Seraphis wallet workgroup meeting #68 - Monday, 2024-04-29, 18:00 UTC about meta HOT 1 CLOSED

<r​brunner7> Meeting time. Hello! https://github.com/monero-project/meta/issues/997
<o​ne-horse-wagon> Hello!
<d​angerousfreedom> Hi
<a​rticmine> Hello
<r​4v3r23> hi
<tobtoht_> Hi
<j​effro256> howdy
<r​brunner7> So what is there to report from last week?
<r​brunner7> I wrote this here yesterday: https://old.reddit.com/r/Monero/comments/1cf8oom/the_monero_core_software_wallet_api_will_probably/
<s​needlewoods> hey
<d​angerousfreedom> After the discussion about deprecating [wallet2](https://github.com/seraphis-migration/wallet3/issues/64) and the willingness of current and new wallets to rely only on the `wallet2_api` (if it has all features needed), I believe it makes sense to me to re-target my next tasks to use an interface for my PoC CLI wallet. The idea will be to use the `wallet2_api.h` and/or create a ne<clipp
<d​angerousfreedom> w interface from which I will base on my CLI. The final goal stays the same. I want to have the most basic CLI wallet capable of sending/receiving txs using a regtest network. After that, hopefully, we will be able to discuss the scalability/compatibility/design and other issues with that approach. I also tried to study FCMP but couldnt find enough time to fully digest kayaba's introduction on it.
<s​needlewoods> looks like only positive feedback there too
<jberman> :waves:
<s​needlewoods> I haven't done much but reading some code regarding the wallet api, but I reviewed https://github.com/UkoeHB/monero/pull/42
<r​brunner7> kayabanerve is fully funded already, right?
<d​angerousfreedom> Yes :)
<r​brunner7> Nice
<o​ne-horse-wagon> rbrunner7: Monero has never wanted funds for legitimate projects that I've ever seen.  It's very strong financially.
<a​rticmine> I am working on the new scaling and fee proposal. This is based upon the latest estimate of 3000 bytes 2/2 transaction for FCMP. 
<a​rticmine> It will incorporate a new sanity median of 1000000 bytes. 
<a​rticmine> Also the use of node relay to control growth, by increasing the minimum node relay fee.
<r​brunner7> I was almost a bit disappointed that I did not manage to flush out a single wallet dev with a reaction of "Whaaaat!" with my post
<jberman> me: implemented a tool to import the Seraphis enote store into wallet2's m_transfers container (+ m_payments + m_confirmed_txs), which is useful to enable using wallet2 to construct txs "in a box" as described here: https://github.com/seraphis-migration/wallet3/issues/48 , and to drop the async scanner into wallets today
<jberman> Putting final touches on that, and will implement koe's review comments in the async scanner next, then open another CCS
<j​effro256> ArticMine:  what do you think of tx-pow?
<a​rticmine> Not a good idea
<r​brunner7> jberman: Nice that you already started to work on that. Progress finally!
<j​effro256> @jberman: nice !
<a​rticmine> It is very discriminatory against the global south
<jberman> the enote store is missing a few things to have complete compatibility, I have a list written up and will share it soon too
<r​ucknium> FWIW I am also skeptical of tx-pow. But I will run some numbers anyway.
<a​rticmine> POW belongs in cold places during the winter. Not in the tropics
<r​ucknium> Probably the set of tx-pow parameters that make spamming very difficult and normal transactions easy, is empty.
<j​effro256> That's about my thoughts as well, just wanted a quick opinion ;)
<j​effro256> Talked to kayabanerve  yesterday. He claims that with a new divisor scheme, the FCMP-RCT proposal is comparable (even for the first layer) with Seraphis in performance
<a​rticmine> Is this tx size, verification or both?
<r​brunner7> Everyday something new out of that corner, it seems :)
<s​yntheticbird> Is this new divisor scheme also compatible for Seraphis ? Or will the performance improvements only meaningfull for FCMP++
<s​yntheticbird> only be meaningful*
<j​effro256> ArticMine: at least for CPU time, I'm not sure about tx size
<j​effro256> verification side, that is
<r​ucknium> Is this the new divisor scheme by Liam Eagen that Cypher Stack decide not to review?
<j​effro256> SyntheticBird: This is where it gets a bit hazy for me, but IIUC, the divisors here are really only applicable to FCMP-RCT since they help keep track of the parallel keys together
<j​effro256> So they don't help *as much* with Seraphis
<j​effro256> Rucknium: I do not know
<rucknium> +1
<r​brunner7> dangerousfreedom, SNeedlewoods , did you actually team up on that wallet API work?
<s​needlewoods> I haven't found the time yet, only looked at the current state of the code for now
<d​angerousfreedom> No. I havent start implementing anything yet. Just thinking how to do
<r​brunner7> Alright. I do think it could be a good idea to build a little team there.
<0xfffc> +1
<dangerousfreedom> +1
<s​needlewoods> if I can close https://github.com/UkoeHB/monero/pull/40 then I have nothing on my table, that would prevent such work
<r​brunner7> That would be most unfortunate, with all the work that we have ahead, in principle :)
<r​brunner7> Really, having that API would open up many work opportunities, as already mentioned last week.
<s​needlewoods> sorry, I think I phrased that poorly, I mean if that PR gets closed, there is nothing 'urgent' that would prevent me to fully concentrate on the API
<r​brunner7> :)
<s​needlewoods> and jeffros #42 does the same as #40 and much more, plus it's way cleaner, so I'm in favor of #42
<jberman> seems like we're good to close #40, I want to review #42 as well
<jberman> I could use help on this idea: https://github.com/seraphis-migration/wallet3/issues/64#issuecomment-2070568200
<jberman> if we agree it's an acceptable route
<r​brunner7> Will have to read it first carefully, and then think about it. Probably it's again a case of almost too many possible ways to go with our move away from wallet2 ...
<jberman> @dangerousfreedom legacy knowledge proofs are an example of something to swap in the wallet API from using wallet2 (m_wallet->) to using the Seraphis lib
<r​brunner7> I guess a first important info will be to see what is actually missing from wallet2_api.h and connected other .h
<d​angerousfreedom> jberman: True
<r​brunner7> Ok. Is there anything else to discuss today?
<r​brunner7> Doesn't look like it. So thanks everybody for attending, read you again next week!
<s​needlewoods> thanks and cu
<d​angerousfreedom> I did an implementation of that a while ago: https://github.com/seraphis-migration/wallet3/issues/55#issuecomment-1696071739. We just have to better structure the work now :)
<d​angerousfreedom> Thanks