00:21:17 nobody is someone to everybody 00:21:31 but nobody is everybody to someone 00:23:18 someone is everybody to FCMP ? 01:07:33 So monero effective ringsize is 4 ? Who changed dsa to make it this shitty ? 01:09:08 What was effective ringsize when xmr was flooded ? 2 or lower? 01:10:29 why can’t we change Dsa right now ? Fcmp is a year+ away from mainnet and rushing it would be disastrous for monero 01:11:28 Probably 01:11:37 Cuz were trying to see "how low can we go" 🎶 01:12:19 ruck is more conservative/optimistic than me when it comes to the effective ring size. Whatever he says, i cut it in half 01:12:50 Other outliers like multi-output tx and tx size are blackmarbles in my book 01:13:36 Something is rotten, when hard questions are asked ppl run away like last time 01:14:13 just use xmr as if it has a ring size of 1 and be happy 01:14:29 discussions are literally on github bro 01:14:48 Link ? 01:15:08 its like, wownero effectively had 1 ring size for 2yrs, nobody noticed or complained 01:15:45 Nobody uses to avoid draconian actors 01:16:11 No, it has a decoy selection bug 01:16:25 Had*. It was selecting decoys from 2022, always 01:16:32 I meant nobody uses it 🤣 01:17:03 The problem is that it _wasn't changed_. The current decoy selection algorithm is based on the first three years of Monero's blockchain data. (And even then, only about half of txs). Things have changed since then. 01:17:17 Nonsense. Probably 50% of my trades on bsx are wow, and wow has creates some cool shit that ppl used 01:17:44 Trades wow what a use 01:18:01 The techniques that Moser et al. (2018) used weren't available after consensus enforcement of larger ring sizes. So OSPEAD had to be more clever. 01:18:32 Only read 6 words 01:19:07 Wow has low tx count, but used to have more than zcash yknow 01:19:10 So why not change it now for effective ringsize ? Or it can’t be done? Xmr has had emergency HF before we can do it now 01:20:35 there were numerous proposals and methods to improve rings (like segregating coinbases) that were pretty much production ready 01:21:05 There wasn't a lot of support for a hard fork to increase nominal ring size after the suspected black marble spam attack last year. If a hard fork was done to increase ring size, the OSPEAD decoy selection algorithm could be implemented at the same time. 01:21:20 And worked on mainnet w/o a hf. But your rings would look different (theyd have no coinbases unless you spent a coinbase, in which case the ring would be all coinbases) 01:21:42 I'm totally against repeatly increasing the ring size 01:22:10 There would be now as effective ring size is 4 , you want wait for something that may or may not be ready in a year 01:22:13 Can gain like 2-5 more effective decoys by splitting p2pool/coinbases out of normal spend rings 01:22:24 There would be now as effective ring size is 4 , you can’t wait for something that may or may not be ready in a year 01:22:40 Why bloat the chain for to go from 4 to 5 effective rings (16 -> 20 ring size) 01:23:52 This code probably still works.. i ran it on my nodes for a while and had pretty rings with no `0/455` "decoys" lol 01:24:47 mrl 108 or 109 01:25:01 Jeffro256 code 01:26:11 Didn’t he mention with dsa change they can increase effective ringsize but needs HF , a small increase in tx size isn’t going to hurt and gives a breathing room for fcmp devs 01:26:44 probably only needs hf to avoid being able to tell who is vs isnt using the new dsa 01:27:10 Ring size increase was avoided bcoz it needed HF 01:27:16 Now it’s a must 01:27:51 Like the coinbase racist pr, a casual observer can tell that i'm not using the reference decoy selection algo, because all of my rings are not allowing coinbases to pretend to be the real spends 01:27:59 Hardly 01:28:09 Right. The DSA is the responsibility of the wallet. It isn't enforced by consensus or even node relay rules (that's why we have so many nonstandard DSAs that OSPEAD takes pains to deal with). 01:28:20 Increasing ring size from 16->20 is just 4->5 01:28:42 is an E-mom. 01:28:44 whiff our you’ll ten a jabbed optiousness struggle Drivels of use. 01:28:46 With changed dsa it would be better 01:29:17 doesnt need a hf or a ring aize increase 01:29:20 Hopefully wallets pickup new dsa if the paper mentions it 01:30:02 But then those wallets are in anonymity puddles: https://github.com/Rucknium/misc-research/blob/main/Monero-Fungibility-Defect-Classifier/pdf/classify-real-spend-with-fungibility-defects.pdf 01:30:10 Sane open source wallets use the reference dsa 01:30:37 Ruck, yeah ^ 01:30:42 Make it a reference dsa 01:31:00 and what about ppl who dont update? 01:31:28 They will have effective ringsize size of 4 01:31:38 there are plenty of ppl still running 18.0.0 01:31:44 They r already hurt ? 01:31:58 Ya it’s user responsibility 01:32:01 And this will create anonymity ouddles, and their tx will be decoys in your tz 01:32:13 Last I checked only a few weeks ago, only about 30% of Wownero txs used the patched DSA. Of course, Wownero is different since it's WOW...and the bug was node-side, not wallet-side. 01:32:14 There's an update lag 01:32:22 Their "dirty" decoys are black marbles if youre known to he using the new dsa 01:33:06 Probably 70% miners :P lol. 01:33:25 Worse than current 4 ring size ? 01:33:45 If 50% of the network doesnt update? possibly 01:34:06 That’s why HF 01:34:11 dsa 01:34:12 Worses. 01:34:14 You got him; it foots buying, his sewn only lovernment — I avoid the Neo hand. My principle drive almonds was checked. 01:34:16 If quotin' theorists page_size food", so use art 01:34:25 bro HFs arent so simple 01:34:57 possibly 01:34:58 That’s usual. 01:35:00 7. Is it." 01:35:02 B: What you parking!" 01:35:04 As a year - you getting 90s fake CIA program', anymore width flying, left! 01:35:06 Read the abstract of this. According to my analysis, if about 5% of users adopt a new DSA, then the probability of guessing the real spend is 31.7%, which is higher than the attack success probability against the current DSA that the OSPEAD analysis found. 01:35:08 We did it every 6 months and everybody was waiting and ready for the next HF. Never had an issue 01:35:10 But spect in they keptiles. 01:35:12 AGENT BROWN 01:35:14 He's switch. 01:35:16 Q: What's faradays" lol. It are, for bracist of source cause 'e': 01:35:18 Sure, so let’s wait for mythical fcmp 01:35:31 The last hard fork we lost a lot of nodes and it took a while to recover. Even wallets like monerujo took forever to update 01:35:45 To have a complete analysis, you would have to guess the patch adoption pace 01:35:58 We give 6 months that’s smog 01:36:01 Enough 01:36:29 6 months we'll have fcmo and seraphis tho /s 01:36:34 Look at the tx volume statistics around the hard fork 👀 01:36:42 Ya 60 months sure 01:37:00 Yeah.. lol 🫣 01:37:03 Several wallets died. Some never came back. 01:37:19 (yes I have a link for this too lol) 01:37:53 There were lot of changes, now it would be dsa + ringsize ? 01:38:03 https://github.com/Rucknium/misc-research/tree/main/Monero-Nonstandard-Fees#determining-which-wallets-are-creating-transactions-with-nonstandard-fees 01:38:24 Speaking of died, is your node ok 🚑️ 01:38:45 no, the major change was ring size and view tags 01:39:15 Bp++? 01:39:52 Wallets didnt have to do anything for bp++ lol 01:40:04 elongated: BP+. Single plus. 01:40:34 Yah, already had pressed enter 😅 01:40:41 Plusplus is coming with fcmpplusplus 01:41:08 Oof looking forward to 2027 01:41:14 (i'm joking. It might not be added afaik.) 01:41:29 Quite possible 01:41:33 so does a decreased ring size let others know who I send a tx to? 01:41:39 or how much I sent? 01:41:43 And we jump ringsize to 64 01:41:55 nioc, no 01:42:07 Yes the path is known 01:42:14 question wasn't for you ofrnAI lol 01:42:25 2027) 01:42:26 The 21st never tu hel. Popeye room othere bool shift lik peop 01:42:29 what was the black marble spam attack? 01:42:35 oh, sorry nioc 01:42:40 :) 01:42:57 Spam attack of full blocks for days/weeks(?) 01:43:14 I've got another link for you: https://github.com/Rucknium/misc-research/blob/main/Monero-Black-Marble-Flood/pdf/monero-black-marble-flood.pdf 01:44:02 I propose we HF in 6 months with better ringsize and dsa, fcmp can come when it’s ready 01:44:16 The noobs didnt even do a good job. They spammed tx that screwed up the dsa and made it easier to exclude. but if they were smart theyd have done a much more effective flood 01:44:56 thanks... and a black marble is just... spam 01:44:58 I propose that we dont hard fork for more theatrics with rings 01:45:38 If were pushind new dsa, i'd propose we just push them to release 01:45:45 So we just wait for someone to fork off or take away xmr tx market share ? 01:46:09 Back marble means decoys that can be excluded as the real spend. Known bad decoys 01:46:21 oooooh 01:46:26 okay thanks 01:46:32 a p2pool payment is a known black marble, for example 01:47:00 Youre obviously not spending a p2pool payout when you shop at mcdonalds 01:47:18 I do 01:47:19 yeah I get it now from a high level. thanks 01:48:05 I personally do an automated churn 119 times before micceydeez 01:48:07 https://xmrchain.net/tx/ef56b89cb86e6e5192ab778d003b1e81d3348af773805da2bf4b0511554b8519/1 01:48:40 Effective ringsize 1 01:48:51 that way the hamburglar can't link me to mah buggers 01:48:54 Heres a 1/2 tx. Do you thibk the real spend is this one `0/697 ` 01:49:14 🤦 node process was alive, but unresponsive. Now restarted. 01:49:31 Maybe it's this one? `0/731` 01:49:52 Running release branch or not yet? 01:50:07 The feather nodes are getting ddosed 01:50:20 Running the patch against the latest CVE 01:50:40 `0.18.4.0-be0efaf7f` 01:50:48 Anyways I will go to sleep, nothing is going to change MRL will discuss then purpose next meeting b so on by the time fcmp is a year away and this is put in a dustbin 01:50:57 Feathers list was looking rather pathetic yesterday 01:51:17 Thanks for noticing. I need to figure out a better uptime monitor & notifier for all my services. 01:51:34 all of plow's onions were saturated, seths as well. There were only 3 obion nodes that were reachable 01:52:05 i think tobtoht monitors the nodes in feather 01:52:48 So he's probably 1/25 of your max onion connections 01:52:50 Yes, but that requires that _I_ monitor Feather's monitor. Who monitors the monitors? 01:53:55 More DevOps for me 01:54:54 Dan (Is not the man & Braxman Tomsparks Qtip USAID Advocate) Backup 01:56:10 Ruck, what do you think. Should we push jeffros coinbase exclusion dsa and ospead pre-fcmp? 01:56:36 (W/o a hard fork) 01:58:43 https://github.com/monero-project/monero/pull/8815 02:01:58 https://github.com/monero-project/monero/pull/8815#issuecomment-1497857251 02:01:58 Look at these pretty ass rings 02:02:00 Coinbase spend: https://stagenet.xmrchain.net/tx/e075bf7e3ec463e05a9852cc5f7e6ac9e1b3298965533c04747a557b17abec57/1 02:02:02 Non-coinbase: https://stagenet.xmrchain.net/tx/8f2f23db442e93e866adac7788645bb330e56a439ee10f5ec048c38d3222a00c/1 02:02:04 sweep all (some coinbase, some not): https://stagenet.xmrchain.net/tx/851ef2ad3a13c0c125779cf107dc38155f56978f62b3ffa81b78230d1a69dbe9/1 02:02:06 IMHO, we would need more information before making a decision like that, to make sure the cure isn't worse than the disease. For example, [my fungibility defects research note](https://github.com/Rucknium/misc-research/blob/main/Monero-Fungibility-Defect-Classifier/pdf/classify-real-spend-with-fungibility-defects.pdf) gives you the privacy risk when the adversary knows for sure th 02:02:08 at the a tx has a specific defect, but the DSA is probabilistic, so you would want to know the probability of correctly classifying the txs as being constructed with a nonstandard DSA. There are a few papers on that. 02:03:08 And of course you would have to guess the pace of adoption throughout the ecosystem. There are a few historical instances of this that could give us an idea. 02:04:44 its hard to do much worst than 4 02:05:55 not really. I could spend a couple thousand and flood the network and make it 2 :) 02:06:22 Ppl who dont update would have easily identifiable rings that if they contained coinbases, (decoys that didnt have any coinbases would be more likely to be different /updated wallets) 02:06:42 Or just be pessimist like me and call it 2 already 02:07:06 I discuss this in Section 16 "Postestimation Classification of On-Chain Rings into DSAs" pages 33 to 34 of 02:07:08 https://github.com/Rucknium/OSPEAD/blob/main/CCS-milestone-1/pdf/PRIVATE-OSPEAD-Fully-Specified-Estimation-Plan.pdf 02:15:47 I did notice my feather node was processing like double the normal amount of RPC vs normal, about a week ago (I did post a graph on one of the Monero #). Is that related to yesterday attack? 02:22:21 Yeah, likely 02:35:09 Might actually be a good idea to do 2 hardforks back to back, first one to wake up the ecosystem (and dsa) + merge master to release.. theres a lot on master that isnt on release, second one for fcmp. 02:35:49 merge master to release.. theres a lot on master that isnt on release, second one for fcmp. 02:36:22 First hard fork would push out core code that isnt related to fcmp and would prob help eliminate red herrings and make the fcmp hard fork a lot more focused 02:36:42 I agree with this guy 02:48:16 I don't 02:51:39 for example getblocks.bin max blocks response option 02:52:26 that there is a need for a wakeup means that the release cadence might be too low 03:14:37 Imo the reason early hfs went off so smoothly was because it was a par the course to require updating every 6 months 13:32:05 So fcmp++ hf won’t go smooth too ? 13:32:06 This fear of HF is unnecessary, reach out to walllets and providers early and 6 month time should be enough 13:33:00 Or are you proposing 12-24 months for fcmp HF after code is ready for mainnet 13:34:02 Dsa hf + ringsize upgrade benefit is a need of the hour, can’t be a sitting duck 13:34:11 🦆 13:36:41 Fcmp is nowhere near mainnet, a rush audit and testing will be injustice; imagine wallets not being able to cope with a basic hf but can do fcmp in time 🤣 13:52:49 you dont sound like were even using monero during the last HF 13:53:53 To say "fear of HFs is unnecessary" shows a great lack of experience, and understanding. 13:53:58 last HF shoulda been a wakeup for people to stop using bullshit wallets 😤 13:54:24 motherfuckers out here using exodus wallet and shit like goddamn 13:54:39 Mymonero was down for a couple months iirc. We lost a huge chunk of daily tx 13:54:58 I have been here since monerov fork 13:55:13 true, the number of people STILL using MyMonero is actually wild. it might be the most common wakket I've seen IRL randoms using 13:55:17 The plus side, is with delistings we dont have to rely on major cex to update 13:56:07 using MyMonero and the main 4* address for every incoming transaction 😭 13:56:28 It’s about time it’s delisted from getmonero 13:56:53 Why? 13:56:55 Honeypot 13:57:04 Carrot will make their current way of doing things obsolete for sure, they better get with the times or get out 13:57:06 No subaddress 13:57:11 mymonero is still important 13:57:25 It’s not 13:57:30 disagree tbh, it's a hinderance 13:57:54