--- layout: post title: Logs for the Monero Research Lab Meeting Held on 2019-07-08 summary: Surae work, Sarang work, and miscellaneous tags: [dev diaries, community, crypto, research] author: el00ruobuob / sarang --- # Logs **\** Righto, let's begin our meeting! **\** Agenda: https://github.com/monero-project/meta/issues/368 **\** Starting now with GREETINGS **\** hi **\** Hi **\** suraeNoether: you here? **\** It's quiet... too quiet... **\** yes **\** sorry **\** hellow gents **\** harrow **\** Let's jump into ROUNDTABLE then, with our small crowd **\** mine is going to be super fast: last week basically right after the research meeting i started getting very ill. long story short, i went to the hospital, and let me just say: recreational pancreatitis is not a thing for a reason **\** but it'd be a cool name for a band **\** yes. yes. \*strokes chin\* **\** so, i'm trying to take it easy and i did very little this past week **\** Glad to see you're well enough to debug :) **\** For me, MLSAG/CLSAG verification updates continue, with PR 5707 open for review and similar changes to my CLSAG branch (to be included in later code for review) **\** i have a few things on my plate for today, but other than that: i'm basically all action items and no progress compared to last week. onto sarang :D **\** I've been in contact with Aram, the author of the Lelantus paper/protocol; he came up with an interesting idea to make the prover very efficient, at the cost of proof size and verification **\** although the verification cost can be batched, of course **\** I view proving time as generally unimportant (to an extent), but it's a very clever new way to prove a 1-of-N zero commitment **\** good prover times means fast construction of transaction for mobile devices **\** which is v nice **\** true **\** but if it costs you both space and verification time... **\** unlike, say, verification time, which puts a constraint on how rapidly the network can grow, which has security consequences for the chain **\** ^ ah yeah that's true: is it faster with a big space tradeoff? **\** Faster prover, slower verifier, bigger proof **\** I think you can batch away some of the verification increase (you effectively do two smaller proofs) **\** There's a non-public draft writeup already, but I assume he'll work it into the main paper once the modified security proofs are complete **\** oooof **\** very interesting consequences **\** Regardless, it's a damn clever construction **\** i'm excited to read all about it **\** I'll ask if I can send it to you suraeNoether (not public though, sorry) **\** I'm investigating a possible modification to Omniring that splits out the range proofs, improving verification batching at the cost of proof size **\** And, because of the Omniring non-batching currently available, am revisiting analysis of RCT3 **\** Which, while it would require a separate output pool (non-compatible key image structure), does allow for batching of proofs (aside from ring member group elements, which cannot be batched unless reused) **\** last we spoke about this, we were still interested in writing up a comparison paper, but you've done all the legwork on it so far **\** still the plan? **\** I don't consider a formal paper necessary, or even a great use of time **\** But analyses of spacetime, totally **\** fair, maybe we can do a blog post on tradeoffs between the three schemes or something **\** Maybe, but it gets subtle and complex really quickly under many different assumptions **\** There isn't really a quick-and-dirty soundbite answer to which is better or worse **\** Depends heavily on input/output structure, use of fixed epochs, batch behavior, etc. **\** and output pool migration is nontrivial **\** Omniring would \_not\_ require this... Lelantus and RCT3 would **\** (or rather, Omniring does not \_require\_ this) **\** see... when you say all that, it seems like it \*is\* a good use of time. maybe not high priority, but **\** A comparison is useful, I agree. But I don't want it to get lost in unnecessary formality of a full paper **\** And a comparison is exactly what I'm doing **\** Any questions on these topics? **\** Just a comment to say simple comparisons are good **\** roger **\** Does anyone else have research topics of interest to share? **\** Righto! **\** suraeNoether: people are wondering what the attendance was at konferenco **\** At least provide a range proof **\** 26 people and 172 sybils. **\** an article in coindesk mentioned 75 which I know is way low **\** nioc: i'm finishing up my post-mortem report on the konferenco today (on my action item list). We had 150 swag bags made, with around 30-40 leftover, but we had 27 speakers and like 10 sponsors. **\** I estimated 120 just by glancing **\** 75 is the number i gave coindesk for the number of attendees on Saturday morning, but more people bought tickets on both days and the totals were higher **\** if you count speakers and sponsors, that's around 110 on the first day, and around 125 the second **\** think you mran Sunday morning **\** mean **\** nope, we sold tickets throughout the afternoon on saturday and a few on sunday too. but coindesk asked for a comment on saturday morning, so i told them what i had sold at that point **\** I'll wait for the report **\** k **\** thx **\** suraeNoether: congratulations on effectively committing yourself to running a kickass conference annually until the end of time =p **\** it was actually some of the best days of my life, but i've been told explicitly by my doctors that i need to take a vacation **\** so i'm planning for that in august, since scary cardio and internal medicine people told me so with stern voices **\** Well then, let's move on to ACTION ITEMS **\** I have many things in progress. Lelantus proof review, modified Omniring split proof analysis, RCT3 analysis, and starting to put together my defcon talk/workshop **\** my action items: konferenco post mortem, research report for previous quarter, funding request for the next 3 months, and some debuggin **\** I'm doing a talk on transaction protocols (very high level), a workshop on simple cryptographic constructions with Python, and a panel discussion at the blockchain village **\** oh and i'm definitely not going to defcon this year. i can ship leftover swag like our USB data blockers with the monero logo and our pull-up banners if someone sends me the information for it **\** :( **\** The pull-up banners would be nice, assuming it's cheaper to ship than to get new ones **\** as would the USB blockers **\** i'll look into it; ordering the usb blockers may be short notice but i can find out **\** unless you just meant shipping the banners **\** which is cool too **\** Yeah I meant the banners **\** I believe there was an idea to perhaps order more blockers for this (I'm not the one to ask) **\** If you have extra USB blockers and would ship with the banners, cool **\** fantastic, i am happy folks liked the blockers **\** Any last questions or comments before we formally adjourn, since agenda topics have been completed? **\** Going once **\** twice **\** adjourned! **\** Thanks to everyone for attending; logs will be posted shortly on the github issue