From Idea to EIP: A First-Time Author’s Journey

Transcript

Okay, we go with the presentation. H this is based on my experience on ERC 80003 uh ERC20 pre-initialization extension. Uh you can scan the QR code to to see more details about it and I will start. So uh why create an EAP? So this is a question that was made in the public I I guess uh from people like me that was why should I create an AP? I see all that specifications. I see how valuable they are and I think I have a new idea that can improve uh the protocol can improve the in infrastructure. There is new use cases that can be made consensus in the in an EIP. So this is very

important. So we need to have something to work on that is needed or is going to improve Ethereum. So we can document parties patterns that are already happening but are not documented. So we make a standard as well or to coordinate the community around a common solution. We saw this in the past presentations. So second research and self validation. We think that we have something that we are going to improve Ethereum somehow. Now we have to research and we have to self validate us. So we avoid wasting time of others uh from the newbie perspective

going to a forum taking the time from the experts that they they they are kind and they will reply us and they will try to explain but we have to to take advantage of all the effort. So don't make uh Ethereum lose time uh for you if you didn't make your own research. So first if you have something that you want to propose if it's code do it do it in your own code make it real not just an idea and go there and write down it it's valid if they if it makes a lot of sense but anyway you have to validate the idea.

Second, after you have the code of you have the idea or the consensus or whatever, start to find errors. You you have to go against yourself and also to your contact, some dev, some Ethereum people that is also improving. Uh find limitations. Uh try to push it to edge cases. uh iterate as it says until it survives your own critical analysis to the maximum. Then create if it's a playable a mini site, a repo, a documentation with exactly a technical rational of it, use cases, examples and metrics and whatever

else that you think will uh go forward to justify why are you doing this and how it works and it really works. So in my case in the example of ERC 80003 I had a problem that was gas fees spikes. This ERC is about pre-initialization token uh accounts ERC20s uh before have balance. So why I did this because I want to prepay the initialization of the slots of their ERC20s. So when you make the first transfer of of of real balance, it will be 33% cheaper. So I had a problem. Then

I was testing. I create some contracts. I deploy it and I start testing. And then I had metrics. So finally I got metrics of how much do you save uh depending the gas price and the gas price now when you pre-initializate. So about this I can show very easily how I did my proposal. Here I started I put my repo then I justify it then the background motivation proposal specification. The specification is very simple but here I put this extension route reference implementation in solidity conceptual savings calculator live and

real testing results that are very important. So in the calculation that is live you put gas price current price calculate and here is information. So this is just to show you that you need to prove what are you doing and very important learn to propose and discuss and this for me was very important because there is something that is called net ticket that I don't know if this term stills working was when I was 20 25 years ago and was like behave on the network as well there is some rules there is some some important formalisms And in every forum that you get there is

a way to communicate. And I found this proposal that is not mine with a lot of activity very useful because was a lot of exchange of opinions many cases because uh some person proposed something other person came to say I already proposed it come with me and then they say no this is different. And then more people came and and start showing why is different why is a different point of view which are the differences. So a lot of justification in both sides happened also some improvements in the middle as well.

People saying okay but I want to add this but it's a thread. So it's very complex because in the meanwhile they were getting a conversation of this should be part of another ERC that is already approved or this is a new ERC or EAP and in the meanwhile I was suggesting to change uh some function and this is in the middle of the conversation and the author has to take care of of all this but also other people came supporting the author or or better say the proposal to answer as well and you start getting aligned at and the end of the day without this alignment while the discussion continues

with sense of course that is in almost all cases you can't move forward because if not it's a waste of time so this is is very interesting how it happens I really recommend this this uh discussion in particular uh I really like it And well now comes the proposal timeline for newbies. We already saw what happens on the validation side. All the iterations, all the logics, all the workflow. But for us that we are making this for first time. First we have to have the problem context an abstract motivation specification rational security link to the repo and following

the standards the EAP1 that is the cure code there it's very important to read it you can read it before you can read it after makes a lot more sense if you read it after a week researching the machines if not it's too much specification for nothing because you are you are standing on nothing uh then you To make the pull request, you will receive an EIPRC number. Erh uh when the the request pass the first filter, then you will include this number in your prop. And here start the problems. You have to rename your file.

You have to do the pull request. You have to change a lot of information inside your your uh MD file and the bot comes. That was a nightmare for me. er the a review bot throwing thousands of errors. It can it can give you eight items but inside an item can be like thousands of errors and then there is one error but when you solve it your you solve it it's fine. So now come 100 errors because the error that you solve open a door for finding more problems.

Uh so the best is to debug manually. Don't use GPT. When I ask a GPT about my presentation, erh GPT try to remove avoid using GPT. He say this is a nice joke but it's better to say in a different way. No, don't trust him. It's very nice but don't use it for this because I tried to debug first on GPT and I generated more errors and it's nonsense. Uh so well uh do it manually. Take the time to read to learn. Um and then you have to attend a call uh with Ethereum cut herers uh to understand the process. Try to go to the calls, listen, listen, listen and then

request to your EAP or ERC to be added into the agenda and you have to be patient. You have to wait for your slot. Maybe you're lucky and and you get it because other other pro other proposals don't show up and you present it to the call. You have to listen the feedback very carefully uh and then address adjust and repeat. And here I will show again what happens with my proposal. When I attended the call I was on this stage just here and on the call this issue was open an edge case. try to do this and show that the AAP the ERC in this case

don't broke. I made it very fast. I made all the tests. I documented it and luckily it it passed the review. So implementations and integrations after you are passing and go and moving forward it's very important to file real world integrations that demonstrates the usefulness of the EAP or ERC. So you have to start working on the traction because this will help you a lot to get approved and added if is your goal to to to give some value. If not, you are just giving an idea that is very cool, but no one is going to use it. And then adapt

the protocol on what you learn and create a sandbot envir environment for others to test. And very important, keep the proposal alive. Nobody is going to make it for you unless you are doing a a super cool great proposal that maybe you are doing, but nobody see it before it's working. So for the year 803 I'm trying to contact open select people to review and implement in part of their contracts the year 20s and contacting wallet providers interested in pre-initialize their users as a service to check if there is internet interest or not. So recap your first EAP where to start write down write down an specific Ethereum problem erh that is

bothering you today or you think that is bothering other people. Uh look for similar APS. I also hear some questions before about this. There is a forum the Ethereum machines. You can search there. You can also search on the repo of Ethereum. Uh and I think maybe in the discord also uh the Ethereum research uh and this will be enough do a deep a deep looking and in the next 30 days if we say I start today create a small prototype try it invite people to try it id on ethereum micians start your post very sorted out with the requirements to

make life easy to the ones that reviewing it and get fe feedback from people who already published an AP and open to talk. So thank you for having me here. I hope I clarified some some people path to publish an EIPRC and well thank you pa. >> Thank you Herman for sharing your experience and giving us a shout out. Yes, of course the call is called as EIP editing office hour. Please join us if you have question if you're firsttime author or if you in general want to understand how the process works. You may have uh followed Sam's earlier conversation. He had a message that we

displayed in the beginning of this summit where he mentions that this is authordriven process. So if you want your proposal to become an Ethereum standard, you have to be around you have to be around to push it to get into a network upgrade process as well as if it is for application if that is an ERC as well. Well, we are encouraging people to come up with different ideas and specs and implementation and be an author and be a contributor to to this big wide open ecosystem. Um, while we have German here, if you guys have any question about like becoming the first time author or trying to maybe document a proposal, anything you would like to ask

with him? AOY, great talk. So my question would be it was obviously a lot of work for you and 33% gas saving is quite significant in some way but were you doing this primarily in an altruistic way or were you doing it to try to learn more about the ecosystem through having direct contact or to see about the adoption of standards or what was your real motivation in some sense given the huge volume of work that it required and given that you don't internalize all of the benefit that you are going to be producing if this is adopted.

My motivation is always saving gas. I love to save gas. I'm from Latin America. We want to save in everything. Uh and and I found this uh that it's interesting because we hate gas spikes. Gas spike for us is a pain makes impossibilitate us to do something. While in other areas h it just means I will pay more in the global south but in Latin America happen a lot means I have to wait and I can do it until the gas go down. So, this is an improvement that I think is saving 33% of gas on something that maybe uh is $1, but maybe on a gas

spike uh writing a balance of uh a new token could be $30, $40 and you want to be in because you are following this launch science long and you are going to get out just for this. I mean, it's an improvement is very very small effort. is just adding a small functionality a one small functionality change how the storage is started on the contract in place of u int bytes 32 and then the interface remain the same and it's a pre-initialization is optional so I think why not >> that's great I know everybody love gas

optimization and whatever form it is if it is for like main network or for applications So thank you so much for your contribution and documenting a proposal. But we still have few minutes if we have any more question perhaps we can take one more that could be related to this proposal as well as if you're curious or you want to learn more about how to become an author if you're considering that one as well. All right, in that case, thank you so much Herman for joining us here and people please find him back the stage and you can have more questions.