<ooo123ooo12345[m> "What are you referring to as "..." <- I was actually hoping YOU could answer that, since you are unhappy with the current centralized option.

I guess your suggestion is that researchers should make due with their personal computers then? 'Cause it seems to me that many researchers benefit from faster number crunching, and I think a centralized resource is better than nothing.

> <@busyboredom:monero.social> I was actually hoping YOU could answer that, since you are unhappy with the current centralized option.

>

> I guess your suggestion is that researchers should make due with their personal computers then? 'Cause it seems to me that many researchers benefit from faster number crunching, and I think a centralized resource is better than nothing.

Call these researchers ?

* Call these researchers that benefit from number crunching ?

* Call these researchers that benefit from number crunching unfeasible on personal computer ?

* Call these researchers that benefit from number crunching unfeasible on personal computer.

ooo123ooo12345: Your statements suggests to me that you do not fully understand Monero's privacy model.

Rucknium[m]: Can you pick concrete statement and explain what is wrong ?

I really don't care to argue about 0-conf, though I do want to note my personal opinion is that it's insecure. I don't believe anyone should deny that. I do ack it's secure enough, especially given Monero, for small payments, especially in an in person setting, yet it also voids the trustless aspect of crypto some people are here from. In the end, it's a personal decision to enable it, and there's no point arguing about it. Solely

about decreasing 'finality' (provable security in this case) time and increasing 0-conf security.

BusyBoredom[m]: I believe that statement was about Monero constantly re-encoding, not my priorities, though that has also come under scrutiny. I can confirm Monero constantly re-encoding is stupid :p

As for what was wrong with Ristretto.... nothing. It's a great solution. It's just arguably unnecessary now that XMR has sufficient other checks (if it didn't, we'd be gone in a few minutes lol). It also doesn't have anyone who's made a PR for it yet.

And also, while trying to comment without fueling further drama, I'd like to note I do believe ooo generally talks more than they contribute. While research is discussion, I have yet to find these talks discussion, rather than people talking at each other (which I've also been guilty of in the past). I hope they're able to find a more productive way to conduct them in the future.

* is wrong according to your opinion ?

kayabaNerve: Any reference example of someone who talks less than they contribute for comparison purposes ?

<kayabaNerve> "And also, while trying to..." <- patches with insecure code would be counted as contribution ?

I can generate 100x per day probably or even more

... a lot of people, who I'm hesistant to ping and drag into this. Rucknium[m], who recently chimed in, most of the people considerable as the core team (though mooo is great at answering random questions about the node which I love <3), koe who spends most of his time working on Seraphis...

Though it's not about finding a reference. It's about this. > Does my conversation have practical effects? > What are the benefits of these effects? > How long is the conversation?

kayabaNerve: "talks more than they contribute" how to measure talk and how to measure contribution ? examples are needed

... this conversation right here.

You're not going to get some perfect textbook answer. Therefore, no practical effect. You are talking and antagonizing me though :p

As one other note, I've seen a notable amount of cleanup PRs recently from a few different contributors, and it's really great to see :) Gets people more involved while reducing the Monero footprint and increasing code quality.

kayabaNerve: examples can be found in real environment too

kayabaNerve: cleanup PR example ? reducing footprint and increasing code quality example ?

<kayabaNerve> "As for what was wrong with..." <- few thing that prevents to be gone in few minutes vs 10000 placebo PRs, which one is more according to your opinion ?

s/thing/changes/

* few changes that prevent "to be gone in few minutes" vs 10000 placebo PRs, which one is more according to your opinion ?

If it wasn't for the fact that amount of epee work from various parties is going to end up as a full rewrite at this point, I'd say ristretto is the extremely valuable one. It's also just 3 functions, yet also an update to serialization code, and updates to all of the existing seraphis work which may be a pain since we hash points into transcripts... Not sure the best design ideas there though. Anyways, not just a few

changes.

I will note one can be done by a few people and one can be done by a lot of people.

While I don't care to say ristretto isn't more notably impactful than rewriting core parts of the system as needed, I never want to suggest quality rewrites of bad code aren't impactful, nor do I want to suggest a 1w ristretto pr deserves more thanks than a 1w lib rewrite with peak C++.

"monero-project/research-lab #100", pretty concrete, verifiable, hard, valuable, isolated goal

it's much more valuable than ristretto

If you will keep silence during few months of researching this task while someone else will do a lot of small placebo PRs

Which one will be more contribution according to your opinion ?

And after 3 months you will appear and start talking and someone else will say "look, your are talking more than contributing;..."

<kayabaNerve> "And also, while trying to..." <- By this logic it would be better to generate 100 arbitrary combinations of static pow algos, instead of developing randomx.

Anything can be compared. If you believe that something is better than define comparison function, concrete examples

* Anything can be compared. If you believe that something is better than define comparison function and concrete examples

<kayabaNerve> "You're not going to get some..." <- "antagonizing" It helps to find edge cases in the best case and make discussion not boring in the worst case.

s/edge/something/, s/cases/interesting/

Also I'm highly skeptical when someone quickly agrees with me. I don't enjoy blind followers

<Rucknium[m]> "ooo123ooo12345: Your statements..." <- "The immediate motivation for this request for 20 Terabytes of SSD storage space is a need to compare the inter-temporal stability of the distributions of the age of spent outputs within and across blockchains. This is important because it sheds light on any risk to Monero user privacy of having a static rather than dynamic decoy selection algorithm." Without explicit

math model to calculate risk It will be just waste of time. And with explicit math model to calculate risk there is no need to look into transparent blockchains. Just immediately develop provably secure way to minimize risk.

Rucknium: Did you learn how to calculate risk ? Or not yet ?

Analysis of transparent blockchains is a way to gather statistics, but it isn't way to prove anything.

<Rucknium[m]> "ooo123ooo12345: Your statements..." <- Are there any comprehensive Monero privacy model already developed ? or not yet ?

s/Are/Is/

<ooo123ooo12345[m> "it's much more valuable than..." <- Arguably. I'd probably agree with that. If no one picks it up, it's worthless to monero specifically, and small PRs were more beneficial though.

<ooo123ooo12345[m> "By this logic it would be better..." <- That's a leap.

<ooo123ooo12345[m> ""antagonizing" It helps to..." <- And while you may find antagonizing others helpful, the people you're antagonizing don't and it decreases your perceived value, and therefore the weight of your word and likelihood people will engage with you. For example, I largely don't believe you ask questions in order to help or in good faith in general. Therefore, I only respond to select comments based on those

assumptions of mine. You won't gain anything if no one talks to you .-.

In the end, this meta conversation itself is reaching the point which is no longer more beneficial than not, and this channel should be refocused to actual research discussions. Feel free to reach out privately. You're also an individual who can do whatever you want, technically.

kayabanerve[m]: Such questions helps to learn something faster. It's the same as critical thinking when you're trying to crack something. In this case it's a try to crack someone claims.

s/helps/help/

<kayabanerve[m]> "And while you may find antagoniz..." <- "For example, I largely don't believe you ask questions in order to help or in good faith in general." relatively to what goal ?

s/what/which/

<kayabanerve[m]> "And while you may find antagoniz..." <- By this logic when you asked about "normalization" for Bulletproofs+ someone could say: "oh, you're antagonizing" and ignore it

Judging by time it took to find relevant line in code It was easy question to answer.

* It was not easy question

<kayabanerve[m]> "And while you may find antagoniz..." <- scammers that don't write code and asking polite questions that don't irritate anyone, is it in good faith or in bad faith ?

<ooo123ooo12345[m> ""antagonizing" It helps to..." <- Unfortunately, the worst case is that we lose devs as they get tired of being antagonized. Most people are here because they want to be, and persistent irritation can mess that motivation up pretty quick.

<ooo123ooo12345[m> "do you mean that work that would..." <- Helooo there!

I have done this many times already and will have to do it for the current month as well: Whenever I see that there's either no demand for my skill set at a given time, or I'm so pissed off by an autistic&dumb person here (that would be ... YOU), I do go to my wooden shack and prefer doing side projects. Yes. In such cases I postpone the payment of my main maintenance proposal (ask luigi1111 ) and await important review requests

from selsta .

So... where's the scammy part here? I still don't get it.

oh dear

<BusyBoredom[m]> "Unfortunately, the worst case is..." <- Oh yes. Then it ultimately makes the Welfare guy the only "Alpha" left in the pack.

So I don't want to suggest any concrete solutions to the mods, but I'm really getting tired of this. Hence my scammy work on my main CCS Proposal is getting postponed. And if not for a few good people here, I'd be long gone. Seriously.

I meant "a few who are audible". Somehow my proposals get financed by silent anons pretty quickly. I feel like I have a duty for you here and for those who just finance me. If anybody wants to convince them that they should stop doing this, go ahead. But I don't feel like I have anything to hide. When I was 14 like you, I invested all my productive time into spaceflight, rather than crypto. So ask me about spaceflight. The things

that I bring on the table are solutions common in both (and from other) fields. If you don't like it, fine. But I'm not you. Period.

What you suggest is a kind of imbreeding the project: Filling it with specialists ONLY of the same kind. The results of this can be seen on my statistics. Plus it's not wrong if more researchers like you join (minus that complete lack of manners), but every project needs an engineer. Some people get it. You don't. Get a job.

Here are the mentioned stats, that you love so much:

Also, by exposing my alias, you really don't scare me, dude. In fact I was considering doing a presentation some day in a future Monero Conference.

What you would do however by exposing other peoples' aliases, who live in restrictive regimes, is organizing them quite a lot of torture. All that they could take actually... and this is a top privacy coin's chat. Have you confused the coins maybe?

please #monero-research-lounge for all meta discussion

Sure thing.

It's very important that this rooms stay research focused and with absolutely no drama. It's one of the few places left where people can come to discuss high quality research and only that. If the conversation is derailing, please just stop engaging or use the ignore button. There are elements that use provocations as way to engage, don't fall for it.

<ErCiccione> "It's very important that this..." <- Agreed. That's the button I'll hit on his alias right now.

* right now. I recommend everybody the same. He doesn't deserve your time.

You can now render mathematical expressions in markdown on github: github.blog/changelog/2022-05-19-re…athematical-expressions-in-markdown. I think it would help a lot with clarity in issues 🙂

(Latex syntax)

ErCiccione: cool thanks

Small in person transactions are a very low risk of a 0 confirmation XMR transaction double spend fraud. Card tap and shoplifting / dine and run etc are a much greater risk for the merchant

Also, I know this from own experience - various customers have various risks attached. This means that with some customers you have almost no risk.

Yes. Customer risk is critical here. For example in Canada there is a debit network called Interac. The debit transactions take longer to approve than VISA. The reason VISA only checks the "naughty list" a much smaller database. Interac checks the actual account balance. By the way debit in Canada takes on average over 1 min for a transaction. So there is a significant probability of 1 confirmation in Monero while a debit

transaction is processing

1 minute is an enternity for a centralized service

Monero is extremely competitive with card payments when it comes customer risk vs time

On a centralized network the size of VISA worldwide.1I mim may be impossible. The naughty list can be sent to the edges of the network,, making VISA partially decentralized. Monero may be very close to optimal here

Centralized networks can be very fast if they are small and localized

<ooo123ooo12345[m> ""github.com/monero-..." <- by "it" here, you mean `socket_.next_layer().shutdown`? I'm still trying to understand why not call `socket_.async_shutdown()` when the server initiates shutdown sequence like in that test

<ooo123ooo12345[m> "MITM + access to client machine..." <- agree it doesn't seem a particularly serious issue in practice

the old shutdown sequence your code in 7760 is fixing hangs on `socket_.shutdown` with your test; your updated code in 7760 just skips calling that entirely. AFAICT it seems the "correct" shutdown sequence would call shutdown directly on the socket, even if the issue of the above attack doesn't seem particularly pernicious in practice