FOCIL, Deprecating BLS, Fusaka releases, Staking Gathering and Devconnect - Staking Nerd Talk - 5

Transcript

Intro

[music] Hello everyone. Um, welcome to Stick in their talk episode five. We are together at Dev Connect, which is quite rare for us, right? We don't see each other very much often, >> but I'm I'm glad we uh were able to see each other here. Um, I'm joined by Yoric >> and I am joined by Remy, beautiful Buenos Cyrus. >> Yeah. And we have some stuff for you um to talk about today. We have first of all fossil. What do you want to say about fossil?

FOCIL

>> What I want to say about fossil? It is not about dinosaurs. [laughter] >> It is the fork choice infor enforced inclusion lists and uh it is not coming into glamsterdam. I am very sorry guys. >> I'm sad about that one as well. >> So why do we why do we want it Remy? >> Uh because that keeps this um censorship resistant, right? >> Yes, it does. So what it is is um clients will create inclusion lists and they say this transaction must go into the next block and you know the block is invalid I think if it doesn't go >> yeah I had a quick read at the EIP and I believe that there's like 16 random validator being selected >> y to select um some transaction to be included well they will publish some

inclusion list and then the next um block builder will have to include them in the next block right >> or the block is invalid that's my understanding as well in my setting as well. >> The uh testing team at the EF ETH Panda Ops said, "Listen, this is going to delay Glammsterdam." Um, if we do this because we need to do a lot of testing and we don't actually know what we're testing for because we have no experience with it. So, we'll need to do a shadow fork and this takes time. Um, and so it gets pushed now to Hegota Hegot. Heka and Bogota.

>> There's no good port. >> There's no good portra yet. It's Heka and Bogotaa. He go whatever we call it um bokeh [laughter] and uh it's get pushed to there. However, not all is lost. Uh POS and team have and and and friends have a really good idea and he's probably going to set us straight that this wasn't originally his idea. He's just championing it that we do this optin after glamadam. So the inclusion list can be created the execution layer client can verify that it was obeyed or not. And that gives them really good data they need to know. For example, so when you think about Amsterdam, right, we'll have EPBS. So we'll have like 9 to

10 seconds to to >> Was that confirmed uh EPBS being included? >> It's still on still the headline. I mean nothing is said it's out. It's it's still the headliner. Everybody's assuming EPBS goes in alongside the block access list of BAL. So you know, I'm I'm going to go with what we have right now as to scheduled for inclusion. PBS is scheduled for inclusion. >> Um, so with that, we have more time to uh do our anticestation in I think 9 to 10 seconds in the 12 seconds slot instead of having to do this within the first 4 seconds. Um, which means that where was I going with this? For sale.

>> Uh, yeah. >> Yes. Uh, >> censorship resistance. >> Censorship resistance. Something. I lost my train of thought. It's terrible. Um at any rate we can then have the abil ah this is what it was. So when we have this much time what will happen is you you build your inclusion list and you say this transaction hasn't moved from the mele in a while. It must go in but you're already like 10 seconds into the slot. There is a really good chance that the transaction is in that block already. um you just doesn't know it yet. So because you haven't seen it, right?

>> Yeah. >> So the um the idea here is to run this on mainet to gather that data to see how many of the transactions that end up in the inclusion list were already in any way and then that informs the testing that pandops can do. So they need to know that kind of stuff to to tweak parameters and do some testing around for um exactly which parameters don't ask me I'm not deep in that EIP but that's what they need and so that can then make sure that it really does go into HKA plus Botaa because they have already done this as kind of like a dry run everything works it's just not

consensus relevant the concern is that there may be some timing iss issues or like they might >> some timing issues or there's a way to optimize it so that you don't have so many things in the inclusion list that already got included anyway. >> There could be like a >> so but because you're broadcasting issue as well. It's not an issue so much, but you're double broadcasting. It seems wasteful, >> you know. Um, so yeah, that's fossil. Hopefully, it goes into the hot fork after Clamster. And the idea is to really test it out on mainet as optional to get that data so it can go into the next hot.

>> I'm crossing my finger because um we need fossil for a bunch of other different uh changes that we want to push forward, right? So um that was just one part of the whole ecosystem issue around fossil. But let's move to the next topic. Let's talk about uh the deprecate bls.com. Right. So this is a website. >> Yes. >> Uh a website and a uh lobby effort to uh enable uh stakers who are still on type zero validator to potentially um include a withdrawal address and have access to their staking. uh value on the consensus layer.

Deprecate BLS

>> Yeah. So, specifically, I think it's a um it's a community member, Wood, I think. Is that right? >> Uh I can't remember exactly, but >> yeah. Um and what he's interested in is helping people who have lost their memonic. So, if you're one of those, uh check out Deprecate BLS to make sure you're not getting scammed. Uh check the uh East Staker Discord there for the right link. Um, you can also know that you're not getting scammed when all you need to do is to with a CLI tool create a signature that proves that you own the validator. That's really all >> and I believe the signature is based on the deposit address plus the key store.

Well, the the validator signing key >> and we're using the ETH staker deposit CLI for it. Right. So, if it's the official E staker deposit CLI, that's the right tool. If it's any other tool, you're getting scammed. Please be careful. So the concept is that if you [clears throat] can prove that you still have access to the deposit address plus you have access to the validator assignment key then potentially you can add a withdrawal address and have access to the >> even though you lost the pneummonic. Now this is not a guarantee that this will go in. This is going to be a lobby effort. The idea is to do this one time only. Um so gather the data now so that the the lobbyists know how many people are impacted. Right? So when they go to

ACD with this, they're going to say, "Okay, you're making a big change. You're moving to quantum resistant signatures at some point in the future, three, four years." No, >> I don't know exactly when, you know, um and it could be 10 years, but Vitalik during Defcon was like, "Guys, we have um we have maybe five years to get it done, so let's get it done." Um I hear him there with with the concern about longtail risks. >> Yeah, it's very hard to tell exactly when when or ever we will have enough quantum comput.

>> So, as part of this effort to move all of the wallets over to quantum resistant signatures, we have an opportunity to do a one-time change to say if we need to do this anyway, we're going to allow you um to get your ETH back from a validator that basically is stuck because you lost the pneummonic. Um, if you have the mimmonic, just use the regular BLS to execution change. You don't need to do anything special. You don't need to wait. Just do it now. Become a type one or type two validator. Enjoy your rewards instead of seeing them accumulate on beacon chain, not doing anything with them, right?

>> And that reminds me that like if you want to get started with staking or if you want to do a deposit, you should probably just add a withdrawal address right away. always matter with RO it doesn't make any sense to it is risky not to >> keep using those types of >> it's extremely risky not to right uh because you have to type that minonic at some point into a computer and that's not the case with withdrawal addresses you only type that into your hardware wallet >> and otherwise it stays offline you know never stored online um so yeah the always create as a type one or type two depending on whether you want liquidity or you want compounding Fusaka is coming. Um, and there has been a few uh different uh client releases

Fusaka updates

and >> this is just a reminder that should you should be updating your clients before the fork which is on December 3rd. >> December 3rd soon. >> Yeah. And then um be careful because like some of these clients they might release updates after they publish their initial client version that supports Fusaka uh as uh we just saw with Lighthouse uh which had a second release. So you might want to either wait a little or like just be conscious that there may be updates. >> Yeah, the second lighthouse release is optional for people who already updated Lighthouse. If you haven't updated Lighthouse, just go to that second release. I think it's version 801.

>> Um, ME Boost had had an update from version 110 to 110. They call 1101 mandatory. Um I'm not sure whether you can probably go through um Fusaka with 110, but there was a fairly nasty timing bug fixed in 1101. So if you know you have your MV boost there and you didn't update to 1101, please do. Um Prism will have one more release uh before Fusaka and that one will fix back filling and a couple other things. So, >> so you might want to wait until >> probably not super critical if you're already sunk because you don't need to backfill again. Um, but it's something you'll want uh

>> eventually >> eventually and or right before Fusaka. Read the release notes once that's out. That should be out well within days, >> Hopefully >> because like December 3rd is quite >> December [laughter] 3rd is coming. So, they said it would happen before the heart fork. Uh, fingers crossed. Yes. So >> you were also talk about webser. Uh >> uh yeah I want I just wanted to mention the things you need to pay attention to. You have your consensus layer client like your lighthouse your validator client that might be another lighthouse or it's a vero or whatever maybe you're using web 3 ser right so that needs to be updated then you have your execution layer client and your never mind bas ref whatever and then you have your sidecar for um me boost or commit boost either

one for PBS. So there's like a a whole little zoo of software >> as we get to expand the complexity of >> whole zoo of software you need to pay attention to these tools that >> this is also a good time to uh shill the automation rappers. If you're in systemd then coin cashew has a wonderful little program called ETH pillar which will take over your existing systemd setup and then you just >> automate everything >> automate most things like the configuration is still something you will manually edit. He will not change it for you.

But so if there's a a deprecated command line parameter flag or something, you need to do that yourself. But the client updates are automated. So you can go into ETH pillar and just navigate through the menu and say update everything and it will um huge timesaver. >> It is >> absolutely. And of course users of my very own ETH docker is just two commands. E the update followed by E the up. Uh what what if someone added a custom flag in their um end uh file >> for Eve Docker?

>> Yeah, for Eve Docker. Maybe the >> custom flags are kept. I'm not overwriting anything there. >> You might still want to check out if that custom flag is still being used or what? >> Yeah, if it's still there. Absolutely. If it's been deprecated, so read the release notes for the clients. You know, if you have a custom flag that's been deprecated, your client will not start. But you will see that with SD logs, you know. So check the logs after you update always. if whether you're in systemd or not, you want to make sure that stuff works. >> Uh right, reading these uh fine print is always um helpful. >> Make yourself familiar with what it looks like when it's healthy so you know when it's not.

Staking Gathering

>> Yeah, we had the staking gathering our own event at uh Death Connect. And yes, we did. It was great. >> What did you take out of it? >> Um I really loved it. A lot of um I would say very interesting talks. I was um not bored. I listened to almost everything. It was well organized by our very own Jonathan. It was in the Mala Museum. So, >> which was like such a nice venue for us. >> What a wonderful venue. We got a little bit of a very um rushed [laughter] visit through the museum at the end of the day. Uh took a couple pictures of a couple pictures. Um so, photographs of a few paintings. Um yeah, it was good. I

really had fun. Um >> you had this great on YouTube. Uh yeah, we we have the live uh recording available on YouTube. The videos are not splitted yet. There's not there there's no >> there's going to be timestamps in there. Yeah. >> But uh we'll add the time time stamp and then we'll add potentially like a all the individual talk will be uploaded on our YouTube channel. So you'll be able to uh >> Yeah. Right now it's like a 5 hour marathon. Yeah. Sorry. >> Almost like a 10 hour thing. So you don't need to get there if you don't if you're not interested in everything. Um I loved your interview with Danny Ryan. Denny Ryan is is such a wonderful person. Um, hard on the right space as far as Ethereum goes and very personable, good communicator.

>> What did you like most about the content of that uh discussion? >> Um, how focused he is, I think, on the health of the ecosystem in general and supporting uh home and solo stakers. >> Even even if like he's more into the business side, right? >> He is completely into the business side. he's helping um Wall Street on board, but he's also saying, you know, the the banks don't understand the risks inherent in Ethereum and we can um we can control these risks with decentralization.

Um that's the right way to think about it, you know. >> Uh we also had a bunch of other event at Dev Connect. This is the official last day of Dev Connect. There's still a few things happening here. Um was there any other event that you uh liked or that you had some good insight from? >> So today was ETH proofs day. >> Um this you're going to hear a whole lot more about this over the coming months and years and East staker will be involved um as much as we can be. >> All the Ziki proof will >> all the ZK proving all the lean Ethereum. So the thing I saw this morning was a demo by Justin Drake.

Devconnect

Really blew my mind. um they took a lighthouse, they modified it so it no longer needs an execution layer client and it can validate. So imagine right you're running um like I don't know even like a like a tiny little computer will likely work or it's an an older Mac or something and you can have a 512 gig um drive and it'll be enough. Um, you still need to store the database for Lighthouse, but you no longer have an execution layer.

>> No, never mind. >> Yeah, you have to rely on third party provers, though. >> Yes, you have to rely on third party provers. There are currently seven provers on mainet that deliver real time proofs of every uh block, which is pretty pretty impressive. Um, and this lighthouse subscribes to a special um, peer-to-peer gossip channel to get the proofs. Then it runs a verification of the proof that takes 100 milliseconds, like a tenth of a second, super fast. Um, and then it attests to that. Now, running this now, I should say a couple things about that. Um, that is not rational to do right now.

>> This is very much experimental. >> This is super experimental. It's not ready for prime time. >> Don't go ahead and run this on your own. We we will actually have one user on Discord who's like, "Oh, I love this. I'm going to take one of my many validators and run it that way, which is cool when you have many and you're rich like that, right?" But no, it's not rational to do because um well, one, it's extremely experimental software, and then two, uh you don't get optimal rewards while you don't have to run the execution layer client, which is great. No more disk space anxiety, no more pruning, none of that, right?

um you also because you have to wait for the proof um you're not going to be timely. So right now in in a 12 second slot you need to attest within 4 seconds. Um and the proofs take well let's say 99.7% of the proofs are generated within 12 seconds. That's great already but it's too late to test in a timely manner >> for most uh >> it's too late to test for good behavior. >> Yeah. Exactly. um after Glamsterdam that may look different. Uh Justin is very optimistic and thinks that maybe 10% of the VA data set might opt into this type of a testing because then with EPBS you have more time to attest and so you can just grab the proof for test and be

timely and get great rewards and not have to run. >> So the future pretty much has a bunch of Zik Zik proof in them, right? >> Yeah, it does. It's it's it's leaning into that ZK proving new future. um if it become and also be careful if this becomes where Ethereum goes. This is the the vision of Justin Drake and the team around him for lean Ethereum. However, that is not currently the official road map for Ethereum. It's kind of being built, you know, in parallel >> parallel track. Yeah.

>> As a parallel track. Uh I like to quip it. Turns out you can just do things >> and uh >> Exactly. >> Yeah. Exactly. and and and and broad buyin by governance uh will have to be sought before this becomes a mandatory component of Ethereum and an actual official component of Ethereum. right now it's just a way to run a validator that is not you know sanctioned by cordiffs as such by all cordiffs right >> yeah um myself I was at the trillion dollar event today where we tried to figure out um how we can have a trillion dollar assets on the blockchain and most of that uh discussion was surrounding

the security of like every every component of the whole ecosystem um but I feel like most of the talk were very much well we had a few tracks and I was part of the offchain social track uh but I feel like the main track that was focused on was the wallet aspect of the whole eos e ecosystem and the blind signing issue that we have oh yeah >> right now so um a lot of this discussion was uh done there uh there was no recording uh no photograph there but uh I believe the slides of the summaries will be published so that will be something to uh take a look at later on.

>> Do you have a takeaway there on what they want to do about blind signing? >> Um I believe like I don't I don't know if you heard about this. I believe Ledger has a client signing uh feature but that one is quite expensive as uh in terms of like as being a user you have to pay um you have to pay for it. Uh but there's a bunch of different >> uh I'd say like a free and open solution that are coming out um from especially this initiative uh when they want to try to on board potentially like the next billion user on Ethereum as well. So uh we'll have to have a look at that eventually.

>> So the idea is that they'll give you a human readable and understandable version of what it is you're signing there. >> Uh the goal is to get there but there's a there's like different approach to to get there. you you potentially need to have access to like the source code of the smart contract you're interacting with or you potentially have to have access to good simulation of your transaction. So there's >> okay so this is non-trivial is >> this is very non-trivial way to uh >> this is a hard problem >> this is a hard problem which which is why like there's a bunch of different approach to to okay >> to this um anything else about dev connect that we uh learned today I think there's one interesting thing uh devcon next year is going to be in Mumbai um are you exciting to Mumbai

>> I am not going to Mumbai >> you're not all right >> I'm not going to Mumbai but I'm I'm sure people will have a blast in Mumbai India um and yeah I look forward to hearing what is happening at Defcon when people come back, but I personally will not be going. >> All right. Um, is there anything else in the DevCon space that we should be talking about that we missed? >> Nothing I can think of right now. >> Um, all right. Well, then, uh, if there do you have anything else we we want to say? >> My last thing I want to say is, um, please help us out here. We desperately need topics to talk about [laughter] >> for sure. So if you have something we like can you explain in XYZ or I have a question please put that in the YouTube

Outro

comments or come to East Staker Discord and ask it there and we will put it into our long list of topics we haven't covered yet which is getting shorter that's the problem. >> Y [laughter] and uh then we'll cover it at S&T 6 or 7 or something. And until then thank you. >> Thank you. And make sure not to stake like Yoric. >> Don't stake like Remy.