00:00:05 not currently, besides the normal cryptography overhead, but carrot also improves on view tags 00:00:26 so the scan time is still constant 00:00:29 (per tx scanned) 00:00:35 Cool. 00:00:40 but 1/2^24 outputs are scanned on avg 00:01:01 currently it was 1/256 on avg 00:01:26 I still need to figure out things I'm curious about like the network's hypothetical maximum scale and capacity. 00:03:17 @jeffro256: This also makes the whole drama over the OVKs somewhat pointless. In a scenario where governments are already incredibly invasive and coercing people to hand over view key information, THEY DONT NEED OUTGOING VIEW KEYS. They could tell when you received money and didn't tell them the key image. You could try to [... too long, see https://mrelay.p2pool.observer/e/ssajjd8KWWhNX1Q1 ] 00:04:55 "oh no outgoing view key? cool, so next step... yeah give spend keys and we will just look around :)" 00:05:01 This is also assuming after FCMP++, because they REALLY don't need them before FCMP++ b/c of ring signature tomfoolery 00:05:08 Speaking of FUD, people were debating if Zachbxt's "news" was nothing more than market manipulation.. Apparently the timeline doesn't add up. 00:05:31 if they want to do X, they don't care it's not supported, they will ask for what gets X in the same process. or ofc, ask for ongoing key images 00:05:56 @pyratevevo:matrix.org: even ignoring the spike, monero's market price has already had an upward trend for at least a year before 00:06:14 governments will cry if they have to hit you with a wrench over a spend key 00:06:14 @pyratevevo:matrix.org: The idea of Monero being targeted from all fronts, spreading FUD and manipulating the market sound all too silly but highly plausible at the same time. 00:06:19 instead of a outgoing view key 00:07:05 @intr:unredacted.org: I think for the price spike specifically it's probably just a coincidence of many things happening, not just 1 bitcoin wallet getting stolen. 00:10:05 Cindy: Either way, it depends on you whether it comes down to the wrench or not. CARROT can't fix a weak mind 00:10:58 Like, let's say that a whole ass country starts using Monero. Millions of people start transacting in XMR mobile wallets. How "big" does the network need to be to accommodate the enormous amount of transactions ? > <@pyratevevo:matrix.org> I still need to figure out things I'm curious about like the network's hypothetical maximum scale and capacity. 00:16:19 1 million people X 5 txs per person per day X 365 days/year X 6300 bytes/tx = 11.5 terrabytes per year chain growth 00:17:38 6300 is after FCMP++ 00:18:02 Or are you asking about p2p network size ? 00:26:44 @jeffro256: Whoa, that's a lot. 00:27:18 @jeffro256: I was asking about how the network could handle that m any transactions at the same time. 00:27:24 Bandwidth, basically. 01:10:34 Light wallets are also obviously bad for mandatory privacy... for the same reasons. Most users will happily give their OVK to a large LWS provider for the convenience of not running their own > <@rbrunner7> The outgoing viewkeys allow much better "light wallets". You can entrust somebody running a server for that with your viewkey, or you can run such a server yourself. 01:56:09 @just_another_day:matrix.org: Like mymonero did for years 01:56:25 Lws should be not be made scalable 07:48:07 yall are insane 07:48:30 btw the niew viewkeys can offload most of the scanning without revealing everything 07:49:20 you could give coinbase your viewkey and have it scan 90% and still have better privacy and UX than we have right now 07:50:24 Its not like Monero is going to get relisted to build a fragile net of known viewkeys 07:51:14 we are already delisted pretty much everywhere, may as well have a better UX for the people that can still figure out how to find it 08:18:32 I'm not suggesting Monero developers are the ones to have to do this, especially as good information, technical information and explanations have already been published on official channels, but the community as a whole really needs an ELI5 / idiots guide to CARROT to clear away the FUD and absolute crap being circulated in the usual places (Reddit etc.) 08:53:26 @kaigoh:gohegan.uk: https://mrelay.p2pool.observer/m/gohegan.uk/tIxkJWnZuzmIFmmRSFojQxsd.png (image.png) 09:01:10 Something like that, yeah 09:26:44 @kaigoh:gohegan.uk: "anyone with the key can see" would be more correct 15:12:22 so will fcmp require me to create a new seed in the gui/cli wallets? or can i use an existing seed but change the derivation path or something in a new wallet 15:12:54 To use carrot features youll need a new seed 15:15:40 Unfortunate 15:17:22 @kiersten5821:matrix.org: Not required, you can still use the old wallet, albeit without the CARROT features. 15:27:18 16:12:54 To use carrot features youll need a new seed 15:27:20 to use carrot-native specific features* 15:27:50 legacy wallets still use the carrot addressing mode and gain some things, too, just not as much 15:52:19 is there like a big doc that will explain everything that changes with fcmp? 15:52:22 i searched but couldn't find 15:59:27 Fcmp and carrot are different things 15:59:41 There should be docs for each 16:02:22 i must have terrible search skill because i can't find these docs and can't even find how much bigger a 2 in 2 out tx will be after the upgrade... i'm sure that's been discussed 200x by now though 16:02:41 Carrot https://github.com/jeffro256/carrot/blob/master/carrot.md + ongoing https://gist.github.com/jeffro256/146bfd5306ea3a8a2a0ea4d660cd2243 16:02:45 Carrot is an addressing mode 16:09:43 thanks, still would like to know the tx size change though, i remember a long time ago it was more than double or something 16:10:22 a bit more than that :) 16:10:40 I can't remember if we have a single document for FCMP++ part 16:11:11 but you can peek around in my stressnet block explorer https://stressnet.p2pool.observer/ that has the current changes (and we are testing FCMP++/Carrot) 16:11:17 you can compare tx sizes 16:12:59 @kiersten5821:matrix.org: 2 in 2 out is about 7.3kb iirc 16:13:38 https://stressnet.p2pool.observer/tx/3ec7827d3241708fc95af81353d79fb101bf6b30d5f871e7349ed98c55097b01 16:15:14 DataHoarder: thanks, really cool 16:19:58 @kiersten5821:matrix.org: Are you trying to build something with Monero? 16:22:27 anyone here used haveno on tails? 16:31:23 @kiersten5821:matrix.org: Write the Qs of a FAQ and it can be filled in by knowledgeable people. Then it can be posted as a blog post to getmonero.org 16:33:58 @rucknium: yeah, but these are mostly just general questions out of curiosity. i asked the stuff i needed to know in the research lounge and dev chat a few days ago, haven't had many problem since then 16:35:08 Want to give any hints about what you're building or will it be a surprise later? :) 16:36:02 though there were a couple things not in the rpc docs that i had to dig into that i talked about in the monero-docs matrix chat, which is that the change address of a tx doesn't necessarily belong to you, and the pending transfers with the get_transfers rpc call include things in the mempool, but don't refresh it 16:39:44 @rucknium: i'm developing a multisig-based bridge for a guy. but yeah those questions about the fcmp stuff weren't related, just curiosity 16:52:50 @kiersten5821:matrix.org: FROSTLASS is an alternative multisig protocol for Monero: https://github.com/monero-oxide/monero-oxide/tree/main/audits/FROSTLASS 17:05:21 @rucknium: some of these have been answered by you guys already or are answered in the doc, but on the user level i think they will be helpful, of course there are many more i'm sure would be enlightening to average nontechnical users that i haven't thought of. most users are not in this chat after all. 17:05:21 as as side note, when devving, the docs for xmr seem to be pretty sparse on details so i have to ask a lot of questions here. also, a huge amount of stuff in https://www.getmonero.org/resources/user-guides/ is extremely outdated, like the multisig article, when docs.getmonero.org is the up to date version. overall i think the [... too long, see https://mrelay.p2pool.observer/e/-KvCqt8KLTR1WjhZ ] 17:05:22 Will I need to create a new wallet with a new seed to enjoy the benefits of the upgrade?[... more lines follow, see https://mrelay.p2pool.observer/e/-KvCqt8KLTR1WjhZ ] 17:11:05 @kiersten5821:matrix.org: Would like to see the answers to all these questions too, in a user friendly format. 20:45:25 And now they've stepped down precisely to push people away from this trust model > <@elongated:matrix.org> Like mymonero did for years 20:46:07 And yet monero-lws is deliberately made to be a high-performance lws server > <@elongated:matrix.org> Lws should be not be made scalable 20:48:20 I might be wrong but I don't believe the full view key is actually necessary for LWS 20:48:23 Lightweight wallets aren't all bad. I have a way better time with my self-custody wallet now that I can 1) connect over Tor and 2) not ever have to sync anything. For those who do self host, it's the best experience with zero compromise 20:49:26 good luck regularly syncing a full-local-sync wallet with a Tor node. It's slow 20:49:38 @sgp_: Does this mean MyMonero's decision to step down was wrong? 20:50:02 no, I'm just saying I love self hosting my own LWS :) 20:51:02 how necessary is it to sync fully if you already know you haven't gotten any new outputs since the last time you synced 20:51:12 so much so that MAGIC built our own wallet for it :p 20:51:34 Cindy: you need to sync to spend, at least at a high level 20:51:48 is it for decoy selection? 20:51:57 there are workarounds, but no wallets use those 20:52:08 at least no wallet I would recommend 20:52:21 on that note, I've been trying skylight + monero-lws but for some reason monero-lws is not giving out subaddresses despite having --max-subaddresses set to 500 or something 20:52:22 or just checking if the key images associated with the outputs have been spent 20:54:13 @intr:unredacted.org: are you on the latest skylight release, and did you install LWS recently? LWS only recently supported subaddresses correctly* 20:54:13 *there are still some quirks, but far fewer than there were a few months ago 20:54:38 yeah, monero-lws is lastest checkout, skylight is latest github release 20:54:46 latest as in, uhhh, 5 hours ago 20:54:49 IIRC there was a proposal for probabilistic view keys for light wallets that filter out 255/256 outputs 20:54:56 is it still a thing? 20:55:15 I can see from the logs from monero-lws that it's not allowing subaddresses 20:55:21 if you bump the lookahead to 20k, let me know if you still see issues with subaddresses 20:55:23 but I have no idea if it's a skylight issue or monero-lws issue 20:55:29 I'll try that, thank you 20:55:55 no ty for the report and testing it out 20:56:09 give me just a few mins 20:58:03 stupid question but I don't see any commandline options or mentions in monero-lws docs on lookahead, where do I set it 😅 20:58:04 21:51:48 is it for decoy selection? 20:58:06 for a wallet, or for a node? 20:58:08 wallet asks node for decoys 20:58:16 and ofc, for checking spent key images 20:58:40 but wallet syncing doesn't build up its own decoy db 20:58:42 it only tracks its own state 20:58:44 that's why you can "skip sync" at a specific height 21:00:08 @kiersten5821:matrix.org: Beta.monerodevs.org those user guides are essentially deprecated 21:00:46 The new keys are only relevant for new carrot wallets 21:01:07 The new address type is not visibly distinguishable from old ones 21:01:53 max_subaddresses in config 21:02:14 --max-subaddresses=20000 > <@intr:unredacted.org> on that note, I've been trying skylight + monero-lws but for some reason monero-lws is not giving out subaddresses despite having --max-subaddresses set to 500 or something 21:02:40 oh that is the same thing, fuck me, sorry 21:03:35 > they cannot see who sent the money, etc > <@kaigoh:gohegan.uk> https://mrelay.p2pool.observer/m/gohegan.uk/tIxkJWnZuzmIFmmRSFojQxsd.png (image.png) 21:03:35 > unless the sender has shared their view key as well 21:03:43 the servers default is 50:200 (50 accounts with 200 addresses = 10000 minimum) 21:08:32 gotcha, so I deleted the old monero lws database, cleared the wallet from skylight, recreated it, accepted the request 21:09:26 with monero-lws-admin debug_database, there's a boatload of "subaddress" objects, but even after I close skylight, open it up, wait for it to connect, hit receive, still only the main address with a warning that it can't use any more subaddresses 21:09:56 I'm also noticing a little bar below the wallet balance that's stuck loading 21:10:10 it's connecting over my lan, no tor 21:12:28 @intr:unredacted.org: do you select "No Tor" in Tor settings? 21:12:40 Yes, that's correct 21:12:52 22:03:35 > unless the sender has shared their view key as well 21:12:55 it's showing the "no tor" icon below the connected checkmark 21:13:05 they cannot see that 21:13:24 not directly 21:13:36 what do you mean? 21:13:37 but you can replace view key with "cex" 21:13:41 @sgp_:monero.social we can continue this in DM if you prefer 21:14:15 #skylight-wallet:monero.social 21:14:18 can you join #skylight-wallet:monero.social ? That's the best place for this discussion 21:14:19 still. in the case everyone shares their view (without balance) keys, you end up with similar issues AND you can tell the balance of each other 21:15:04 not carrot, today too 21:15:17 FCMP however does cover this 21:16:32 @sgp_: why this link is not in readme? 21:18:03 FCMP patches vulnerabilities of (current) incoming view keys, but CARROT simultaneously introduces outgoing view keys to undone this improvement (i'm assuming "less optional privacy is better") 21:18:34 FCMP++ is about removing rings 21:18:47 you don't need view keys, CEX can do this with outputs they sent you 21:19:16 if you sweep multiple of these at the same time in same tx, they can statistically attribute these back 21:19:18 this isn't only about CEX, merchants can also be pressured to require view keys from clients 21:20:14 what I mean. if you use CEX or merchants and they are just telling "I sent X tx with outputs to Y,Z" they can look back and do that 21:20:24 without view keys. you don't have to bring them to the conversation 21:20:48 it's hard to do, but on certain conditions you could pass this along and tag the outputs of the next tx 21:21:14 will it be the case with fcmp? 21:21:14 it still doesn't give you information on its own, attribution is what they are looking for 21:21:20 no, fcmp++ removes rings 21:21:39 so you cannot do stuff like https://p2pool.observer/sweeps 21:22:02 this is great 21:22:15 (p2pool outputs are known attributed to a specific miner, so you can tell when they sweep in a single tx, that's why we recommend using a different wallet for mining. solo miners also have unique signatures so the same rec also applies there) 21:22:38 but when fcmp is implemented, view keys become relevant 21:22:43 @basses:matrix.org: no good reason. omission 21:23:08 suddenly they NEED them to accomplish stuff like this, besides tracking source ips that dandelion++ attempts to make hard 21:23:39 or at least, attribution. they don't need keys 21:23:44 they may ask for a list of txid and "is this yours" 21:24:12 remember you can't outdumb this, they care about the end result not how it gets there 21:24:55 if you remove any viewkeys in a special wallet (by making a wallet like old bitcoin lol, you lose wallet.dat and it's all gone, no seeds), besides having issues with quantum opponents, you'd ... be asked to just provide proofs 21:25:21 and they can assume you providing these no more = you failed at providing and are evading 21:28:43 or, they ask for a spend key via some third party "safe system" :D 21:29:23 even worse, forced to do multisig with them! 21:31:38 it's harder for them to ask for a spend key 21:31:57 the right to self-custody is relatively well-established 21:32:06 but the right to privacy isn't 21:32:51 it's really a question of how many people will choose to comply with a specific request 21:33:15 i didn't get the "provide proofs" part 21:33:23 tx proofs 21:33:36 it can establish when you received funds like a view key 21:33:45 you can generate these with a view key, too 21:33:46 but how do they know what tx i have made 21:34:36 I'm assuming you gave them a no-balance view key (like a legacy wallet one) 21:34:47 so they know when you receive funds, or someone else sent them to you and they know 21:35:00 i think he means the system shouldn't have view keys at all, like cash 21:35:09 they can ask for key image of each one 21:35:39 view keys are part of how you can have an efficient wallet and scan outputs, it's part of the design not just ... an extra feature 21:36:38 as an example from last conversation https://blocks.p2pool.observer/tx/9c78c972c295f31cf3b30240a22c300aff96ac9d4bd886ef96639fcf3b328271 21:37:08 they know you received a transaction in 9c78c972c295f31cf3b30240a22c300aff96ac9d4bd886ef96639fcf3b328271, they may or may not have view key for it 21:37:17 they can ask for an InProof that shows you indeed received it 21:37:19 but they would need to repeatedly ask for key images > they can ask for key image of each one 21:37:23 and for you to provide key images 21:37:32 and only one time for a full view key 21:37:42 yes. they'd force this and non-compliance would be same as you moving elsewhere 21:37:52 so on https://mrelay.p2pool.observer/m/gohegan.uk/tIxkJWnZuzmIFmmRSFojQxsd.png 21:38:14 "forced to share" it means the same that you got view keys or not 21:38:26 you should switch wallets 21:38:42 @kiersten5821:matrix.org: my point was that we should make the compliance as cumbersome as possible so that it's harder to compel people to do so 21:39:15 this is important specially around a PQ Turnstile https://gist.github.com/jeffro256/146bfd5306ea3a8a2a0ea4d660cd2243 21:39:17 which explicitly needs the generate image key 21:39:35 like, currently many wallets don't even support exporting key images 21:39:47 i hate whenever i make a transaction, i have to export the key images 21:39:54 you can import seed elsewhere :') 21:40:04 or give the server my spend key (which is retarded) 21:40:04 if someone asks to do so, most users will just go to another merchant 21:40:18 again, they will put the burden on you 21:40:19 this is just for a fundraising wallet 21:40:20 they don't care 21:40:22 Cindy: multisig too :D 21:40:24 That (we should make the compliance as cumbersome as possible so that it's harder) is an interesting way to see it. 21:40:25 but when we have a single button "make regulators happy", most will just press it 21:40:50 legacy wallets as they exist keep being supported 21:40:55 if i could have an outgoing view key, i would publish that in a heartbeat 21:40:59 rather than having to stick with a hack 21:41:03 for my fundraising wallet 21:41:05 If we assume a central auth wants to demand it, then: if it is at all possible, we can assume it will be demanded, wnether easy or not. 21:41:20 compliance as cumbersome -> also means you make using monero as cumbersome 21:41:22 so it can't be used efficiently for multisig, used with cold spend keys 21:41:25 Then making it hard to do will just mean fewer people will use monero. 21:41:53 which also means cycling wallets is harder 21:42:04 i'll copy my comment from reddit 21:42:06 Making things hard is effective with stuff like "should I use my node or a stranger's", not "can they demand this if it's a pain for me to do". 21:42:08 "The scenario: a merchant has started accepting Monero. A regulatory agency is unhappy about the inability to identify the source of clients' funds. It recommends that the merchant asks clients to provide the key images to ensure the funds are clean - or face complications. Not necessarily legal complications, maybe just a ban [... too long, see https://mrelay.p2pool.observer/e/hfK3st8KMXNyUlpE ] 21:42:08 The merchant faces a choice: they either comply with this recommendation or refuse. Now consider two hypotheses: the clients have a one-click solution to provide the key images, or no such tool exists. 21:42:08 In the first case, the merchant would likely choose the first option. Most users would comply. Regulators are happy.[... more lines follow, see https://mrelay.p2pool.observer/e/hfK3st8KMXNyUlpE ] 21:42:23 and again, it's not making it transparent and sharing these no longer allows looking at ring elimination as a risk factor 21:42:23 yeah, outgoing viewkeys would make multisig easier 21:42:25 (FCMP++) 21:42:37 wait, this will cook my multisig setup? 21:42:40 In case of coercion, what counts is outright impossibility. 21:42:55 there's already a one-click option, give seed wallet, and users comply for the scams :P 21:43:12 with outgoing view keys, everyone automatically gets the latest key iages 21:43:13 images* 21:43:17 In that case, the centrl auth decides whether they still wnt to demand, If they do, you can't legally use. But they might not demand in the face of impossibility. 21:43:25 22:42:37 wait, this will cook my multisig setup? 21:43:28 without ability to generate key images internally you can't see what's spent 21:43:30 without importing key images or making rounds for all participants 21:44:00 this also simplifies hardware wallets, too 21:44:30 which many can't export key images too 21:44:32 but hardware wallets already work well, don't they? 21:44:40 The salient point here is whether the party bearing the pain is the one making hte decision. 21:45:04 they don't work well, specially if you need to know balances 21:45:06 for that you need to have them online 21:45:29 @just_another_day:matrix.org: It’s not easy, you need the device connected while wallet syncs 21:45:35 not offline 21:45:37 you need to do this ... on each device you want to keep track of this 21:45:41 connect, sync, etc. 21:46:09 you can extract view keys atm, but that still gives no balance 21:46:37 when i'm using a hw, i just plug it when I want to interact with the wallet 21:46:49 i remember a long time ago i was using an airgapped setup 21:46:53 and the importation thing cooked me 21:47:01 and suddenly it couldn't send anything lol 21:47:04 For other coins too 21:47:09 so much more complicated than bitcoin 21:47:32 @just_another_day:matrix.org: For other coins you just need to connect when you want to sign a tx 21:47:38 how would a merchant know their balance in hw wallet in cold storage? 21:47:40 you putting it online is a no-no 21:48:05 @elongated:matrix.org: no. You need to get the address from the hw wallet too 21:48:10 that is no longer a "cold storage" setup 21:48:12 ^ 21:48:22 to receive funds to it 21:48:32 or maybe the saved address was altered 21:48:41 if you don't reverify it 21:48:42 it makes all offline and multisig work as intended 21:48:44 nope 21:48:46 wallet extracts view key 21:48:50 nope 21:49:18 that is an extra check 21:49:20 i'm talking about other coins 21:49:47 like ledger app doesn't allow you to receive funds without plugging the hw wallet 21:49:49