Casper Community Calls, 19 Jan 2021

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Community call, Introduction to CasperLabs

Description

Welcome to the new format for our weekly community calls which are live every Tuesday at 9am PST

Learn more about CasperLabs today
https://casperlabs.io/company/see-our-vision/

Join our Telegram
https://t.me/casperlabs

Follow us on Twitter
https://twitter.com/meetCasperLabs

A

Hi everyone um I'd like to just quickly briefly introduce myself. My name is joe davis, I'm from london, england, and I am the community director at casper labs. Basically, my job is to manage all of the casper communities.

A

I also post content and announcements from our various social media channels. um With that being said, I'm very excited to welcome you all today uh to the new format for our weekly community calls, starting today and running every tuesday at 9am pst.

A

We will be following a more casual approach to these calls with an ask me anything approach, you'll be able to ask us questions beforehand, and we will do our best to answer them for you. uh We all have. uh We will have weekly guests uh like partners and casper labs team members.

A

um So with that being said, uh don't worry. We will still include our weekly technical updates, as they have been in the past, um so I'll pass that off to meadow and welcome everyone to the to the new community call.

B

Thank you, joe very much for that. um I'm metta parlikar, the cto and one of the co-founders of casper labs and I'm joined today with a few folks in the community and, of course, ashok rendezvous. So without further ado, I'll talk a little bit about my background.

B

I came into technology in the early 80s and started really working as an engineering executive in large software as a service companies back in the early 90s or mid to late 90s with mp3.com divex, adobe and avalera, and I served as an engineering leader where I was responsible for driving teams to ship product right to solve real customer problems and I've been responsible for multi-million dollar.

B

You know implementations and large accounts where uptime was critical, so I come in with a real understanding of what it takes to build professional software, what it takes to serve real enterprise, customers and businesses in a 24, 7, uptime environment. So I'm very excited uh for this new format. Joe thank you for you know, kicking it off and I'm going to turn it over now. uh Next slide to ashok.

C

Yeah thanks maida thanks joe nice meeting everyone. My name is ashok ranadive. I joined casper labs about one and a half years back when meda convinced me that blockchain is the next big thing for mankind.

C

uh Prior to that, I was with google for about 12 years in different parts of google's organization and different locations, fortunate to serve in three different offices of google. I spent five years managing program program manager, teams in pixel software and data center operations.

C

uh Three years in singapore and four years in india running a pro services team prior to google, I was with indian navy, I actually retired, as a commander by education.

C

uh I have done my master's in nuclear engineering from indian institute of technology, kanpur uh iit kanpur and have my undergrad in mechanical engineering, and I also have an mba.

C

I was part of a very prestigious project in indian navy for over a decade I built the first unconventionally powered so to say um submarine for indian navy excited to be here. um Thank you.

B

Yeah, nothing less than a nuclear engineer, helping to lead and guide the building of this blockchain, which is kind of funny, because I actually find there's a lot of folks that are physicists that wind up in blockchain, because the blockchain is so ephemeral right. So those minds are really well uh well designed to work with blockchain technology.

C

Exciting to be here, yeah.

B

Excited to have you here: let's go next slide, so we're going to talk a little bit about the primary barriers to blockchain adoption right. So what is what makes casper really really unique right?

B

And if you look at all the other blockchains that are out there, you'll find that there's a few really big problems in terms of you know blocking adoption right, one, it's hard to get started because you've got proprietary programming, language and the blockchain itself is a different enough machine to work with let alone having to learn a new language on top of it.

B

So this is something that really makes it hard for developers to get started and lack of pricing predictability is a big issue, because if you don't know how much something is going to cost and you can't control your costs and you can't budget for your costs, then you really can't build a business using the technology right and the third thing, that's really a big deal is unclear. Upgrade path, mechanics right! So when you think about it, you know almost all the software that you've ever experienced till date you've had control over right.

B

So every now and then you'll get your iphone update right that pops in and apple pushes out an update or you will. You know if you're, using a software as a service like amazon.com they're, pushing out updates to their website routinely, but they have full control over these things right well on the blockchain, because you're using shared infrastructure when you're running a dap on a blockchain, you don't necessarily know or really have understanding.

B

You really have to manage how that blockchain software gets updated and sometimes it gets upgraded and updated right out from under adapt developers and they're having to struggle with a contract, that's immutable or forks where they can't even do an upgrade. So how do you manage your customers and your business if you can't manage upgrades right? So these are real problems uh with around blockchain adoption.

B

And so you know, when we set out to build the casper blockchain, we really took a hard look at what did we want to do in terms of the core virtual machine? What kind of experience did we want to provide developers and, most importantly, we wanted to build something. That's future-proof right, because once we build that blockchain, we want to make sure that in the future it continues to sustain and evolve over time, and so we went with web assembly and rust provides excellent type safety, which is really really important for financial applications.

B

You can do really great things with rust and it's it's a very secure and you can even do formal formal. You know formal methods using brust, so we felt that that was a really good choice and you know blockchain pricing changes are really inconsistent with industry trends. Right, if you think about um using aws, aws has reduced their prices 47 times over the you know its instantiation since it started so usually prices start off high and they go down as you get economies of scale.

B

Well, the blockchain is actually inverse of that right as a blockchain gets more adoption. Pricing actually goes up for use of the blockchain and that's really not what you want right and we are going to be doing a technical integration with chain link and we're also going to be building a gas futures market that will enable people to basically predict and budget for their future execution. So we'll create a mechanism where block space will be tokenized, and then you can just basically book future slots in the block up to six months in advance.

B

So you can have some guarantees around how much your transaction fees are going to cost and on the upgrade mechanics. We provided contract authors, uh the ability to upgrade their contracts, including complete internal state upgrades, as well as defining which protocol version their contract works with. So in the event of a protocol fork, you can actually control which version of the protocol fork.

B

Your contract is going to go with, and this is really important, both from a governance perspective, but also from a customer support perspective, because you don't want to see you know multiple versions of your contract out there, one that could potentially have really you know bad effects, because an upgrade took did something to the internal into the mechanics of the contract that you don't have control over. So um the is upgradable, no matter what your business brings you'll be able to make those changes and support your customers on casper.

B

So you can learn more about our recent announcements. We've recently published the highway protocol and we've cleared our trail of bits, security, audit and very amazing findings. There. I'm super proud to be able to say in recent days we're one of the few layer, one protocols that has been completely transparent about our audit findings, so you can go to trail of bits on github and see what their security audit findings were and you'll.

B

I'm pleased to report that it was very low or low or informational findings, so we always put security first at casper, we've been very clear that the technology and the security is the most important thing we're working towards and the trailer bits audit really proves that out. We're excited to announce a matisse and casper partnership. Matisse is building uh decentralized uh governance protocols that businesses can use to govern their business right. So the governance problem for corporations is a real problem and blockchain technology can definitely bring a lot of value there.

B

So matisse has signed an mou with us and we're very excited to be migrating them over to the casper.

B

B

Announcing our mainnet launch we're very excited, I'm so thrilled about this- that we'll be launching mainnet in q1 of 2021 of this year. So that means february march by the latest. Our mainnet will go, live we're presently in our delta test net and we've had fantastic engagement. um Over 100 110 validators will join. The network is permissionless, so people can join at any time um just download the node software join us on discord and follow the instructions, and you can bond onto the network, we'll be pulling uh the rewards information off of the protocol.

B

You know the network data itself, so we're really going to be looking at people holding on to their public private key pairs, so they can prove their participation because we're only using the keys as a way to know whether somebody's participated in the network or not so join us on telegram and wechat and, of course, on discord. You can find our discord url our discord, server information on the website, okay, we're gonna go into the ama joe, is that right. I've got some questions here.

B

A

Yes, we have a few questions that came in um and I have sent them over to you.

B

Okay, fantastic, so I believe the first question is uh is coming from unkith on telegram. There are so many projects who support decentralization and there are plenty like hbar, uh soul and pond. I'm not sure what pond is, but soul and pond. When we talk about their partnership and their use case and their supply, we see that hbar is 50 billion, so is 80 million et cetera. So how will casper beat the system and give a good price at the time of launching investors need partnerships they'll, give a push trader needs good supply.

B

How will casper come with its plan? So we haven't been talking a lot about the token plan. Token supply you can definitely for for regulatory and compliance reasons, but you can look, look up the token supply on docs.casperlabs.io.

B

We show what the total supply is. Renault comes from a background of traditional finance he's intimately familiar with you know. What do I want to say how money works right? So we just talk about how do currencies work? How does money work? This is something that renault has been steeped in. uh He has a professional experience with this and you find that a lot of blockchain protocols don't actually have someone that comes from traditional finance markets at the helm, and this, I think, puts us in a very unique position given renault's experience.

B

So what are some of the things that we've done? One? We have a very small token allocation for the core team and advisors, and if you look at a lot of other distributions, there's some there's some links up on twitter, some folks have put out there. You'll often see see team distributions of excess of 25 percent.

B

Our team distribution is only eight percent. The other thing we've done is our developer. Dao grant program uh the developer dow. The devdao is a completely separate, a swiss association, a swiss entity that has all the developer grants for casper.

B

Like a large chunk of tokens, um I think at the launch they're going to have a half a percent of all tokens allocated specifically for developer grants, and the agreement outlines that they are incentivized to give away as many tokens as they can to valid grantees um in an aws credit style of mechanism to help drive adoption.

B

So this is like specifically grants around building utilization of the network.

B

We're doing a lot of really exciting mous and partnerships and ashok is heading up that you know adoption program both through business, business development and strategic partnerships in that we'll have at least half a dozen projects that will be building on casper by the time of launch. So we've got uh ipwe that we're gonna, we'll bring them on here and you'll, get a chance to meet uh the folks at ipwe and we'll talk about their solution in future.

B

Community calls um we've been working on that that uh strategic partnership and implementation for a few months um we've also got uh hera that we'll be bringing on board and we'll be talking about that mou. Coming up, that's a very exciting partnership. That's coming and we'll also, you know, all the tokens at launch will be locked up for regulatory compliance reasons, and so those persons that will have freedom freely trading tokens will have a pretty long window of time where they can experience some token appreciation. We hope. But again no promises can be made.

B

This is not a security and, of course, u.s investors will be excluded and nothing, I'm saying here constitutes investment advice. um So when we talk about the current scenarios, many projects do partnerships before it's launching how many partnerships casper lab has done so far. So we talked about metis that's been signed, we have the ipv that signed we're working with broadly for enterprise deals, and we have at least uh half a dozen par such partnerships planned or that we're in final talks with.

C

We also have plasma pain.

B

Yeah, quite a few interesting ones. um How will you come with some country's law barrier? I believe this question means it's: how will we overcome some country's law barrier? Well, the network is decentralized, so our goal is that the security and permissionless properties of the casper protocol will eventually prove out over time such that it will be looked at as one of the top leading decentralized, truly decentralized protocols, both from a governance perspective as well. As you know, a code maintenance perspective.

B

So if you understand rust, if you know how to use rust, we're looking for community contributors to work with the protocol and to become core developers, we have 15 percent of tokens that we've given to the dev dow join the dev dow, and you can earn tokens through grants right.

B

um This is the nature of decentralized uh development for blockchains, and our intention is that once casper launches that it will become a decentralized protocol, and that is the way the real way that it's going to get adopted in countries that have legal barriers, the u.s not being the least of them.

B

Okay, let's see those are three questions. Any more questions. There's some q, as here right.

B

So what is in the cbc casper specification, and what does it make? What makes it special um everett if I'm going off of the script here, just feel free to send me a send me a ping here and I'll do I'll. Do my very best to stick to the script, and so it's it's a good meaty upgrade. You know update for the community, so the cbc casper specification talks about not only casper as a protocol right the base safety protocol. um That's the 2017 paper that vlad put out.

B

That's the paper, that's being used in gasper and casper ffg, which is part of ethereum 2.0, which we could protect we'll talk later in another in another call more about. You know how we are not like ethereum how we are like ethereum. We share that in in common with the ethereum casper ffg. Is this notion of a fork choice? Rule right? um That's that's the casper! um You know cbc specification.

B

The cbc portion is what we call correct by construction, and you saw earlier. I talked about the highway paper and the trail of bits on it. These are the portions that really talk about cbc, which is correct by construction.

B

So we did that in the same vein as cbc is we started with formal mathematical proofs, and then we constructed code that met that formal mathematical proof, and then we had the every you know all of it audited by a trail of bits right so trail of bits completed an audit where they checked the paper and the formalism in the paper, and they verified that it was correct, and then they took a look at our code.

B

They looked at our code and they verified that our code matched the mathematical paper and initially they had some findings that you know our lnc. You know piece that was specified in the math paper wasn't implemented, so we were actually able to confirm that they actually did check everything correctly, and so this is the correct by construction, formalism, right and so andreas fackler, who was is one of the key authors on the paper along with daniel kane and cardinal cryptography.

B

Andreas is the one that wrote the bulk of the highway protocol right. So he actually wrote all the rus code, that is in the protocol. He actually also participated in the paper and he was responsible for translating the mathematics into rus code and then, of course, he worked with trailer bits to ensure that everything was audited.

B

Any other questions in coming in now from the members on the line.

B

I'm happy to take questions if you guys want to hit me with any questions right now, you're free to include them inside the q, a.

B

B

I can proceed with a tech update if there's no questions, how does that sound to everybody? Joe sounds good tech update.

A

B

All right, let's do that all right!

B

Let me share my screen again.

B

All right, can you guys see my screen? Okay,.

A

Sure yeah yeah.

B

All right so we've started our 21.02 release cycle and we are working on protocol upgrades performance and hardening improvements, and we are also planning on completing our integration with our custody provider. During this release cycle and we cut tag v0.63, we are working through a few issues. uh The delta test set is online with a block height of 3404.

B

The network is waiting for a bunch of validators to catch up and sync, so we found a synchronization issue. We dropped the brand new lib p2p small network component, so we replaced the original pre-production or pre yeah production ready small network component with lid p2p, which is what is used by ethereum and a variety of other systems.

B

The lib p2p component didn't have a lot of documentation or unit tests which made the integration a little bit of a challenge and now we're seeing that the joining is really really slow. um So we strongly suspect that there's an issue in the p2p implementation, because we haven't had any issues with joining for quite some time. In fact not since, like september of october, and so this sudden regression um leads us to believe that there is an issue um inside the lib p2p component. And so what we've see.

B

What we're seeing right now is that joining is very, very slow um and the network is waiting for some of the bonded validators to catch.

B

Up and our current focus is we're going to evict the non-participating validators. So if a validator hits a liveness failure, they will be evicted and uh ashok remind me: will they be evicted for one era, the current era or or they're just evicted completely out of the auction contract?

B

And then they have to wait for, like uh unbonding delay to rejoin.

C

Yes, they are evicted for um completely evicted, yeah they're.

B

Completely affected yeah, but their their tokens are not on, will not be unbonded, so we will have a 90-day lock on all bonded tokens. So there's not going to be a loophole around that right, so they'll just be evicted, but not unbonded.

C

Not unbonded, yes, that's right, correct.

B

um And so we're doing some, there are some ceps, so those of you in the community, if you're running a node and if you're a validator, please do take a look at the ceps. That is where our enhancement proposals are being dropped.

B

So these are the requests for changes, um and so you can see the ceps inside the pull request. So there's one ledgerly submitted one for validator key management. This is crash recovery, um designated node operators, validated a lock up delegation caps, disabling bids on liveness failures. um This is the gas spot market, which is what we're going to going live with, but not the futures gas futures here so there's quite a few important ceps that need to be reviewed by validators. So we highly highly encourage validators to come in and check these.

B

uh These requests for changes. We want to make sure that we're as transparent in the community, so you can go ahead and take a look here and then you can view um you can view the uh the cep right in here right. So this is where you'll be able to see it.

B

Okay, and so that's what the review crash recovery means is. We need to review the cep, we're going to replace gossiper with the lid p2p gossiping and then we're implementing a public key in account. So this is we're removing the account hashes. We discovered that not having not using the hashes in the account in the account semantics in the global state was actually preferred because then what happens is you have this account mismatch right.

B

So, if you're, using a public key and you're trans you're you're transferring money from one public key address to another, there's an expectation that you're going to see that same public key in the global state and seeing an account hash really makes it hard to audit. So we just got decided that we were going to roll back that account hash change. It's.

B

We will still support multiple key types, so we still support ethereum keys, as well as the casper keys, um so still supporting p256 k1 keys and ed25519 keys, and our implementation actually allows us to support as many different encryption schemes as we want in the future and those would probably go out. I think they would probably go out as hard forks protocol hard forks we're gonna pro. uh What's a prototype joiner, so this is the fast synchronization. Do you know what the prototype joiners?

B

I don't know what the prototype joiner is.

C

Not not at the top of my head.

B

Okay, I'll, I owe you guys an update on what the prototype joiner is and then an optional state route hash in genesis. So what this means is if we have an equivocation catastrophe and we need to do social consensus, that means we will restart genesis from another root hash. Besides the genesis root hash right. So when you start from genesis, the genesis root.

B

Hash is implicit, but if you have to recover from an equivocation catastrophe- and this is similar like a 51- you know- 51 attack 51 attack in proof of work protocols- they basically roll back the chain to the last good block and restart.

B

This is basically the same thing, so you would basically pick the last good state route hash and you would put that in the chain specification and then you would boot the network from that state root hash and go forward on test and sre we're going to start implementing some kubernetes deployment for node testing. This is to enable uh testing of large networks on in a local environment, and this is the uh elastic kubernetes clusters uh for ci and local testing and then we're going to create a monitoring stack deployment strategy.

B

So this is really for being able to monitor the network more broadly on the ecosystem front, we're doing a lot of work for tutorials, we're focusing on the multi-sig and key management tutorial.

B

So I did create a tutorial early on there's a youtube video that basic functionality hasn't changed, but the contracts have updated since we did the contract headers work, we're building a dsl for upgradable contracts and our javascript sdk is available on npm and we're adding a query method, and this is because we actually built this js sdk, because our custody provider requested us to do so.

B

um Alexander is working on the econ paper, basically integrating that with the consensus paper and we're going to be adding the staking auctions chapter to the econ paper. So the idea is to show a paper with the economics of the casper blockchain. um What is your date for that alex? I really want to have that done by you know the first of february, mid february, at the latest.

D

No, I mean it will be done by that. I mean we need to. uh I mean a lot of this is feeding through the consensus paper, because we need to make sure that the reward actually makes sense. Given the changes in the uh in the consensus paper and, of course, the reward scheme was initially devised by owner, uh and you know so for for me, it's somewhat uh unexplored territories, it's taking a while to actually do that.

B

Okay, are you looking at owner's math he he did. He did have some mathematics around that.

D

Yeah, that's exactly what I'm looking at yeah.

B

Okay, yep. Well, we certainly don't want to redesign it. um The the idea is to eventually like I'm, working on the balances, the starting balances of the genesis block and I'm actually looking at the distribution of tokens in the genesis block as well. So we are working to reach out to those persons that participated in the vfta and finding out whether they intend to run a node or not. Our intention is to have 200 validator slots at the launch of mainnet, and we've been pushing really hard to get up to 200.

B

You know we're looking for 200 nodes right in the in the test net. um I think that the most we've had is 110 nodes. um So we definitely want to make sure that this thing scales out to 200 plus nodes before we launch on the contract runtime we're going to be making a small change where we're going to support an immutable contract. So this contract will never upgrade.

B

We are ripping out the system contracts. This enables folks to build from source and run from source with just a chain. Spec right, there's system contracts in there that are vestigial. We don't need them and we are going to uh we're going to eliminate them. We really want people to be able to run from code directly and implementation, we're going to evict the non-participating validators. So if a validator has a significant liveness fault and they've not unbonded, they will be evicted from the auction contract, and this does not mean that they'll be unbonded.

B

From my perspective, they will still need to send a specific unbond request, wait for the unbonding wait period and then they can rejoin.

B

And on the governance status, we're also working, you know, in addition to the cep process, we're also setting up, there will be a casper association and the validators will have um basically stake in the association, so they will be full-fledged members of the association and that governance process is being flushed out as we speak. So um there will be a casper association that will be separate from casper labs and that bifurcation will happen. Thereabouts. The time of mainnet launch.

B

Any other questions, joe any questions coming in from telegram.

A

Not at the moment.

B

Okay, great so in future, community updates expect to meet some of the core engineering team. I'm gonna make them get on here and put on their cameras. um Today we had originally tapped ed to join us, but unfortunately he is in a series of meetings, so I'm gonna have to give him some heads up and advance notice as to when he can join. So you folks will get to meet ed hastings, who is a close friend of mine and one of my former colleagues from avalara and he'll be joining another military man.

B

If you don't see a trend, um he'll be joining us in a future call. Hopefully the next call I'll try to get him on here and then, of course, you'll get the chance to meet some of the other core members of the team. So thanks everyone for dialing in love, to see the engagement and uh joe feel free to share the the link, and let's get more folks on the call itself.

A

So, thank you. Everyone who watched live thanks to those that watched on youtube after the live event. uh Please feel free to subscribe to us on youtube with the bell notification on, so you never miss when we go live and we'll see you next week. My dms are open, so send me a dm with any questions you want us to cover in next week's community. Call thanks. Everyone.

B

C

Thanks everyone bye.