EIPIP Meeting 21

Transcript

okay um so we can go ahead and get started this is eip ip meeting number 21. um the first agenda item is single source of truth for eip uh what's that agenda item referring to uh that is coming from the previous discussion i do not see brent here on the call but that was something related to making one source of you know information like at this time we have like multiple resources for getting information on eip one is eips.ethereum.org another one is ethereum dot org slash eip and then we have readme then we have eip1 so there was some discussion on it and

yeah micah if you would like to add something on it uh like um i i don't have anything specific other than what you said that we have four things that all describe the ap process and they'll describe it a little bit differently and they'll describe it a little bit wrong because they're not in sync and so each one is out of date in its own way it would be nice if we didn't have that problem okay that's that's um yeah i think that eip1 serves a different purpose than the readme the wiki and the website though uh yes it should but it's for reasons of history it currently contains a bunch of overlap with them just in terms of what

content it does contain and so the idea was that we would clearly delineate what should be in the readme versus what should be in the ip1 and then once we have that line drawn we can then delete all the stuff from the readme that should be in the ip1 and vice versa and then those two are at least now in sync but then we still have the other problem of we have two other sources that are basically just description of the ap process which again are like draw lines figure out what should be where and then enforce that brent was going to work on this but he's not here so i don't know if he made progress yeah we can probably skip that one if brent's not here to give progress otherwise we might be talking about something he's already worked on yeah sure okay the second agenda item

are edson's comments um they are somewhat some of them are related edson do you want to just go over them and we can talk about each one we've talked about a lot of these before but if people have new commentary or updates on some of these i think that would be good and i wasn't here for the entire last one so if there was stuff that happened last time i'm probably a little out of the loop i should have read the notes my bad uh sure i can go ahead so the first one is talking are you on board more eap editors i think paying them would help and i'm and i am talking to um or kind of help thinking about how to

do that i'm not talking to anyone specifically who can pay us yet but um and even then i i think it would be more prioritizing people who aren't getting paid for anything right now except for eip or they're doing mostly eip editing and that's like their primary thing or something they spend a significant amount of hours on getting paid hourly or something like that i think would be nice um so like i wouldn't want to get paid for that but maybe um greg or micah or other active people would i know grega said that he would prefer to not be paid and so i don't know if we should ask him what his arguments are for for that and maybe see what other

standards organizations do in terms of paying people involved in it because i'm not sure how the ietf handles those sorts of things yeah i think that's that's a good idea to ask him about that micah did you have an opinion on that is that something like if there was an opportunity to get paid for this would you um accept that maybe yes if someone gives me money for something i'm doing anyways then i will accept said money great i'm i would be in the same boat as you except i since the ef is kind of like i don't think i push hours for this but i also don't do it enough to really push hours for it beyond the few i do every every few

weeks so yeah i don't really need any money from it um so that's one thing um i think it's going to be kind of tough during the holidays because people are kind of all slowing down i think that's just something that's just gonna happen over the next couple of eipip meetings and maybe we should have like a 2021 push to recruit editors once we find funding source and look for a funding source i think malloc dao would be a good place for that um right now like no one's submitting stuff to malakdow and

everything that's getting submitted is getting passed like the very few things that are getting submitted um like i think since june there's only been like two proposals and they both passed do you know what the do you have like a short description of what their mission statement is or what their their goal is nope okay okay it's a little bit in um they have money and they're giving it away kinda yeah uh i mean like originally it was really focused on or it was focused on a lot of like non-core protocol like privacy stuff like tornado cache and but then it would like range to other

things too um so it kind of as time went on it didn't really no one really picked up the torch to adjust it for what it needed to be today um is what i'd say but right now uh there's someone named yalda who is starting to do that so there is hope that maybe it um could start speeding up again and kind of becoming a new force but they have like a million dollars in their fund so i don't think it'd be crazy to try to use molok dao for that and we could do that as early as you know early december that might be something to consider so

one thing i was wondering like while we are figuring out like about this budget thing and all but the one burning problem we have is a lot of issues and less additives i i was wondering if we can have somebody who is already looking into it like i mean in my mind like client was there and it was earlier proposed as well he has been actively looking into that so is there a way or if he agrees at all then is there a way that we can come up with this onboarding and at least yourself for the time being yeah if someone like like clients interested and even taking any amount of responsibility in this i'd be all for giving the pro like right permissions

for the repo um for people that we trust even doing small tasks and maybe not wanting if they don't if regardless of if they want to be called an eip editor or not we can kind of work that part out but my viewpoint is if anyone wants to help there shouldn't be a hurdle in doing that for bureaucratic reasons as long as we trust them i'd be happy to help out i don't know if you think it would be valuable for me to like go through this onboarding process like i'm pretty familiar with how things are going as it is because i've been going to all these calls for the last six eight months and i've been involved in the process but maybe if we're going to look at even hiring more editors in the future in the very

near future maybe it would be beneficial for us to do like a dry run with me through this like on-boarding process so we can get some feedback to how that all works because i don't know if we've actually had someone go through it yet since we've like been talking and developing it yeah that's a good idea i'm okay with that uh what do you think everyone else so we have another editor on call oh just micah i think i'm fine with it cool there's a chat with editors where i will bring that up today let me just make a note of that um and that would be adding use an editor um

and practicing onboarding cool that would be that'd be a good way to do it thanks so much for offering light client this is definitely that's definitely really awesome of you to do yeah looking forward to it um and i mean i guess i can't confirm it so i talk to the other editors but i doubt there will be anyone who says no um okay brent just showed up um so if we want we can go back to his agenda item real quick just in case there's an update on that um no problem um it's the one that says single source of truth for eip so that would be having something that

accurately describes each or basically having um each eip1 the readme the wiki and eips.ethereum.org having their purpose well-defined and the text with them well defined uh and different from the others yeah mike appointed me to um one of the guys that's working on that and so i'm coordinating with him and we we've opened a conversation but there's still more work to do on that i'm still work in progress okay great thanks for the update edson we can go back to um what you're saying but to summarize the first point basically funding is a good idea and we have malik dao as a potential option and light client is now probably an editor

okay great um should we run to the next question or should we yeah that sounds good to me how can we on board and assign more eap champions i don't think it's our responsibility to really push that i think if you want to be an eip champion well the one thing is that it might be the responsibility of people who are helping to re-factor the eip process is um clearly defining what an eip champion is so that people understand the roles and

responsibilities and then having people champion when they want to champion i don't think we should be saying hey you can you champion this that's a little bit out of scope i think we should just let it be i feel like one way that we could help improve that process is this kind of goes back to what i was thinking about this summer was some way of having like informal working groups and maybe doesn't even need to be that overbearing it could just be some people who have a lot of experience in specific eip areas like i you know i think like alex berasozzi is like a good person from like the core side in a lot of ways

saying with peter sology and and martin and these are sorts of people that could see like a core eve come through and if it's not so that they have the ability to push for themselves they could say this is something that is probably going to become finalized it just needs to have that champion to go through and i think this happened very organically uh you know about a week a week ago whenever alex and i it might have been on the core devs call might have been some it might have been the breakout room but basically there is an eep that is supposed to limit the size of the account not and this is something that's like mostly specked out already by alex and it just needs someone to take it to all core dev and try and like iron out any potential issues

and just because it came up i offered to like go ahead and and chaperone it through the rest of the way and so that happened you know in a very organic fashion but i think that if we had if we just talk to some of these people who are already like in these roles of overseeing a lot of things that are happening in that aspect of the ethereum network we might be able to just have them begin to think in terms of how to help delegate that work to other people okay hmm i'm gonna have to think about that more before forming a full opinion something that may help is if we can somehow make it more socially acceptable

to for people to create essentially forks of vips so right now we have many eips that are all but done as as just described but just have been dropped and part of the problem is is that only the author can edit it and so i think if we just kind of make people aware that hey it's okay to just copy the eip put your own author on it change the eip id you can now champion that like you don't need to if the other the original author isn't moving forward and you want to you can take it over and that's okay i think right now there's like this feeling that people are like oh i don't want to i don't want to step on anyone's toes i don't want to take over their thing and clearly this person doesn't care

anymore because they haven't been working on it for a long time so some way we can just make that more make it known that's accepted like that's okay it's okay to just to fork a thing just like with open source repos it's become acceptable to fork someone else's work and we just need to do the same thing somehow with the ips i think that sounds good to me that could just be a new expectation kind of set by editors from an editor's perspective yeah i try to encourage people to do that when i see it pop up maybe it's something that cat herders could also push for with their various news sources i don't know if it could be talked about in

on their twitter or just in those videos you guys make or something just something to get the word out there that hey if you see an ip you like and it's stagnant just fork it yeah sure that is a good suggestion we can uh definitely mention that in our newsletter on the people series that we talk to different authors one other possibility i'm not sure if this is the right idea but forking is always problematic because then you lose the single source of truth you lose history and it seems like another possibility would just if if someone's not active on an eip if someone proposes a change and then they're silenced silent implies consent so and then if they come in later and say we miss something they can pull us

back but anyway that's my feeling yeah i don't like forking okay that sounds good is there any other comments on this question it does seem like it's a it is important um messaging problem rather than concrete actions something that just needs to be continually said until it kind of like permeates throughout the community i'm curious what other people think

about the suggestion of that no no comment means acceptance instead of denial currently we are operating with no word from the editor means denied but there's no word from the editor is more common than word from the editor i feel like it's sport you know to me the the outcome that we're kind of hoping for i guess not really hoping for but the outcome that makes it sane to work in ethereum is that unless we have a concrete agreement among people we won't move things forward and so it's different at terms uh in the terms of editors

pushing things through and finalizing standards if we don't have an editor responding then that does not mean that we can move that forward but in terms of the eep i feel like 99 of the time if the person doesn't care enough to respond to someone who is updating their eep then it's unlikely that they're going to have a problem down the line whenever that e is actually pushed over the finish line by this new champion and to me i don't know that's the that's acceptable because even you know if that person comes along later down the line the editors can choose to figure out how to handle the situation appropriately

exactly i think that that's you know the you know less than one percent of the time and the other 99 of the time we get to have people come in and start championing eeps i guess like one of the problems is that maybe they could leave this like path of destruction behind them where they you know change each and then they also don't take it to the finish line so now you have this original author this eep that's like not really what they said but it kind of still has their name on it and it's in a worse place than it was and so that is a bit of a concern yeah i don't really have an opinion um right now on that uh yeah on micah's question specifically i don't have an opinion

right now i guess i'm just like in terms of forking the thing i don't necessarily like about just copying the eep and starting anew is i don't like the idea of having potentially five to ten versions of eep just continuing to be forked along and along and maybe that's just like something silly i have in my head because realistically like we are unbound in the number of eeps that we can assign and there's nothing stopping us from having all these numbers i i don't think that that's also something that's going to happen often i think the most times an eep would fork would be three and that's rare because the first time it would be like oh this eep didn't get much traction let me try

and then it would have to then the author would have to abandon the new author would have to then abandon it and that would be the second abandonment and then it would happen a third time and by the third time there would be a trail of people not wanting it so there could even be like some hinting from the editors of hey this has a history or you should look at the history that's true it's just you know it's not as easy to like look through the history if they're just copies of eeps because you kind of have to go to the eep site copy the title search for that title if the title has changed it's even harder to see if that even has been in copied case it doesn't matter then because if there's enough changes then it really is a new eep

that's true also we do have we do have the ability we do have the thing in the that we just introduced where we actually are marking eaves as abandoned eventually and so that i think will hopefully help reduce the frequency with which we see like many copies that are in some sort of state that is under like anything other than abandon and you can just ignore the abandoned ones or the withdrawn whatever we decided to call it so if if an eve is abandoned do do we should we assume that it should be if someone wants to work on that should we say that if the champion is no longer around to really respond to anything should they copy that even create a new one or because it's abandoned is it okay for us to say okay because it

was abandoned the champion didn't care to get into a place that mattered and so we're going to allow them to like become champions of it i believe we when in this previous discussion we agreed that it would you'd have to create a new one so that way we don't have to deal with the problem of politics and people like saying oh i banned that on purpose because i don't think it's a good idea and i don't want my name attached to it i don't want my history um and i think there's there's a fear of like this can get complicated whereas if we just say hey if you see an abandoned eep that you like fork it like that's much simpler simpler politically i mean probably slightly harder in terms of

mechanics but i think simpler in terms of politics that seems fair also i believe we indicated abandoned eats eeps can be unabandoned by their authors and so that's another layer of complication if we allow someone to take it over like it gets abandoned and the person comes back and they're like oh i want to resume this after all and then but someone else started taking it over and so we tell them to fork like it gets weird if you take a petition model that that may not make mistake make sense here but in canonizer people can explicitly sign a camp and join a camp but they can also leave the camp anytime they want and so

so to me it's it's not a governance problem political problem just a communication problem how can that person either leave his name on or remove his name off the camp so anyway yeah and i guess the thing is though with canonizer though like it's a little easier than an eip where you're having to trace someone down manually that's not like they have an account you can just message like on canonizer good point um i think the other thing like client that i would say is if we if we're doing it kind of the way it already is where we just fork the eip and it turns out there starts to be a problem we can we can you can propose a change or anyone can propose a change to the process um but i don't think

there's a better process that's or i think there's a better balance of process right now that we wouldn't be able to predict things already yeah that makes sense uh are people okay if we say that the official messaging is that people should fork and eep if they want to go ahead and work on it and the informal messaging is that people should make an effort if they want to work on you to try and reach out to the original author and try to take over that eep if possible you know a courteous way of like going about it and i feel like that helps avoid this sentential scenario where

someone is heads down working offline and they start working on this eep and then someone else comes along and you know these are like you know one in a thousand but it just seems like that would even resolve more potential problems i think that also aligns with open source software uh kind of etiquette where generally people first try to go and propose a change or a pr or file an issue with some project and then if they don't get traction or whatever they eventually they'll say okay i'm just going to fork this and do my own thing yep that makes sense sounds good okay um next up how can we reduce the backlog um oh you mean the eip backlog right

edson i think cloned micah a couple more times actually that's a real answer that's good i want a copy of mike each clone is a little bit dumber than the first you should have seen the first one it was awesome that's really funny uh okay so i think having more editors will reduce backlog and that's really the only real solution i think between more editors and then the bot that we've added recently that basically goes through old issues of prs it just says that this is stale

and it's going to be closed i think that that's like brought a lot of people out of the woodworks or not out of the woodworks in which case those pr's were just closed yes i think i agree anybody else have comments we are getting down like we were at i don't know 10 pages of backlogged i think we're down to two or three and the that's mostly due to the bot like the bot has just been slowly working through something keep in mind the bot actually does have a limit on how many it will flag at a time and so it's been iterating through and as things get closed it then opens up a slot where the bot can then flag something else and so it's still making progress i don't know if it's reached the reach the end or not in its first pass

and then of course as things get just abandoned naturally they're also going away on top of that in since uh whenever i started like in july or something mid-july um i believe i have been responding to every but i think i have responded to everybody it's like i think that there is no longer a problem since july of people submitting eips and getting no editor to look at them which was a very big problem previously and so i'm my hope is is that between the bot and me actually responding to people we will get to a point where um eventually all that's left in the pull request list are eips that are actively being worked on i think

that combined with the automatic move to abandonment which will kick in what in a month or something like that um i'm hoping that we have done enough that as long as you keep up with all that we will get down to something that is manageable manageable for one full-time individual um not guaranteed but that's my hope thank you so much for that micah i mean i personally have observed this like a lot of pull requests are now closed um for whatever time like whenever we have requested and personally i have received all the messages from you getting response so thank you so much for this hard work great any other comments on this

ah my phone sorry does does anybody disagree that our current trajectory looks like we're going to get to a good place eventually like is the is the worry that we're not going fast enough or just that our trajectory is actually wrong like we are not heading towards healthy state i write it more as a check-in but edson did you have a specific um meaning behind the uh question uh it was just a question after a long long time ago just uh checking okay okay uh next up how can we reduce the turnaround time um

that actually is like i think that kind of goes with the previous one and um that that is uh just more editors micah's doing good like clients getting added any other comments on that or did you have something else a direction to take that other than its literal direction edson i think we could move on how can we organize working groups for different eip categories oh what's the idea behind this one um i think it was um having

uh i i think something similar to editors who specif who focus on a specific type of eap so someone that focuses on ercs and stuff like that or i think it was a group that creates those kind of eaps i'm not i'm not quite sure cool any comments on that i don't know if we're at the point where we want working groups for different eip categories personally i think we have enough else going on i would like uh just separate repos for the different categories personally

so i can ignore one of them i really i really i really don't like ercs we've talked about separating them before and it always has come up how messy that would be and if we want to split it because then there there could be cause to split it again in the future and it would be a precedent set that we could um between like and then we could split it again for smart contract versus like api standards versus other types of ercs and there's pros and cons and if anyone wants to bring that up again for the future feel free i just have never

i just always like shiver at the thought of going through that you have to create a new um naming or numbering standard so right now we have it with issues and prs just add add 4000 to every issue number that's the new numbering standard or whatever whatever your fork block is nice um that won't that won't write us later [Music] we've run out of space um uh let's see um uh any other comments on that one i may have a question uh so

not directly related to this particular question but uh somehow like we are talking about having different repositories and uh different categories of that so my understanding is we agreed that no more meta no more informational eips would be there but at least for one eip which i believe would be going forward about that serenity phase zero would there be any any other place where it will go or if we are just going to get it in the final form because yesterday we had some talk with the author and we are moving i mean the author is proposing to move towards like changing the status of the eip to the last call and then eventually to the final status so

is there anything else to be done or that is just defined are they already already in review so for review section uh yeah that that was the case and we discussed that ideally it should have been in review two weeks back when they actually added the spec the final respect for uh this upgrade but since we did not do they did not do it is it okay to just skip this stage for now and put it in the last call because there are only two weeks left till they go for final deploy so that was a question i mean i i just didn't realize they were already in review

um that's all like they should i would say a move to review right now if they haven't yet and then b are they okay to move into last call soon it's up to the author i would say i don't think we put any hard requirements on length of time you have to be in review right uh yeah go ahead so the author currently is proposing that if it is fine to skip the in review for now and right away go to the last call still they would be getting 14 days which would be like you know the time by the time they would be reaching for that genesis so is it okay to go ahead with i'd say so from a process standpoint i think so as

well i'm still not a fan of that eip but i've lost that fight so i'll back down okay cool i'll just communicate that okay um i think that's the end of edson's questions edson did you have anything else uh no i think we should track the uh funding part of it and just more progress on getting more editors uh i i guess you said next year so we only have like a month and a half left we have a month and a half left there's definitely things we can do in that month and a half but nothing i doubt we can do anything substantial based on just the previous years when things have slowed down during the

holidays kind of a thing um okay thanks edson for the questions those were good next up is the item three the network upgrade process medium post puja do you have an update on that uh we already have shared this network update process with the all codem meeting and this would be live today i have shared the link of the medium it will be live and shared with rest of the people we haven't received any comment or any suggestions to make uh i mean like a corrective suggestion to edit on the process part that we earlier decided on or maybe discussed on so yes that would be going live today

great um anybody have a comment on that okay we also have item four eip status medium post i think that's probably puja again do we have this is that something that's in progress i don't i vaguely remember something like that um i think that is a medium post in progress uh with edson and uh uh it's in if you have anything to add or maybe next week i don't know uh i have a draft i should be ready for review and posting this week cool um onboarding eip editors

expectation guidelines um we already went over that and as one of edson's questions i think unless there's something else that this item refers to no i think that is pretty much covered there but yeah if anybody has any other comments all right review action items from the previous meeting um uh the decision 2021 is actually i don't know if there's any action items there's decision items i think that's the same thing uh no that's just decisions we made so we don't have any action items from the last meeting i don't believe yeah i think so um is there anyone with any outstanding things that

didn't make it on the agenda or anything that they want to talk about i had this one thing in mind like now that the network upgrade repository is uh separated from this entire process i had a question like uh should we i mean is it is it the right thing to do to start considering issues and pull requests that comes over there i mean in this eipap group also that eipip is uh like we are talking about in eip editing so can we bring up issues which are there and heavily discussed bring into the agenda from let's meeting onwards is it okay sounds good to me so wait can you uh rephrase that last

part real quick though like so my question here was like uh whatever issues that keep on getting discussion on other forums and is related to definitely of related to eip or process or the network upgrade is it fair to be a part of agenda i think as the as far as the agenda is staying this small yes if it starts to get unruly we might have to cut some stuff that might be lower priority or less relevant who is um so for the eap repository we have eip editors which are the people who do the like editing like making sure things flow through the process is there a similar

group for the network upgrade process or is it just james pretty much james puja and uh tim baiko okay james's lead but tim and pooja help yeah we are just helping with uh keeping it up to date and uh there was this like a pull request that we received so i was wondering if that is something that we can consider here and talk about i don't know i'm not sure because i don't see james here in the meeting so [Music] okay uh if there's nothing else i think we can uh stop the meeting and

the next meeting is going to be on december 2nd at 1 500 utc i think i may have one other thing i'm just scrolling through chat history i thought i asked something couldn't didn't get a response oh yeah the the merge bot the emerge bot seems to randomly turn itself on and off i do not know where it lives or how to make it turn on or how to kick it um it'd be great if those instructions were somewhere that i could find and more people have access to it who are reachable because when it goes down it usually just stays down for days and it kind of breaks the process and requires a bunch more work from me for example hmm who runs that do we know i think i do

not i think nick johnson nick wrote it uh nick does not so like for that nick wrote it nick may also be technically the operator but i haven't actually seen nick for i i heard at some point in the last year that nick has it running on like a heroku or some some like you know platform at the service that could be wrong but that's what that's what i thought that i had heard a while ago that would not surprise me so then the follow-up question is is can we bring that under the wing of someone who is more active and available that's a good idea um do you have you do you have a like a

talking relationship with nick micah like do you know how to reach out to him on twitter or otherwise i can i can talk to him i'm currently boycotting twitter i do see him in the ens discord like i know i just said that i haven't seen him for 12 months but i have seen actually seen his name pop up and he announced this chord like three times in the last two months so maybe i can reach him there oh i didn't know he had discord huh yeah he i mean there's there's a person called nick johnson who's answering enos questions sporadically in the unis so i think that's him real is it the real nick johnson who we will never know you'll never know suddenly someone gives you like a broken heroku box that doesn't have real code in it they're just like there you go

uh okay anyways um cool anything else from anybody does anyone want to volunteer to take that over i would rather not be the person in charge um i feel like i don't have the skill set required maybe i do like what would i don't because i don't know the code or anything um i don't either like id in a perfect world we would have someone who actually can make changes to it as well not just who knows how to turn it on and off but in a less perfect world but better than the current world we at least have some one and knows how to turn it on and off i will talk to nick and see what's required to jump start it and then when it freaks out i'll learn how to just restart it i think that would be a good medium right

now yes that would that would be i think a notable improvement just just literally anyone who is reasonably available and knows how to kick the box and this is the merge bot right yeah the merge bot like client has already done wonders with the uh ci piece okay sounds good i can do that uh anyone have anything else i was gonna say maybe longer term we can look at like malloc to see if maybe i don't know if it's an editor that's also like you know really inclined to work on the merge bot i you know i would like to work with the merge bot but between full-time work helping with the repository doing the ci bot

another thing is a lot of a lot of extra stuff to have on top of the other responsibilities i have maybe as things cool down i could help more on that but i don't know if it would make sense to have like another editor who's kind of like pet in the eeps the organization is the merge bot itself yeah okay so we can think about that longer term like if you talk to nick and we can figure out a way to just have it under one of our own control and i'm happy to help like be a part of that and you know be available to like help kick it to get it started going again i just don't have the bandwidth to like really contribute to it to make it do the things that we want to do right

now [Music] no problem yeah it's understandable yeah on top of just being able to kick it we have discussed a number of improvements that i think everybody has agreed would be good there are a lot of great things we could do to it yeah it's one of those things if we had money to hire a full-time engineer there's a bunch of cool things we would like to do with that engineer but we do not currently have an engineer or rather we have a bunch of engineers none of which okay uh you went mute for a second micah was at the end of your talking it kind of sounded like it cut out i believe so i don't know where it cut out but i

think so okay cool okay um awesome anything else from anybody all right thanks everybody talk to y'all in two weeks thanks thank you