Ethereum Cat Herders EIP Editing Office Hour, 20 Sep 2023

Previous Meeting

Next Meeting

⏯

youtube image

►

From YouTube: EIP Editing Office Hour 25

Description

Agenda: https://github.com/ethereum-cat-herders/EIPIP/issues/262

A

Welcome to EIP editing office, our meeting 25. This is issue number 262 at eipip GitHub repository. Thank you Sam for joining us today, and thanks to all authors who have joined so um Sam. If you would like, maybe we can yeah go ahead with screen sharing.

B

Sure yeah, so, let's get into this, we have um do. We have any authors here today that want me to look at their Eeps.

A

They are right, so we are joined by uh Stephen. Stephen had three proposals for today's and I think one is already merged. So let's go ahead with the.

C

Second, the pr.

B

Number, this is the one here yeah. Okay, yes,.

D

This one here is for my team, perfect.

B

Good I think you've been through this this Rodeo enough times, I, don't think, there's gonna be any problems with your proposal.

B

um Good test cases.

D

Yeah, this is the second one where we discover a huge Improvement, so yeah.

B

One yeah this was the uh like multi-character emote problem. Yes,.

D

Yeah like after getting to final, we discovered like we didn't support. We supported.

C

Only like 40.

B

D

The Korean emojis.

B

Yeah, okay: this is totally fine.

D

Yeah with the next one we are we, we learn our lesson and we're actually uh doing something in production first before trying to move it to final, but we think it should move to review now. Yeah.

B

Okay, that makes sense that was 7409, and did you have a third one or.

D

Yeah I wanted to move to review. uh It's 7508 sure.

B

I can take a look at that one here.

B

Okay, any utilities for 721 all right.

B

Can we expect a smart contract without the need, for the token to be aware of the external smart contract, for example, represented game character with a set of attributes? It can be used in unrelated game with the same stats, without the need for treating this attribute from an opportunity Source. This ensures that the data the game is using is legitimate, not tamper with, in order to mean touch. Okay,.

B

So is this like a key value store, that's like two levels of keys, one being like the token and address and then like then, a key Value Store per token address.

D

Yeah, the idea is that you can set any yeah any kind of property or attribute into any existing token, but it is the creator of the collection, the current owner, of the collection, who sets which attributes are valid and who gets to update them. So.

A

D

Could be only the issuer of the collection or a collaborator they can configure or the.

C

Token, owner.

D

Or some specific address, it's very flexible, makes.

B

D

And for to store string attributes so for the keys uh when they are strings, we we have like an extra mapping where we map, uh like the string keys to an ID, so we use that ID from then uh for the mappings instead of using a string which is more more expensive, yeah. That makes sense.

B

So are the um the keys like namespace, so if it's set as a string initially, can it be updated to a non-string later or.

D

uh That's a good question: I, don't I, don't remember right now, I think I think you could use both actually okay, so you could have the same name for for two different types of attributes because on both Setters and Getters they Define the type. So, okay.

B

So, like the same key with a uint and the state like so, if you had like the key, a and a u n could have a different value than the key, a string that that makes sense. Okay, okay,.

B

B

Pre-Signed attributes, so this looks like 7 12..

B

Is that yeah that this has to be 712.

B

Let's call it, you call it the pre-signed.

D

B

D

Should include it definitely.

B

Yeah because it looks like um 712, but it doesn't explicitly say: 712. yep,.

B

B

Yeah otherwise I think it seems great. It's perfect.

D

B

D

A

Cool, so the next one is um two proposals: actually uh they are six eight zero, eight and six eight zero, nine. They both are merged to last call, and here Arthur is Nick. He had some questions for you sure.

B

E

Yeah, so the speed at which we emerged was fast and anticipated um um with the pull request uh before the deadline. How much sooner can we actually merge into final.

B

uh So when you set your last so sorry you uh which PR did I just merge for you.

E

um It was six io7685 and seven six, eight, six, all right for ERC or eip6808 and.

B

Yeah so um anytime, after your last call deadline, you'll be able to open up or request to move it into final okay. Yep um make sure that there are no changes left that you want to make, because once.

E

F

B

Final, there's there's no, no editing it all right.

E

Okay, thank you. No.

B

A

Next, we are joined, By Conrad and the pr number is 7484.

C

4 all right.

B

Did you have any particular.

F

Questions: hey um no I didn't have any particular questions. Actually I was just wondering if you could take a look at um kind of yeah kind of merging the initial initial draft. Basically yeah.

B

Sure, let's take a look: where's Mark.

B

Oh So, based on the description in your abstract, this is like a like. An auditor can vouch for a particular smart contract and say you know, I believe this is secure and then another um smart contract can check this registry when somebody adds a plugin and then determine whether it's secure or not,.

F

Yep yeah exactly kind of more generally, it's like an attestation registry, so you could also use it for non-security related kind of about the stations. But but that's the main use case.

B

Have you seen the ethereum attestation service.

F

Yes, yeah, our code is actually um or our reference. Implementation is actually um inspired or took a lot of inspiration from the.

C

Easier specifications.

F

C

B

F

Yeah proposition.

F

um Actually I had a question on that comment. Is this true, even if kind of we mentioned um 690 as an example of of kind of like one implementation, yeah.

B

Like yeah, so, okay, like, for example, if you mentioned 6900 and today it's an example of you- know a 4337 account or whatever. um But you know in in two weeks the author of that EIP could completely change the meaning of it and then your link wouldn't make any sense.

F

B

It's just about, like you, want to kind of move them together to final.

F

That makes sense. Okay, that makes sense. That makes sense, because we didn't uh specify that this EIP requires it, because it's not in the specification, um but but that makes sense anyway, yeah. It's.

B

Okay, first solution: security considerations.

B

Of the market was required certain statements about these security modules and can require chain.

B

S create on geneticizations, okay,.

B

On transitions made by the security of module, not General, okay, well, that seems fine, so and adapters means for a little excellent contracts to Smart accounts. To query the modern virtue before using a module in order to create account must implement the adapter either natively in the account or as a default module. This is on the center and account deployment.

B

B

Choose this okay.

G

B

So if I put the comments over here, it's like not at an editorial related, it won't affect getting merged or not. um So it's just just my my very poor technical review.

B

F

Good we've actually towards the comment. We actually got that feedback already and implemented. uh I think it should be just above the comment you're looking at now.

C

F

That's what the registry must have heard, yeah.

B

Well then, thank you uh just get rid of that. Okay.

C

B

Okay, so that all looks good, so the only other thing is.

B

um This is always like a bit of a subjective kind of concern, um so um you don't really need there's no objective answer to this is basically what I'm going to say, but uh why does this need to be an EIP? So if you were to, you know, deploy your uh registry somewhere and uh you know tell everybody how to use their published documentation.

B

um Everybody could read your documentation and like integrate with your adapter. So generally, eips are meant to coordinate between. You know, independent people who um like need to interrupt, but for this it seems like everybody would be interrupting with with this one registry. So uh in your motivation, you should probably set out like some kind of explanation, for why this needs to be an EIP, and maybe the answer is.

B

It needs to be an EIP, because you know that's the best way to get visibility for it and that's an okay answer, um but that is just something that I'd like to see a little bit more in here is: um if there's going to be one registry adapter, why does it need to be a standard.

F

Yep yeah, yeah I think that makes sense. We actually thought about that quite a lot as well like before kind of uh taking the the step to to make it into the Erp as well and actually I. Think, like the most straightforward answer, is kind of a bit weird in that. If the EIP works, then there's only a single registry that everyone uses and because of that like the adapter, is actually like, very straightforward to build.

F

But um there could also be a case, and this is quite likely, even if the IP kind of is there, that people build alternative Registries and then the kind of like the use case of the EIP is that, even though these Registries kind of like do their own work or kind of like resolve these attestations or whatever in specific ways, all of these adapters for the like smart accounts or smart contract wallets, are kind of like make all of these existing wallets kind of backwards compatible with the Registries that pop up.

F

um So it's actually kind of weird in the sense that like if it succeeds, then there isn't. They need to kind of like standardize, because there's only a single entity right but actually like without the proposal. There might be multiple more entities and then.

B

Yeah yeah yeah.

F

This is a bit weird to explain, but but yeah I think that makes sense to to kind of like rethink the motivations and bring that to life.

B

Perfect all right.

B

Where are we off to next Pooja.

A

The next one is uh with the author bofu here, and the pr number is 7682.

H

H

Yes, hello, hey.

B

Did you have any particular questions about.

H

ah Yes and I saw Lair is a change request from the um Panda pip, so he mentioned that we should specify which metadata should be extended. Then, and I also provide a feedback, so I, so I think I just want to clarify um I pass a link in the zoom chat and it's a common and I reply. um Other Pips input so I want to know, and should we should we mention the idea in the common creating or should I really narrow down this VIP tooler, specifically at Peaks.

B

um So that, whether it's specific or not, is up to you, um let me give this a read and I will give you a better answer. Yeah.

H

Thank you very much.

B

B

So mining here is not in the blockchain sense of the word mining.

H

uh Yeah, it's more uh AI ml data mining, so this I think the idea of this EIP is, uh if I'm the content creator. So when I um create my digital content, then I have some ways to register my prominence. The one is I can just register and as the outtrend metadata. So let's uh so I can in this case I can Leverage The, eip-70, 53 and or I can also meet the ownership over the license. Lp. Then it will go to the traditional ft, an EIP like the 731 over 10 50 55.

C

B

Okay, you said you're supposed to clear mining preferences for digital media content. The business units for ERC is.

B

B

's kind of concern here, without giving like an exact uh like which metadata you're basing off of it's hard to tell where to insert this mining preferences. Blob um so like in in I. Think.

B

Where's the metadata Duty, so I'm, assuming it would just be in this properties array here for 721 right.

H

um Yes, so for 721 and for like uh 1753.

B

H

So because I I, think and I I'm not sure it's like like because like uh 731 is super popular, so many uh senders or eips Will based on it. So uh um so the idea is um I trying to mention as the first layer of like then the lower layer as possible, so that I don't need to mention a lot of all the standard speakers. It's not possible and always they will have the new standards coming so I try to make this a proposal as a festival enough as possible. So I think that's the idea.

B

Yeah so like I I, do understand why you don't want to tie to a specific um EIP. So maybe what you could do here is just include a few examples of um like in integrating with other. So you could be like here's an example of a 721 token that has a mining preference. Here's an example of 1155.

B

um that might satisfy a panda pep here. um If you haven't already, maybe you have I'm not sure exactly.

C

B

B

Okay, of course, okay, accessibility, so.

B

B

Okay, relaxation.

B

Camera, it's okay um definition: okay,.

B

Yeah so I I I can't really force you to because obviously it's your standard but I would very much recommend finding a different term than than like mining preferences.

B

um Just just because mining is is a very specific term in Block space, but otherwise I think this looks really good, um just a few small things and uh it should be good to um yeah.

H

And yeah and thank you for the input and for the um Russian red snail here I want to cancel. You is about the adoption so because, like I I mentioned in layer is actually with a there are. Actually there are some ways we can mention how the data be used for the AI ml training, so we're trying to refer to many existing centers, including the c2p standard. So that's why uh so we're trying to Leverage The Extreme standards and trying to summarize the which we think?

H

Let's continue this one of the best way to mention and the usage of the data here. So it's a it's a it's a kind of rationale or is still like the motivation.

B

um So like so, the motivation would be um why you want to include AI consent into smart contracts, um whereas the ration that would be like why you chose to base it off a particular standard versus a different standard. Does.

H

B

Kind of clear it up a little bit.

H

um Yeah so I I think I I. um My understanding is that an original is an explain. The technical decisions, so yes, so and for the adoption part I, think our decision is we're trying to respect the existing standards. So that's why we do not divide the like some format by ourselves. Instead, we try to extract the existing standard from the existing existing standards yeah. So, let's in the adoption part so but.

B

Yeah, yes, I, think yeah reading this again, I think adoption seems fine here.

H

B

Yeah, so what is this? Is this like a standards, body or.

B

Noticed interesting.

B

H

Yeah uh yeah listen to described how to inject the metadata into the digital media locally. So it's just like um just like every different media format like jpeg PNG levels have different headers, yeah, so c2pages you can think it's kind of the spec mentioning how to inject the metadata in the content. Header.

B

Okay, Okay, so yeah I would just maybe let me go back over here.

B

Okay, yeah, so yeah adoption is, is totally fine to stay in rationale. I think um yeah. That makes sense. Okay,.

A

B

A

There is a yeah, please go ahead, I.

B

Was just gonna say, I think I think it's just um it might not have be like moving it. That's important, it might just be rewording it so that it doesn't sound like you're talking about the whole proposal um and I think that would probably be the best change. Let me just update my comment here.

C

Yeah: okay, okay, cool.

A

So I think there is an issue uh that is a 7700 which is related to this uh pull request. I wonder if it is by team of tofu, and if that is, is there a question.

H

H

um Actually, I I don't know about this. um This issue actually because I I also don't know uh who is the uh Mike h33.

B

Yeah there is I think there's a Spam bot that just scarves people's comments, yeah yeah.

H

Sorry about that, but.

B

C

H

Actually, I really don't know who they are. The only thing I know is that I saw a this. Eip was posted on the Reddit, so I guess it attracts some people from the crypto Community to be here.

A

No problem, we just wanted to make sure it is a spam and not any member of your team, so we can relate it. Thank you, Sam for deleting that.

H

Yeah I can probably slide that we didn't do any spam Bots. Oh.

B

No, we don't, we don't think it's you. It's just.

C

We wanted to make sure they weren't asking a question.

H

Yeah, we're not just try to pick it yeah.

A

All right, I think there are a bunch of other pull requests. Those are added in agenda. um I, don't see author present on the call, but they have requested this to be reviewed in the chat I mean in the comment section of the agenda. All.

C

B

So we have PMP get back to the agenda here.

B

All right so just go through these ones here or were there ones in the comments.

A

I think there is only one last call: let's go review that and then we'll come to the yeah. We are sleep requested by matthas.

B

Let's Call 0927, okay.

B

Okay, so this probably actually needs to be updated unless that one's also being pull requested. Let's check that out first, so five, five six eight.

B

Okay, so 5568 has not been updated yet.

B

It's in review okay, so this can't move until that is changed.

B

Okay: five: two six, nine.

A

So the next one requested here is 7535.

B

Right, what do we got going on here?.

B

Okay, so we have somebody adding themselves as an author, do we have an author approval? We don't.

B

This is core okay, so this has already been approved by gender, so this should merge. It just needs.

B

Okay, so they approved it. um Okay,.

B

So, let's see everything's passing there good.

C

B

A

And the next one is 7336.

B

C

B

Domain namespace interface, allowing for interactive capabilities across multiple registers like ens.

B

Ure to facilitate on train registry entry, as you change the structured up, some nothing type allowing for interactive capabilities across multiple.

B

Accounts: okay,.

G

B

Ty accounts.

B

The necessity for you to find method of data retrieval across variation is apparent.

B

S seamless reading data from every branches to another. Definitely only keychains information is proposed.

B

Single the plug contract or treat records from different combinations of Registries, such as CNS, okay registry. We don't care it's a signal. Contract enables to retrieval.

C

B

C

B

B

B

B

Okay, the circumference implementation neat and then down.

C

B

B

All right so there's interactive across multiple registries.

A

B

I think I'm probably going to call it for today unless there's anything else that needs to be done.

A

So I mean it's fine, it's not like. We don't have author present here, but there are a few miscellaneous PRS which are not a status change, but they are just a small changes to existing peers. Yeah.

B

I can go through some of those.

A

Six two yeah.

B

Did you see this one's here preventing circular block finish with initialize.

E

In that case, I'll, probably head out, uh thank you for answering the questions and you guys have a great day. Yeah.

B

You too thanks for coming. Thank you.

E

A

Looks like they are just waiting and they are not waiting on any specific review.

B

Yeah, so this is technically waiting on an author review um and it looks like the author does not want it merged.

B

So I'm going to say: oh yeah, okay.

A

Actually I see Oren as an author listed for erc6551 and he had made some changes here.

B

Yeah, it's like some spelling spelling changes right.

A

So I mean like.

B

Jaden Wendell and author, yes, so yeah I, think I'm I'm, just gonna prefer to close it and then, if they want to reopen it, they're welcome to.

A

A

All right, in that case, where we do have couple of more core proposals. They also seems to be like small changes but again approved by author I. Guess.

B

So this one needs an approval from axic, but.

A

um I think gumbo is also the author for that.

B

A

Me see: oh no, it's only expensive, okay I'll. Let them know I have a call with the team. So I can let them know.

B

Yeah I can just put a comment down here. Oh yeah.

B

Okay, so we want that one.

A

Almost similar I think.

B

A

For the same proposal, yeah.

B

Okay, I won't uh make the same comment again.

A

Okay, so the last one here is for chance: I think they are just waiting to be merged.

B

Yeah, it looks like uh just need to review from the authors, um yeah I'll, just ping, a panda about this.

B

C

B

So I will take care of that one when I hear back from Panda all right. Thank.

C

B

C

B

Guess I'll see you all in two weeks.