EIP Editing Office Hour #113 (EIP + ERC)

Transcript

[music] Heat. Heat. >> [music] >> What's up? [music] >> Thank you.

Heat. >> Heat up here. >> Heat up here.

>> All right. Welcome to EIP editing office hour meeting number 113. Today, September 15th and agenda for this office hour is added to Ethereum/PM issue number 2216. We have modified agenda a little bit and the current office hour aims at the new authors support more education and taking questions which are process related or blocker related which can be answered in this meeting by EIP editors in order for new authors to be able to move their pull requests forward. So today we have the first one author's

questions uh live Q&A. If anyone on the call has any question particularly related to process please feel free to drop it in the chat and uh after that we would be going forward with ACD related PRs some quick actions and then we'll get into the review queue where again status changes would be prioritized over looking into new proposals because new proposals are generally lengthy. it makes sense to take it async and if there are blockers we can address it in the next meeting. We are joined by editor Yohham today. Thank you Yam for joining the call and yeah maybe we can get started. I don't see any question on

the chat yet. So maybe we can start with the item number two and we will take the questions as it comes. >> All right. So um yeah I will go to HD related. So uh this one EIP7716 move to TW. Yeah please. Okay so right. So this is a new one. >> Um actually this is not a new one. This is an old one but there was some blocker. In the last meeting we discussed that uh previous authors needed approval. So that has been done here now.

>> I see. Okay. Okay. So, this is an existing slightly stagnant. Um, yes. Okay. >> Mhm. >> Uhhuh. Okay. So, that's interesting because do we have an approval from the one of the offices? Yes. >> Yes, we do. >> That's that's good. That's good. Okay. Um Yes. Okay. Okay. So, I want to quickly go over this EIP uh to see if there's anything I can see is alarming. Uh I will also say that I am currently I'm I'm using an large language model to assist with this. But I also want to explicitly read the EIP. So what I will

do now is I will in the background ask an large language model to perform a review on this one. Um and then actually see what it produces. And in parallel I will also read the uh read the EIP. Um because in practice if I uh check the large language model output then uh they sometimes have like they flag things which which are like not right. Okay.

>> Sure. That's a good approach. >> Yes, I think so too. uh be this. So I'm doing that in the background. So now I will read the CP because like uh if you see if you see these changes um just just for for the record, can you also see my mouse on the the >> Yes. >> Okay, that's great. That's great. Okay, because like this looks like a lot of changes. >> Yeah. Yeah, >> that's fine. But uh if this will be like a new EIP then I also want to to to eat it and it should not be that if it's like a stagnant EIP that we uh just allow the uh the um yeah okay so this is already

okay um uh some things to flag because it should not be like there is an EIP which is stagnant and that we then uh just approve these things without like this. let's see if the live flag will also flag this. That's always funny.

Don't do that. Okay. Interesting.

That's quite interesting or stop it. Okay. All right. Yes.

this maybe. >> No, it also does not have any reference to all the ERP. So, okay. Uh-huh.

Very interesting. Apparently the large language model has found distinct items and it wants to override my pending review. So let's see if it actually did that. Yeah. Okay. We can see that these these ones are definitely larger. So these are what >> the lifeline is reflected. >> Yeah. Even though it is not a completely new proposal being a new proposal is always interesting. Scale. Okay.

Okay. Let's see. That was my conscious now. That was also conscious now. Yes. Yes. Okay. What's this? Let's see this add more.

See where I left off. Look, it's Interesting.

Let's see what it's like to post. All right. So what we doing here is um [clears throat] we flex some things. The large language model also flags some things. So, All right. So, we have some comments

which should be reviewed because then we can move it back to to draft. Um, yes. Let's see. Yes. And add more the large leg items. So that's uh that's good. >> Let's move to the next one. >> Yeah. Uh could you maybe refresh it? Okay. Yeah, I I see it is refreshed. Fine. Um we have added the link to ACD related proposals. There are a few more but uh it can be looked into async and uh we can move forward. So the next one is quick actions which is EIP website related. It's a stagnant bot uh waiting for editor's approval I guess.

Sam has reviewed it couple of more times but uh he still needs more time. I mean it was there in the last office hour as well. I'm not sure if you would like to review it on this call but yeah just checking if it is simpler we can take it. >> Yeah. So this is has to do with the CI which was started >> 10 minutes ago. Interesting. >> Yeah. don't see an anthon approval from Sam. Uh so it depends on what this actually does because it looks like 3004 lines. Yeah. I I will not give you this uh this here.

>> Yeah. I mean if Sam had approved this then I could also have approved it but Sam has not done so so I cannot approve it here. Why? What's this? All right. >> Yeah. Right. in the last meeting uh he just mentioned that he would take some more time to review it and this is just following the next office hour. >> I see. Yes, >> I brought it to you. All right. Um so the next we have typo that can obviously be taken care uh asynchronously and um yeah if you go to the review queue what I have done is like in status changes whatever are eligible review PRs listed them here some of going yes some of them are going to review last call and final.

Yep. >> Yes I see. Okay. What I will do with these is um because wait okay >> maybe I should also filter it put the final one on the top so you know that will be faster to review final to last call to review. Yes, but uh I will actually push a little bit back on that because if these EIPs move to final then uh this is also like final in the sense that if you want to edit these EIPs or ERC's then uh all these changes to these EIPs or ERC's needs an editor approval. So we actually need to check if this document is actually final.

>> Yeah. Yeah. Right now it is sorted based on the date it was received. >> Yes. So it it is in the order. >> That's that's good. That's good. So um yeah, what do you think? Because like I think for by moving these ERCs or EPS to review, do you think I need to re-review the entire document or should I just approve it? I would lean to just approving this because it is like open for review. So people can just add their uh comments. Let's see. >> So in review, I think mostly we are looking for if there are any more to-dos. So we don't want to allow any to-dos to get into and obviously the review will be like multiple times. Now there will be text edit now there will

be comments. So a lot of things would be done while it is in the review. So mostly we look for if there are any toos or if there are any like you know EIP which is not advanced as review. >> Yeah I see. I see. Okay. That's that's actually a very good point. So, let's see if there are any to-dos into this document. That's that's actually a very good point because I think we should write this down somewhere if we don't have this written down. So if uh editors move this uh some document to from draft to review

then we should explicitly check if there are like no uh to-dos in in it because only then uh we can actually move it to uh to review. >> Yeah. So editor is not specifically looking into specification. they are just looking into the formatting and it is following the steps that is required from uh for progressing into the next status. So just making sure of that. >> Yeah, that sounds good. Okay, I'm sure is this one second.

All right. Yes, we can we can this. The next one is this. Great. That looks great.

Okay, that's fine. Ah, that's nice that uh Beville is is

moving this to to last hallway. That's that's really good. Yeah. Cool. So, this is an EP for uh forms then. So, that's nice that uh that they Yeah. Okay. So, I think Yeah. Do we have like any requirements for by moving? >> Yes. I I I was about to check like is it for Glamsterdam or is it for uh uh Hegod? If it is for Glamsterdam um I mean of course they can move it into last call but we should wait for like you know uh that uh to move into test net like the meta.

Yeah >> is it I mean uh because this is like an EMI process and not an AC process. So >> I would say that this actually does not really matter because um yeah okay that that's okay this is interesting. Uh >> yeah just for understanding purposes what we have uh shared publicly is if something is going on to mainet we want to move it to the last call only if it is in public test net because then only we will know it for sure that it will be going in even if we are not making 14 days period mandatory because it will be more than 14 days for core IPs we should

have those proposals at least on the test net >> h >> yeah but I see but I see this is almost there because that is the next step I see every every proposal in glamsterdam metab is in review so I think one by one each one of them will start moving to last call so I think this is the right time yes >> okay this is about this >> yeah because there is nothing waiting on stages side be All of them are already SFI. The next stage is included. But for a status, we have two more stages. Last call and final, >> right?

>> Yeah. So, moving to last call seems like a good idea here. >> Yes. Okay. Let me quickly check if this is okay with this ERP. So, the network. >> Before moving to the last call, right? Okay. For one another cycle. >> I think that particular section is for EIP 7723 like when 7723 can move to the last call.

>> It's not for it's not for general EIP here. >> Right. So um that means that the hard fork meta for lamster >> can only move to loll if it has at least two full network bits. >> No no no no uh no I think this particular line as I mentioned it to you earlier 7723 requires a lot of edit and this is again additional correction that is needed. So this particular section in rationale that is that suggests that as assuming it is being adopted EIP 7723 can be moved to last call when it is adopted by two full network

upgrades. So that's what it means here. >> I see. >> Yeah. >> I see. Yeah. Okay. I I get it. I get it. >> yeah this needs heavy text editing. There are lot of like you know things which are confusing at this point. I see. I see. Okay. But since there is like no reference of these uh EIPs, I think then this is fine, right? >> Yeah. Yeah. I think it is the right stage because I checked the trip 73 which is meta e for glamsterdam. All of them are in the review status. So the next status is last call. So it makes sense.

>> Yes. Cool. Okay. That's that's great. Okay. Then we will approve this. yes. Yeah. Yeah. Yeah. Yeah. Okay. I will help with this. Yeah. Cool. Okay. Last call, right? So, I will first do this one. This is review.

>> Yeah. Yeah. >> Oh, okay. Interesting. >> Let's see. Let's see. Let's see. Let's see. Okay. What did I do again? Ah, okay. >> That's fine. To review. So, we move this from stagnant to refuel. And aha. Okay. That's somewhat annoying.

because the uh new rules from the uh EIP bot they will now flag this uh document that has like the smart quotes. So we need the offer to edit this and that means that we have to do this smart quotes. Now we make them non smart quotes. I also don't know what a smart quote is, but I will convert them to apparently a dim quote. I think it's this crazy question.

So, there's two of them right. let's see what I will do is um because I think I will approve this what I will do is I will just merge that thing in if I check what gender has re okay that's fine okay that's cool issue. All right, I will do this uh comments then I will commit this uh right I can activate here

and what we will then do is then it will rerun and we will check if it passes and if it passes then we will accept this uh this this this uh this one. Yep. Cool. Okay. So, we will get back to that in uh well within this call because like 12 minutes, >> right? Um Yam, we have quite a few comments um in chat. If you would like to maybe take a look at it if there is anything that may require your attention.

Yes. Okay. Let's see. Uh I need see things in the chat. Okay. From Cyber Sentry, one question on this uh this one 194 1 1942. Uh oh, okay. That's a comment for for Sam. So Sam has like a pending review on this, I think, that he needs to submit. Um so yeah please uh please please ping Sam like you have a pending review please just uh hit the the submit button. Uh then we have this one a follow up one from Kawei a ami let's see I will do the same thing again

I will point the large language model to this is an very large update I see okay which is fine where do I ERC. All right. So, you cannot see anything on my screen now because I'm doing this on another screen. The only thing I'm doing is that I'm pointing the large language model to this uh PR to review it in parallel. Um, let's see.

Yes, that's this chat. Oh, what notice? Okay, great.

Let's see. Updated. Updated. Updated. let's see what happens here. Okay, there's no comments. Okay, so what I will do in the background, the large language model is now running. Uh I will move to the next point from cyber sentry. So that's this one 952.

Uh this one second. >> Yeah, you on on this one. Sorry Pie, you first. No, no. I was just saying it's the same one now. Cyber is just to give the Yeah, just to give the background. So, um on the EIP and ERC, sorry, the ERC um repo being uh moved, this stagnant bot was failing. And I know that you did one um on something else.

>> Um Sam Sam did this one. It started running through but then it errored. He did the pull request himself. Um I don't think anybody saw that it had errored. So I did another pull request which is this one here 1952. >> Um I can see it's been reviewed by somebody. I wasn't sure if this was somebody internal or somebody that had just reviewed it. >> Um but yeah, it definitely needs looking over. >> Yes, >> I see. Um Puja can you maybe schedule this uh one because this is actually very interesting. So ERC's are not being moved to stagnant. Can you move this?

Yeah. PIP tomorrow maybe. >> Oh yeah, sure. I will add it. Yes. >> Yeah. Cool. Cool. >> Thank you guys. >> Yeah. Yeah. So cyber centry uh this will be reviewed by more than one editors and yeah we will that one. Yeah. Brilliant. That one's Sam's. >> Ah, right. Of course. >> Yeah, I see. >> Sorry. No, that that wasn't Sam's. That was the one that I initially proposed, but it wasn't a pull request. I just put it as a as an issue and then Sam ran it through, but it errored. Yeah, >> I see. I see. Okay. Interesting. Very good. Good catch, by the way, because

this is Yeah. Okay. Interesting. So, this will be like a mass stagnant. Okay. Yes. >> Okay. Cool. >> All right. We will address this uh this tomorrow. But but thanks a lot for this. >> Yeah. Yeah. on on this what loop. Says on this smart quotes. Yeah, I I I also feel that maybe most large language model might be publishing these kind of quotes because if I write something down

like here, I also get like a smart quote. This smart quote, but then I can use it to convert it to like an character. That that's that's I guess nice but not very nice if you publish smart quotes um on your EIP or ES because now the B will flag it. Yeah. Okay. In parallel the uh last language model is uh is working. So let's see on this one how far the B is 6 minutes ago. Okay. Six more minutes.

Um let's see. Okay. Okay. So, we have this and this and this to final. okay. This also has Oh, wait. This is the smart some smart quotes. So, um, but yeah, for for these kind of uh because this is like a very important move of course to move these to final. Uh, but what do we have like a document somewhere to to see the required checks what we should do here as editor? We don't have it documented but I have noticed generally it is a comparison with the fellowship of Ethereum magician if we see any unresolved comment or uh comment that author should address.

>> I see okay interesting >> we also take a look at the open pull request if there are any other open PRs for that particular EIP. So if someone has proposed any changes it can also be encompassed. >> I see. I see. Ah okay. So this does not look correct.

Right. zero for the very schema. Yeah, I see. But this does not look correct because it does not render correctly. >> Yeah, maybe uh this is something that we can add on fellowship of Ethereum magician instead of just this PR. >> Yes. Okay. >> Yeah. And I don't know if there are only one comment then generally uh editors suggest that we should get some more comments to ensure that people are interested in this proposal and finally you know seeing some adoption >> that is not a criteria that is not a

criteria but that is generally encouraged because we are making a standard here and if nobody's engaging with the proposal not very good sign >> I [clears throat] See, I see. Yes. Okay. I will at least point to the table because this table does not look correct. >> Um Okay. And then I also check so when the uh the ERC's moved from the EIPS to the ERC's repository, um what did we do with the PRs open to ERC's on the EIP repository?

They are closed and they are supposed to move here. We just dropped a note to author and it was moved here. >> Okay, great. That's uh that's that's all right. So we check here and then we check this. okay, Let me actually check how this renders

on the the site itself. So it's this there's other one this right and that's This time yes.

Okay. And what I will then do is directly link this on All right. Cool. Let's see what happens to Yes. 1957 update.

Okay. It should post like a pending video, but it doesn't. Okay. Let me first check this one. Okay. still fails. That's not good. Oh, wait. What? No, it's in different, right? I always have too much tabs open and now I don't know what I just Which one was it?

That was this one. No. >> Was that 1997? >> 1997. >> Yeah. Could you maybe take a look at it? moving to find it. >> No. Yeah, it's so it's not this one. It's the one where I just uh push like this commit. is it this one?

>> No. >> So there are only two proposals moving to final in this list we have today. Uh the next one is uh oh no actually three 204 and 207. Yes. Okay. This ah okay does this one. Yeah. Okay. Cool. Sorry. Like I I was referencing to this uh one where I fixed these smart quotes to fix the bot. And that's nice because now the bot has uh finished. What I will first do is I will No, I think that

yes, I will dismiss the genders if you um because yeah I would not know like any reason why we should uh schedule an EIP4 fork. So that's I think was generally requested. So I will dismiss that that okay.

Let's see. Then we will first move to uh this one from Kami U to see what the large language model actually says here if it actually post this thing. Okay, let's see what it what it does. great. So, this is all addressed.

Three small items. Right. Okay. Okay. Okay. So this actually looks very good. Um what I will do is I will push one commit here to uh apply these uh these URLs.

So let me directly do that because I cannot suggest them. RC 8.07.

Let's do it again. going to be interpreted in RS.

See, let's see. The other ones are also referenced. Capitals. Oh, done.

And yes. Okay. I just push that commit. And what we will then do is there. Okay. Let's edit this. Wait. Yeah. Okay. So this is uh this is why I don't like large language models sometimes because I explicitly asked the large language model to not publish these documents because I want to keep them pending and it actually just publish it. So let me actually edit this document.

That's all it was already. because what we are going to do is we are going to approve this uh PR because these are like very um small comments.

Uh I will also trust the larger language model that the points from SAM have been addressed. Um I think this is uh this itself is fine. Uh unless you think this is not fine then I will dismiss it and then I will ref no I I think it is fine. I mean I'm sure that this is going to be reviewed at least one more time. >> Yes I think so too. check the comments here. Wait.

Apparently my edits did not go through. Great, good, great. Uh, yes. Let's prove this one.

Okay, great. Four minutes left. yes this other one should also be nice right okay great um so there is one which is suggested by the uh author that it is a blocker I don't see that as a blocker except for it is ready for review uh point number five if you would like to okay take the comment is already added here the question that he wants to ask Yes, the really issue. Okay.

I see. Okay. I see. Okay. So, the request is in it on re-review. Um, okay. What we will do is we will do the same thing as we did just now because I asked the um it's very yeah somewhat funny because I explicitly asked the live lang to publish a pending review or not publish anything publicly yet because I want to review it. But that's fine. Um Oh, it also said nothing has been published. So, it definitely has. Um, I mean that's that's fine. So, what we'll do is we'll point it to this PR.

Okay, that's runs in parallel. We can also check it ourselves. So, this one is not outdated apparently. Uh, okay. Right. Yes. Okay. Okay. So the question is is this value is this in there? So we have to see this thing

Yeah. Then we have this value. screen. Okay, that's good. So, this seems to be addressed. That's nice. Is this a good So what's the judge?

>> [clears throat] >> See? >> Okay, we we will just go slightly over time because I want to finish this this one. So, the large language model says that all the previews from SAM have been addressed and we will then see if it flags anything else, but I don't think so.

Ah okay. Yeah this is actually a good find. So um okay let me first check if this is actually true. >> Yeah. Puja can we accept this this license? I think not. >> No we only accept CC0 and if there is anything else call for input. >> Yeah. Okay. So so we cannot uh we can we can merge this right. >> Okay. Uh and and what license do we need? A CC0, right?

>> Yeah, CC0 we prefer. >> Um yeah, it's also this one. Yes. Yes. Yes. Yes. Yes. Yeah. Okay. Let me directly comment this uh

yeah Google and likely this is also CC Apache. Yeah. I see. Okay, that's a it's a good find I guess. yes. Okay. So I will directly block Yes. So this will block the uh DPR from

getting merged. Um and so I I have to also I have to go but uh I will do that when I get home again. Then I will publish this uh this large language model assisted review. Maybe it will find more but at least this licensing thing is like a good find. So yeah. >> All right then. Well, thank you so much Yan. Uh we have reviewed quite a few PRs in this call and yeah we hope to look forward to have more questions incoming in these office hours rather than just simple reviews so we can address more and people will be able to navigate their PRs on my own.

>> Yes. I I like this form by the way. It's very good for the office hour. Yeah. >> Yeah. I I would like to thank Sam for the encouragement here. He suggested that we should do it something differently so we can better make use of time in answering questions. Well, thank you everyone for joining us here today. See you in the next office hour every Tuesday at 1600 UTC. Take care everyone. Heat. [music] >> What's up?