I will work on this one, any contribution is appreciated bounties.monero.social/posts/91/twofish-addition-to-monero

You only mention speed as a criteria to decide which of several implementations to take. If their speed is roughly comparable, I think it would be a good idea to take other things into consideration as well

Like code quality, whether it seems there is still somebody in charge, etc

Amount of unit tests

And then licensing questions ...

The first .c file you link seems to be part of something that is under GPL. Not sure that will work out.

<rbrunner> "You only mention speed as a..." <- For sure. I am going to consider any suggestions. If you have suggestions let me know (or comment on the bounty).

Alright then :) Will do.

<rbrunner> "The first .c file you link seems..." <- I have to talk about this with koe too.

rbrunner: Thanks. I appreciate it

(The comment can be better because we can keep track of the discussion).

<rbrunner> "Like code quality, whether it..." <- Very good point. This is going to be a little subjective. What do you think best approach to find out the better implementation with better code quality? Something like poll between core developers?

Yeah, "code quality" is of course subjective to some degree. "I know it if I see it" comes to mind :)

Yes, why not a little poll. For example, you come up with the 3 best candidates in your opinion, and then people can have a look, with this in mind.

Can't remember however somebody went to such lengths for a selection process, most of the time it seems we trust the dev making the PR ...

I see. I will try to create a small poll. It is better to ask people for my first contribution, instead of contributing and then realizing it was the wrong one.

monero-project/monero earns you a lot more respect than bounty hunting

M0xffc[m]: lol

Try things, get the criticism on them, then use that to improve

ofrnxmr[m]: What do you mean? it will be contributed to monero-project/monero.

No need to "create a small poll"

Could of done a css

recanman[m]: Okay, that might not be accurate on this one. Because we are choosing one.

For example, you find this arbitrary thing in this file. Make an issue if you cannot solve it, or make a PR if you can. Then, people will respond, suggesting changes, until it is good.

<M0xffc[m]> "I will work on this one, any..." <- May I ask, where would TwoFish be used?

I have a basic understanding of the math used in Monero, but not technically

M0xffc[m]: I mean, "my car has problems. Needs a mechanic. i ALSO want a a new sound system"

Oxffc gets me a sound system

> May I ask, where would TwoFish be used?

Koe is better equipped to answer that.

Instead of fixing or reviewing anything

good morning

morning everyone

<M0xffc[m]> "I see. I will try to create a..." <- I would maybe start with a GitHub issue that brings up the problem that twofish solves and get some discussion going on the topic.

Not sure I it’s a Monero issue or a Monero research lab issue

0xffc doesnt understand monero

It’s okay he’s new and wants to contribute

Bounties arent where you make steady income or earn respect

Its where you go to find

a) mostly wants

I don't think he understands the flow of contribution, which is fine

The best way is to do some existing issues, present new discussion topics /issues if you’re capable and then just create a ccs for dev funding if you’ve proven yourself

recanman[m]: This is what i mean

0xffc: you should attend monero research lab meetings

chesterfield[m]: This us what i mean as well

Or any other workgroups, so you can understand Monero's community better

ofrnxmr[m]: There, you can talk with the other MRL members and understand the application of TwoFish (if there is any), and begin to plan stuff

oops, meant to reply to myself

Creating bounties is not a good way to start for a number of reasons- bounties that get a lot of funding are typically things the end user really wants and that want is shared among a lot of people. A highly technical bounty like ‘we should add two fish’ will not get a lot of bounty funding, and it bounties isn’t a good place to help people understand why it’s needed.

Have a look at some of the bounties there 🫠

0xffc: It may seem like a very new concept of this volunteer-led thing. I've only really used Monero and been in the Matrix rooms, but never participated. Within a week or two, I began to understand how everything works

* I've only ever really used

If you found that twofish could speed up sync time by 20%, then you could maybe get ccs funding or something for that

I think this discussion here went off the rails pretty quickly, with some serious misunderstandings.

In my opinion, it’s best to start with the problem that the feature solves

ofrnxmr[m]: that bounty is not for steady income. Just for contribution.

Twofish is set as one of the crypto things we need for implementing Seraphis.

See, that’s good to know.

Now somebody needs to go out and find a good implementation.

rbrunner: Yesyes

M0xffc[m], in my understanding, is doing that now.

M0xffc[m]: I just realized you made the bounty

I would say this is more of a CCS thing, because this is something that needs to be done

But 0xffc (ive spoken in private) thinks this route is efficient

* just realized that you made

when bountys is largelt fringe work

recanman[m]: Id agree 100%

Since this CCS seems to be backed by koe and rbrunner, I believe it is going to be approved quickly

It's too small for a CCS. It's a couple days work max, to search, evaluate, test and judge and handful of existing Twofish implementations.

I must be psych 

Should I propose ccs for this? Does look much of work to be honest. Thought small Bounty would be enough

rbrunner: Oxffc Wanta to contribut a lot

This seems like a gray area

rbrunner: Exactly

But keeps looks at bounties for work

And thinks bounties are where the work is

Too small for CCS, too important for bounty

In this case I think bounty would be fine

M0xffc[m]: We have ccs for 1xmr before

ofrnxmr[m]: Yeah. Later. For now I want to finish Twofish thing

Do you have a resume 0xffc?

But the reason its goos as a bounty, is competition

* resume 0xffc or list of work you've done?

s/goos/good/

Github repo? As Reśume

ofrnxmr[m]: In this case, I have changed my mind. I think a CCS is fine.

recanman[m]: Look. I am not going to expose information about myself. And these contributions are gestures of good will. I am ex big tech engineer working at system software with more than 500 million customers.

DanrdarkIsnotthe: monero Github reviews and prs work as a resume too

recanman[m]: Thank you.

M0xffc[m]: I also have a very strict threat model, but giving me some sense of what you are good at isn't bad

At a high level

Not where you have worked at, things like that. Resume is too broad of a term

The reason why I created it as bounty is I don’t want look like it I want to milk the Monero community. It is gesture of goodwill. And if there 5,6-10 monero as money. Awesome

Snowden level

Well you failed there

DanrdarkIsnotthe: Not at all, but that's a different discussion

A gesture of good will is helping other devs with their prs and reviews

Can i just say have a look at other funded proposals, 

Css holiday fund 

Not "only working if and when there is a bounty"

I don't understand the bounty that way. I understand it as something like a donation address. Nowhere I see "I only lift my sorry ass if the bounty gets high enough"

Lol

thats exactly what it is

Ngu fix

Bounties are bribes

Exactly?

Are you serious?

And the bribes go up if nobody tries to capture it

I'd rather 5 xmr go away and there be a confirmed person working on the twofish implementation, rather than having anyone do it. As long as the entity is qualified and the implementation is needed, I'm willing to donate. But in terms of what 0xffcc is saying, I am not sure if it is for 'goodwill'

Bribe is a stronger term, but yes, a bounty is dangling carrots

For anyone to catch.

ofrnxmr[m]: I want bribe 😁😁 (kidding, everyone is so serious around here)

Again: Nobody has to *implement* Twofish. There are more than enougsh implementations around. It's a vetting and selection job.

rbrunner: Oh

M0xffc[m]: Im trying to bribe you 😁

M0xffc[m]: It may seem so, but we all have a sense of humor. We are humans, after all

rbrunner: Exactly. I will just use already available implementations without single line of code inside twofish itself

#takemymoney not "only people who csre about twofish's money"

M0xffc[m]: What is the criteria for selection?

For nioc

Can we give this hopeful new aspiring Monero dev (something we need, oh boy, do we need that) some benefit of the doubt, please? Just for the sake of an orderly discussion?

thx dan :)

They are not here to ripp us off, and scam us, but maybe just chose the wrong tool?

Should have given their Monero address for donations while they work on this Twofish selecting job?

recanman[m]: The information about objective ones will be on my test program. I am going to write small test program using all the available twofish implementations doing latency, bulk etc measurements. For subjective ones I suggested a poll between core developers. But those are suggestions. I am fine with any suggestions you have :)

As long as TwoFish is needed, I would say that a CCS or bounty is fine. This is some gray area.

Instead of making a bounty

Bridge is awake?

* As long as TwoFish is needed, I would say that a CCS or bounty is fine. This is some kind of agray area.

* is some kind of a gray area.

rbrunner: Yes yes

Yes, he is new, he needs some guidance, but seems to have the right intentions

rbrunner: Or just claim twofish bounty as part of their ccs work

ily? Fixed

As long as he gets it done, that will be beneficial, and he will also learn more about the system

recanman[m]: At the end of the day, whatever core developers decide. I am on board. CCS/Bounty. Even without money. I want to get this done. Bounty was some extra money 😁😁

(Rbrunner is on irc, i believe bridge is awake)

Core devs dont decide anything

Community 

ofrnxmr[m]: That is very important ^

Koe isnt a core dev, hes just like you

There was a page on getmonero.org describing this, not sure where it is

ofrnxmr[m]: Who decides? Community? Interesting.

It is community based

M0xffc[m]: s/core developers/community/

But yeah there is big brain people here

M0xffc[m]: For 99% of things

recanman[m]: I like that approach

recanman[m]: For some things, centralization is unavoidable. Like ownership of getmonero.org.

Ban

And generalfund and core team handle that stuff

Shamir key getmonero

<rbrunner> "Twofish is set as one of the..." <- What’s the main reason to choose twofish as block cipher vs others like chacha20? Nothing against the twofish. I’m just curious

Not sure. I think for the situations that it gets used, you simply have no "place" to store an IV

Wasnt luke aka maid talking about it dev or research lab?

* about it in dev or

Just to empathize it again to avoid any confusion. Particularly it is sensitive topic. I am not going to implement nor make any change to twofish implementations whatsoever. Just use the available open source ones. And the process of choosing will be as open as it gets (poll in github? Or something simple).

No Wallet Left Behind or Monero Research Lab

Not a poll. Just ask

Discuss in ^

<valldrac[m]> "What’s the main reason to choose..." <- matrix.to/#/!toFcRZtpaiwiyapgVO:mat…g&via=libera.chat&via=monero.social

Discussion from here and onwards, koe's messages in particular

<JulioTTC> @everyone @here PRIVATE TIKTOK NUDES 18+💋 discord.gg/7qa7m7WGKD

anyone willing to pay for porn of a cute tgirl in monero? 🤔

.. i am not a bot just like curious xD

like custom request things

gross

ew

any recs for graphic designers/artists that'll take xmr?

vostoemisio: has done the graphic design for monerokon

 

Old reddit theme of lemmy