeting no one told me to bring water, was very dehydrated at the end 18:24:08 Lol 18:24:51 I think we can close for today. Thanks everybody for attending, read you again next week! 18:25:07 thanks everyone, cu 18:28:03 [@jeffro256:monero.social](https://matrix.to/#/@jeffro256:monero.social) for wallet devs, if fcmp++ basically plug-and-play? 18:28:36 (For wallet_api wallets) 18:29:18 xyproblem: if i built a wallet like monfluo and swapped in the fcmp repo, would it "just work"? 18:30:26 Yes, mainly. Hot/cold wallets will have a hiccup in the API regarding fetching private tx keys, but regular wallets should "just work" after PR #73 was merged AFAIK 18:30:37 That's the goal 18:31:06 Please lmk if that's not the case 18:34:06 just had a quick look at the `fcmp++stage` branch, kinda surprised how little has changed in the /api/ subdir 18:36:01 Yup that's the goal. Carrot is backwards compatible with the existing address scheme, FCMP tree building can/should be done during normal wallet refresh, and FCMP++ proof building is more or less an internal feature to transaction construction 18:48:33 Which is good because all of this lws/lwsf/wallet work wouldve been partially wasted 18:49:05 I just have to figure out how to get fcmp++ working with lwsf, etc 19:00:18 The main difference at a conceptual level for light wallets is that they don't do a full refresh, which means that they wouldn't be doing FCMP tree building, which means they can't construct FCMP++ tx proofs normally. 19:00:53 During transaction construction, light wallets should fetch FCMP tree paths from the daemon for outputs to be spent (or all owned) 19:01:23 This somewhat assumes that spending privacy is already gone for light wallets from the perspective of the daemon 19:01:56 Optionally, you can also fetch FCMP *decoy* paths to get ring-signature-like privacy against the daemon 19:55:54 I mean I was assuming that the view-balance-key would be used by lws in the future anyway. its just an unfortunate tradeoff 19:56:19 technically the server effectively knows when a spend occurs anyway (now)