Defining Active Users in Crypto: Solana vs Ethereum Metrics

[00:00:00]all right everyone this episode is brought to you by manad an L1 bringing performance to the evm with parallel execution and both a custom consensus engine and new database solution you'll hear more about them later in the show before we get moving on today's episode just a quick disclaimer the views expressed on this podcast by either myself my co-host or any guests are their personal views and do not represent the views of any Associated organization nothing in the episode should be construed or relied upon as Financial technical tax legal or any other advice okay let's jump into the [Music] episode all righty everyone welcome back to another episode of light speed it's we got a Roundup episode with mert and I today and uh over the C last few episodes you may have noticed a a a new co-host here quick intro my name is Dan Smith I'm leading the analytics product on our uh on on our team over at blockworks research and um I have been asked to fill the big shoes leaving that game Garrett has left he has uh done a great job creating and and hosting this podcast alongside mert um so I promise I will deliver that Excellence that Garrett brought to you um but we'll jump straight into things here M there's a lot to talk about uh going on in the salon ecosystem but I think one of the notable things to jam on is Coliseum or this salana hackathon can you give us a quick intro on what exactly the Coliseum is um and why I guess the motives behind why it's an important to see for the growth of the salana ecosystem sure yeah so salana is known for having pretty large successful hackathons that have led to the formation of many ecosystem teams today so teams like tensor jeto mango dialect and Crossman some few other Venture funded companies that are actually quite large today came from salana hackathons and so salana has about I think two or three large hackathons per year and they're quite large they get very impressive sponsors uh very solid branding messaging distribution so they're a whole thing and I actually like to think of them as the best ions in all of Tech um personally and I know that sounds biased but that's actually partly how I found out about Sal in the first place like the hackathons are actually that good and so Coliseum says like okay there's clearly a lot of talent here and it's kind of a repeatable process how do we now turn this into uh maybe more of a more of a like a YC type deal where you take the winners or people who Place High in the hackathons and then you incubate them and you give them advice Mentor networks uh intro uh intros lessons and you know other things that you might learn if you're building a company you give them all those resources and then you give them a better shot of becoming startups and and and and and companies that actually solve real problems in crypto uh um and then I believe it's so colise is actually a fund and they also take some sort of they they fund you as well and get some sort of equity in return uh so pretty much as similar as like Aon Stow or or YC and it's headed up by Maddie who is the ex- head of growth at salana Foundation uh so he really knows salana and um he's seen a lot of uh hackathons and he's actually organized I think pretty much all of them so far as well and so that's that's what it is and I think it's super exciting because you know it's it's hard to do hackathons for like other blockchains because generally you have to focus on infrastructure and scale there right like if you go to like an eth conference a lot of people are talking about scalability and like l2s and bridges and and da and all this stuff whereas on salana if you've been to one of those conferences sure people will talk about that somewhat but the relative frequency is much more skewed towards products and consumer facing stuff and when you have that sort of environment it's exciting because we're increasing the probability that there's more consumer facing products that are actually useful as opposed to the necessary infrastructure to scale right where that's kind of the biggest benefit of salana which is you don't have to focus on infrastructure as much obviously I do but most people don't and so um super excited for that and basically uh another part of it is that um they they're super solid for leading to like VC rounds right like a lot of VCS are Judges and so if you have some sort of an idea or a product or like a good presentation you get a lot of chats with VCS and you really increase your chances of being able to raise raise money and kind of start this whole company you know Journey um so that's what it is and then I think I believe today is the 27th they have a big announcement um not too sure where it is I think they're just announcing the criteria maybe the tracks um but yeah so I'm I'm pretty excited for that I think um the the shortage of applications in crypto by far been my biggest problem with the space and something I talk about frequently and I actually talked about it on the like the media coverage for our series a around for helus uh explicitly like it just really bugs me how the focus is so much on infrastructure and not on applications yeah congrats by the way that's a huge win and it's great to see an awesome team getting getting the additional funding they need to keep moving um so so again congrats on that um you mentioned a lot of cool things there so the focus on application is That explicit through Coliseum or like if you had you know a great infrastructure idea could you still go through the Coliseum application process and and it's just like do you just mean it more generally the teams building on salana are are more application focused because the the base layer itself kind of lends to being able to build those things there yeah like I'm I'm pretty sure there will be an infrastructure track because like salana is more scalable but it's not like the infrastructure work is done or anything right there's still a lot of work there um but so there'll definitely be infrastructure teams and infrastructure tracks so for example jeto right came around and salana obviously has no mempool and they literally Built Me infrastructure for a Blain with no mempool that's pretty wild um so that's a good example of infrastructure um and so yeah there there'll definitely be infrastructure um but like you have more room and freedom to actually experiment with products and applications because like there's also a lot of enshrined functionality onto the L1 like there's token extensions where you can program tokens to be pretty flexible like you can be conf they can be confidential token transfers uh interest bearing yield bearing Etc um Frozen soulbound Etc all these kind of interesting functionalities where you'd have to maybe write a contract for on ethereum they're kind of already built in to the Lan on salana and so that's just slightly less time that you have to focus or or or or and like you obviously don't have to audit it because it's audited centrally du you have Sal's program structure works versus some of these other chains right you don't have like a bunch of smart contracts implementing the same interface you have the same program which obviously has its downsides if it's if it becomes centralized or there's a catastrophic bug but the upside is that when it does work it does save a lot of time yeah no that that that makes sense and one of the things that I you know when you look out at the landscape today there's a lot of l1s that do a lot of similar things with different twists to them right and when you're you know as an analyst you're kind of tasked with judging the staying power of some of these things and the in the long term I think things get a little hairier but in the short to medium term I think one of the most important um factors is the developer Community not only around the application layer but also the base layer so on the Bas layer and you look at salana today you know you have multiple teams building clients now between Labs jto the team that's spun out of labs I think it's Anza if I have that right and of course fire dancer and the jump team then you have application developers like Eugene Chen and his his team at Ellipsis thinking about uh you know the pho Fe the the the evolution of the salon fee mechanism with that recent paper from umra Umbra research um and that is a very important quality right it's not it's no longer anatol and the core team like pushing this thing forward from all different angles it's multiple teams that have you know their own Mo motives to improve the base infrastructure um so that's really really important to see uh at the kind of that core level but also at the the application layer right if you have multiple teams that care about the future of the chain uh that really gives you you that staying power because those teams are going to invest their resources uh into into Building Products that people use and I I really agree with you that you know on your your just core idea that we need more applications in this space right like we see a lot of this right A lot of these cheap da layers are making building rollups easier right but we we just keep creating more block space for for who for the same users and I just don't see how that's solving a lot of the core problems that uh we kind of are facing right now okay guys quick break from the episode to talk to you about onad the L1 optimizing the performance of the evm team is working to materially advance the efficient Frontier in the trade-off between decentralization and scalability the internal devet is currently alive and public test net is coming soon and testing on the internal devnet indicates that the chain can handle up to about 10,000 transactions per second significantly increasing the throughput capabilities of the evm this of course opens doors for new applications and more interesting use cases even those with greater complexity and higher usage to run in a decentralized manner important importantly manad is fully compatible with the evm and the ethereum RPC API which provides evm developers with the seamless portability for their applications given the popularity of the evm today this is really a no-brainer to stay up to date with all of the latest developments join the manad community by following them on Twitter and jumping in the Discord they're a lively bunch so hit the links in the description below all right let's get back to the episode yeah uh two things there um so the first point is um one of the in differences between l1's and l2s is like people talk about fragmentation of like liquidity but I think like the much more interesting part is actually the fragmentation of like social collaboration right like on an L1 let's say on an integrated one where a lot of different activity types are are occurring um if something is wrong with the chain all those people are working together to help fix that problem right so like for example Eugene and and Ellipsis they're helping with the fee mechanism because it affects them directly um and there's other teams like infrastructure teams but also nft teams and standards people who want to help because they're all building on the same shared infrastructure which can can be bad obviously in the cases where that infrastructure is clogged up but if it's scalable then then you're in this case where all these people are collaborating on fixing these problems in a unified way whereas maybe in the other approach extreme approach where it's like the full modul modular thing they're not they're not really they don't have the same incentives to work on that shared infrastructure because it's like an external dependency in a sense right like if I'm on arbitrum um or not arbitrum but like let's say a rollup that uses can't decide between like Celestia or ethereum for da or something I don't have the same it's like well I can just actually switch these things around I don't really I'm not I don't have skin in the game as much right which is good for me sometimes but you know there's trade-offs in in in in both directions there so I think that's like one of the interesting things about having a a a community Comm uh uh uh of an integrated ALB one with shared activity because people have skin in the game to work together and your point about like how it used to be an to and then now it's decentralizing more is absolutely correct I set up like a research Channel where now we have people from like ethereum Cosmos fire dancer all these different client teams like collaborating on talking about all these different problems whereas before it used to be just like people just YOLO [ __ ] into PR like two three years ago um so that's cool and then uh this the second uh point is the D5 point right so I tweeted something similar um in I was actually inspired by Toronto uh which which is a you know Toronto is not a very inspiring Place uh but but uh the the economy of uh uh Toronto feels fake because it's it's just real estate and Banking and like they're just trading things between themselves but there's not really anything being produced and that kind of reminded me of defi uh and this is not a knock on defi by the way this is just this is actually a knock on crypto um where for defi to actually work you need to produce Goods that are actually have some sort of value in some regard and you trade those around and you you know uh add some financial instruments to you know borrow land Etc and then that's how you make markets more efficient but if you don't have those goods being produced like real world assets or whatever then you're kind of just trading nothing amongst yourselves and you're able to aage just farming points because there's nothing actually uh worth trading like uh it's you know I mean at least with a house I can live in it but in defi what am I going to do I just have these tokens that are just tokens for usually governance but governance or what who knows that's where the doubt decide right it's just so uh so I think like the the point that you made is actually quite correct where you need to produce things of value for defi to really work and I think salana has a good uh starting point here because of deepen right like uh like helium and high mapper give you some actual like tokens that at least somewhat tie to real world value um and S or and then there's obviously more of an rwa push these days as well with like parcel um and there's like baxes which is like just whiskey like it's it's just once you have those kind of primitiv start building up then I think def5 starts to make a lot more sense um versus on other chains just pay you money to use dii um so this like a weird chicken neeg problem but yeah so tldr is like the the the the hackathon Coliseum if it does produce these real consumer applications that have real value then defi really benefits yeah there there's no doubt about it and one of those consumer apps that's pretty interesting is drip and you guys had drip on the Pod about six months ago I think it was and great episode we'll put the link in the show notes to that one but if you can just give us a quick one two on on what exactly drip is doing and then we can jump into uh some of the like their user metrics numbers because they do some interesting things on that side yeah so people get confused about what drip is like pretty frequently but I mean it's actually really simple it's basically the same thing as YouTube or Instagram just maybe more crypto which is to say you go on it and there's creators right and you subscribe to them and then those creators give you content and that content is tokenized so basically subscribe to the creators and then creators give you content in the form of nfts and in salon's case it's compress nfts because that's much more scalable in terms of cost so like uh there's like a there's like an artist on salana called djan poet it's this guy who literally makes art with typewriters right it's actually my pfy on Twitter and you can subscribe to him and then he drops you free art um I don't know what the Cadence is but like pretty regularly and then there's like other gamification mechanisms here where like if you really like this person's art you can like like their stuff um and it's like a little bit financialized with these things called droplets and um if you have like droplets you might get like a more rare uh like exclusive drop from that person um and then you know obviously if you're the type of person who collects collecting you collect if you just want to consume the content maybe it's like a comic book then you just consume or you actually want to sell it people do that as well and so um uh uh you you can do three those things and then Vib obviously is a pretty established founder he worked with like Tony Fidel right uh one of the early people in Nest um he founded beta the kind of the IRL kind of uh trying out Product Store in SF that was actually quite big um he did Start Salon spaces as well and now he's doing this so it's like a super high caliber team and I think the idea was like quite unintuitive at first because everybody body was so used to The nft Meta of like okay here's 10,000 nfts this is all you get it's probably a animated ape and you can sell it for money if the influencers talk about enough whereas now it's like well actually this is just YouTube and Tik Tok except the content is tokenized and the creators can now make more money and if the creators make more money chances are they're going to produce better content um and and be have more skin in the game for that platform but also they can't be deplatform right because it's crypto you can just kind of move to if Twitter's like you know I I don't like how M tweets I'm going to I'm I'm going to ban him he he he talked badly about gemini or something they you can't do that on crypto right uh it's it's on the same shared Leisure and anyway so that's um that's a high level idea I guess let's zoom out for a second out of drip before we jump back in getting the active users of any protocol on on a blockchain is a very challenging thing to do because the bar to spam is quite low right spinning up an address is like a CLI command that takes 5 Seconds right so it's very very challenging to determine what a user how to define a user first of all and then how to count them right because you can count very simply uh using something like Dune the you know the amount of addresses that have interacted with a certain contract it's very easy to do but that number is inflated like guaranteed no matter what across every blockchain um the active addresses is not your number of human users we're also getting into a world where are human users really the only real users like where we don't have ai agents that are functional today we have ideas of AI agents but when we do is in you could imagine a world where there's a protocol built exclusively for AI agents and so we're getting into a very gray area of how to even Define what we're trying to measure uh so I kind of think it's an important thing to preface before we go into this conversation because you know drip does do some very interesting filtering on their total active addresses to get to a number that they feel confident calling a user can you kind of walk us through what some of that filtering looks like in the case of drip oh okay maybe maybe to set the context for why we having this conversation is because um uh I I tweeted about so I saw vibs charts and it said they have something around like 100k daily active um which obviously like that number is not perfect right no number is uh but you can see kind of the trend of the number of users increasing over time and people then started because I said like it's I'm like pretty uh let's say loud about not misleading retail and and I've also said multiple times that daus are totally useless stat on salana because they are uh on salana you can just create Accounts at at uh at infinum right like you can just create millions and since the transactions don't cost you anything you can just make those accounts look active whereas on ethereum it's not the same because obviously the transactions aren't so cheap and the friction is higher and so I always tell people to not use that account uh metric right and so people then said like uh they they now contrasted me talking about drip to me talking about daus in general which is just you know I don't want to say brain damaged but uh C certainly certainly quite uh uh uh uh dishonest because a blockchain is obviously not the same thing as an application okay and uh not sure why I need to say this but like drip for example the way to count as an active address there is you need to be invited with an invite code from another person right you need to uh you need to sign in you need to use the app you need to get past Cloud flares ml stuff like Turn Style you need to have certain number of page views and your user agent uh because like the metadata on the on the on the site will can generally detect on the internet whe whether you're bought or not and they feed that into their ml algorithms antibot algorithms so at the very least obviously that's still not perfect but at the very least it's quite different than a blockchain okay it's it's an app with business logic that is determined to to make a best effort because this obviously affects them right like when you go to raise as a company like this VCS will ask like Hey how do I know these users are real and so you need to show the VCS here's what we use for antibot mechanisms and this is why it's correct Etc right so it's like this affects Vib more than anybody right like not some Rand people on Twitter this actually affects him a lot so he probably thinks about this more than anybody uh obviously and so anyways so that that's my rant about how it's not even remotely comparable thing um but uh so yeah but it it seems like uh they do have quite a few users even with all of that obviously again the 100,000 number it's it's not the actual number of distinct users right there's it's some fraction of that obviously uh but in in Vib actually if people are interested he did release a video on his Twitter about exactly how they run this stuff like the process of antibot like he actually went through all the charts and the mechanism and process they use and so you know there's there's a lot to be said for just somebody who's like that dedicated to even solving this problem right uh and again I just want to and like you know let's say uh the the worst thing that you can do right is is you can go on drip or because you're misled right and and sign up and get free nfts okay that's that's it okay like whereas on a blockchain you buy the token and the token goes down you lose money like on drip this is not the case like the worst case is you receive free nfts so anyway yeah know so on on the on the first piece I want to jump in as well and I strongly agree with you which this shouldn't be a hot take but uh counting the number of active addresses on a blockchain is a it's it's like yeah it's a silly metric because it's very manipulatable so on blockworks Research as Analytics tool we we track all this stuff they're kind of they're they're you can't take any one metric in isolation but youve put them all together you can create a trend out of that and it's very hard to compare something like salana daily active addresses to say ethereum as you mentioned because there's transaction cost differences um and even if you try to compare that to say like Bitcoin like that the utx O model and an account-based Ledger has very different ways you would count and what an active address is or an active user right so you can't even like cross- compare these it's mostly to say okay if I look at you know where this chain was three months ago and where these metrics are today can I'm you know am I am I identifying some Trend I I actually was looking at some of the the active address data for the polygon uh the mtic side chain and it was interesting because I found a trend where essentially there was like a start wallet that had you know some 2,000 madic balance and it would just send it from one address to a fresh wallet send it from the next address to the next wallet and this was like basically every other block it was doing this and what that's doing I don't know why it's doing this I have I have you know you can't really tra I didn't trace it all the way back to the this on the this is on the POS chain yeah the POS chain and uh it was just one after the other sending the full balance leaving nothing behind and sending it to the to the next address and just continually creating this chain and so was creating like 20,000 addresses a day basically and that's like a prime example of why this is a very silly metric um and you know no clear motive to this but it's a relatively cheap to send a transaction on uh polygon's POS chain especially that's just a transfer of the base token um so it's like very cheap to do this but the and again I don't know why it's doing this but the outcome that is being achieved is address spoofing like it is inflating the number of active addresses on the transaction on the Chain so that is why it's a cetric and then zooming back into drip here drip probably has the best user filtering uh methodology in crypto by any application today if there's someone doing it better I would love to talk to them I have not met them yet had the privilege to meet them yet um because it's a very challenging thing to do and so when drip is saying you know 100K daily active users I do take that very seriously it's like to your point yeah it's probably not perfectly 100 it is likely a fraction but they've done some extreme filtering and it's not just like a you know like ah this like someone with one more than one transaction is a daily active user some silly metric like that it's a it's a rigorous process that has multiple checks and um there's another thing about like collecting the the droplets uh for drip like you can't just like buy those you have to actually be using the platform like they have ways to kind of do this and I guess the issue with why it's so hard to or why user metrics are quite inflated often times especially on the application Level L the so the blockchain level is you know airdrop farming is this huge meta right now everybody thinks that they're going to get a free tokens from someone um so it's like oh I just go use this you know this this blockchain or this application and I get free money um and it's it's true like a lot of applications are doing this and so that's why we are facing this wave of like how the hell do we filter out who a user is it's good that you mentioned the the metrics and the trends uh uh concept because that kind of gives me two ideas that I want to slightly touch on uh one is like uh like the the account based model versus the utxo model right that leads to some interesting things inherently but but account based uh uh blockchain is actually also differ right like so for example you have something like a s orsus something like a salana whereas onu the transactions um are encouraged to be as decoupled as possible because that's how object the object model works and that leads to many uh a different CPS like what what what is a transaction right transaction per second but what that t is it's super different on salana where it can be like um for example if you do something on Jupiter it might be like seven eight different routes that it takes through different dexes whereas onu that's actually all different transactions right and then people will say something like well you know it has much more through but it's like well I don't know if that's the case uh it's kind of just packaged up a little differently uh and so it's super important to be careful about that um you know uh again like it's it's not I I think the best way to use these metrics is to compare it to like the chain itself like salana is at like 1,000 TPS let's say and then it's now at 2,000 over one year like okay it's clearly improved upon itself um obviously uh in the concept of Twitter and I'm guilty of doing this you'll want to compare like you need to have some sort of things to comp compare um but those comparisons aren't perfect and that brings me my second comparison is um uh I I have I have a good one here um somebody uh compared um uh oh okay here we go so assuming you believe monolithic chains have a bright future ahead and that liveness is not a critical criteria say is faster and cheaper than Soul it is compatible with evm and it's only a fraction of Soul's market cap why would anyone want to buy soul instead of say bridge to say that's my answer to that well let me rephrase that try to bridge to say and then I I think I replied like assume you believe in modular chains that have a bright future ahead and that life is not a critical criteria Tia is faster than eth why would any why e inste up T right and and like anyways like I'm obviously trolling but like the the like you when you're comparing These Chains right like you're saying say is faster and cheaper than soul I mean I mean first of all it's not cheaper than soul that's just that's just like objectively false I don't even know where that comes from uh but the faster aspect is interesting because it it runs tender right so uh there to say say has like 39 validators and I was actually digging a bit deeper they're almost all in Europe so they're like pretty collocated like they're very near each other um and even then the TPS which is actually a measure of demand in most cases uh these l1s is still not higher than soul and then obviously when you have 39 nodes versus 3,200 nodes your final time is going to be slightly better uh and so that is to say you can only compare these things like say in Salon let's say if Salon had 39 nodes and they were collocated in Europe right I'm I'm pretty sure that if that were the case Salon would be like the fastest shaine by a few orders of magnitude right like 39 collocated uh uh servers uh uh in fact I was talking to told about this a few months ago but uh like that would be like if somebody Forks fire answer when it comes out and does that that would be and and somebody will probably do that that would be like the most OP chain of all time um anyway so that is to say uh uh when comparing metrics it's it's probably better to just compare it to where the chain was before instead of where other chains are today because you have to normalize for so many different things that and like they're totally different models usually right like maybe you can compare like arbitrum to optimism because they're both orus and you know there's some sort of similarities but even in that case like optimism doesn't have proofs uh there's a lot of differences there um so it's it's just very very difficult to compare these things properly so really gota when you read these things on Twitter you gotta be like wait a minute like what is actually being compared here uh and just practice some statistical uh uh skepticism let's say great points and it's like the human it's in human nature to want to compare these things right like that's ultimately what a lot of people are trying to do they're trying to be like should I buy this token or that token well let me see which blockchain has more activity whatever that word means to you on it and like I get the the thought process there um but you're right it's just super super hard to compare these metrics and I kind of want to take us down a tangent here because it's something I've been thinking about and you just brought up the validator account so I thinku is really really interesting Tech I uh I'm pretty excited about what that team is doing one of the things I I think about there is the tra off that they're making is their consensus mechanism um like you send transactions to every uh validator in the network and so you can't just like scale that from I think it was 10 sixish today uh you can't just like add a zero on that without having throughput or excuse me like latency issues or or communication overhead um and so tender's kind of actually the same way it's it's a pretty noisy uh Gossip protocol and they I think the largest tender mid chain today is roughly 150 and they like that's kind of the the the ceiling right now you could do some work to improve on that which the the the teams are very aware of but it's like do we need to like is 150 enough and I'm curious to get your thoughts on that because you know obviously ethereum's at the other side of the spectrum where it's like you got to run this thing on a Raspberry Pie baby uh and then salana is like honestly somewhere I think people think of salana as being like way to the the opposite side of ethereum but like there's 3,000 nodes like that's it's not a small number yeah yeah we we uh I we we had anol on to talk about this because it's certainly interesting like when people say like things scale with Hardware they think that like it just scales when you add more course to the machines or like add higher Ram which is like one dimension of it but the other dimension of it is as you keep adding more machines right like most people who aren't technical don't get this but it's much much difficult to achieve high performance the more machines you have right because those machines have different variants they're in different locations the the information has to go through all of them like the information has to be synced right and so salana though due to and by the way this is actually due to like the vote transactions that people say are just inflating uh but they're actually there for a very good reason and so if you can create a system where adding more machines doesn't slow you down and can actually help speed you up in certain cases that's crazy right and that's kind of what salana does it's not perfect but that is I think and so like we we had totally on a few months ago and I asked him like okay what do you think about these chains with the with the with the low value accounts and his answer is something like okay well like you know if I'm some banking executive in in Silicon Valley or something I want some guarantees or some high level heuristics where I can say like okay this is decentralized enough this is secure enough now if there's 39 nodes I'm sorry there's just no way anybody can actually say that's good enough security characteristic right like the Swift network has more people uh different banks than 39 nodes and that's already like come on and so you you need to have at least because like you're using a blockchain blockchains take more uh uh you have a lot of different assumptions you have to make when you're using a blockchain and so is 39 nodes even worth it right like I I don't think so um and then like 150 again I I don't think that's enough personally because like you it's it's all about guarant and kind of these confidence intervals and like uh one of n assumptions right like the the the greater the number of n the the the the better your chances are that when something does go wrong somebody will detect it right and that's actually vitalic I was looking at some some of his Alex old tweets and he really cares about like he's like uh he says something like pretty funny but he's like he's gu so detest this notion of mountain men that like um there will not be even a single honest person out of like thousands of people to like flag like a faulty State transition or something and that's also what toy believes so like so a lot ethereum actually are quite uh similar in that regard where we do both believe that the number of full nodes or the number of nodes that are running on network is super important now ethereum um says like okay well to increase that number you need to make it as accessible humanly possible uh and I think that is obviously an important factor and it's one half the equation for me but I think the much more important part is the incentive to want to run the node itself right so you can't just have Supply without demand you must have supply and demand right so for example if I run ethereum node or whatever chain X node and it's super easy for me to run like a cardano but the chain is so weak due to that node requirement that there's nothing that happens in it why would I run it right uh whereas on salana it's kind of the opposite it's like it's it's really valuable there's a lot of B activity economic activity I want to run a note to capture it can I make it accessible enough so that I can run it right it's kind of like how uh for example Tesla right like first Elon released a roadster right super expensive it just kind of wanted uh it led to people wanting the car and then he slowly reduced the price and started making more accessible and once people actually already wanted it and and you you saw kind of how alluring the car was everybody now wants to buy it now Tesla obviously has crazy sales numbers so I don't think any way is correct inherently obviously there's different paths anybody is can take but I but I do think it's super uh naive to say like oh well Salon requirements are high therefore it's centralized it's like no no no like alltech kind of starts in this phase where it's less accessible at first and then you slowly make it more accessible uh over time as the demand gets there or the supply get right so like and and I think like ethereum has kind of noticed this too in a sense where like Metallica has that endgame piece where it's like okay maybe block production is done from Specialists and maybe slightly more centralized entities but then verification is trustless and and uh uh sens resistance is also added right and so those actually ethereum and salana which is what it's interesting because they're much more aligned than for example something like a salana versus say or like an ethereum versus like another um eth competitor for example so um anyway uh but I I do certainly agree with um your your original point about like you know is that even enough about like 39 200 validators uh and then the Aptos guys actually comment on this too and they say well actually the number of full NOS doesn't matter it's actually the stake that matters and that that could be um you know I I think it could be I think there's some Merit to that um I don't what's the what's the angle they took on that like Stak is in the economic security driven by that stake I I think um I think they were saying like if you have a bunch of nodes with with a low stake you kind of just have like replicated stake and that doesn't really contribute too much to security as what they say or safety because you can um like uh cause like a livess failure right like if you have some sort of the stake uh some some percent of the stake um but I think it's I don't think it's flawed but I don't think that actually works that well because at the end of the day like so to's version of this is like you just need one node and then let's say that attack does happen right you you can reboot with different Quorum and some people think that's just like a big no no but I don't think you can ever get over that right like I I I think that's just a fundamental part of blockchains where you can't just eliminate the need for social consensus alog together you can scale it and I think that's kind of the vision that toally has and I think just when you play like these games with stake as opposed to like full nodes I think it just becomes much more um like it's I don't want to misrepresent their position um and and so maybe it's it's good that I'll send the um uh their actual tweet directly and then we can put in the in in the notes um I I think it's like okay but like I think they're focusing a bit too much on like temporary like liveness failures or like a not even really less failure but just like a you know the leader is just not producing blocks for some temporary amount of time whatever that amount of time is and that's super expensive to do right the more Stak you have obviously and so it's like why would somebody do that and to be clear I think like most people disagree with totally in like Solana's version of this where we like people say like Economic Security is a meme it's like it's not like a meme it's just like extremely over indexed on at the expensive other aspects of the chain that also matter um because like I think it's super important for proof of work blockchains right like Economic Security there is is life or death basically but on a proof of stake like or or like a chain like salana the worst you can do with like a economic attack is temporary liveness failure at Great cost that you only really get once because people will figure out who you are and then you know set up a new Quorum um so the econom activity better be really really worth it there and maybe it is right so I can't I don't know um yeah yeah know I loved your analogy um at the beginning there of the the Tesla Roadster uh Evolution right to like the plaids that we have today and uh how they got more accessible through time it's like you you just have to have that proof of concept and then continue building essentially which I think is super helpful context um you mentioned vitalic at one point there and I think that's probably a great said way segue to jump over to some of his tweets so we'll put the link in the show notes here but he was talking about Bitcoin so I don't want to take this too far out of context but he had an interesting tweet that said honestly I'm about 3x less confident in the simplify L1 even at the expense of more complicated l2s concept than I was five years ago the challenge is that when you trade off the challenge is that when you can trade off between L1 bug risk and L2 bug risk it's not actually clear that the ladder is better if you have an L1 consensus failure stuff breaks cev Scramble for a few days but you eventually get things uh things all right again with an L2 bug people could permanently lose lots of money so when I say it can actually be worth adding some pretty sophisticated L1 features to reduce the code burden of l2s and allow them to be reasonably simple you know it kind of like that that seems to be a reasonable position um and again he's talking about Bitcoin here right because you there's no Bitcoin needs another op code to kind of get these true l2s that we see on ethereum kind of uh able to launch on bitcoin so it's interesting that he has that perspective and zooming specifically in on the I'm about 3x less confident in the simplify the L1 even at the expense of more complicated l2's concept again he's talking about Bitcoin um but if you just think about that mindset and then look at ethereum you know it's doesn't necessarily contradict the the current rollup Centric road map of ethereum but it does like consider you know is that really how he thinks about ethereum now like I wonder if he's any percentage change different from now in five years ago on ethereum's particular road map yeah I mean I I think like certainly like he is because like I mean he also released another blog post earlier about enshrining more stuff into uh uh the L1 itself and so like he obviously can't say anything like well I'm less confident e theum like he can't say that there's just no way he'll get he'll get tortured uh but like you can obviously read between the lines um you don't need to quantify it but like it's it's quite clear and like many people on ethereum also agree with this right like I don't think like for example the scroll guys like Toral like he also agrees with this um and there's other people who who I've talked to from ethereum who agree with it which which is to say that like pushing everything off from the L1 to the l2s is like one approach but that approach I don't think works long term and and the analogy I use for this is like a race car right like let's say the engine of the race car is the L1 okay now to make the car faster once you built the engine maybe you add better aerodynamics you add different kind of uh weight coefficients or Dr coefficients maybe you add really good tires some hydraulic systems Etc okay once you do those though you have to come back right like the engine at at some point says wait a minute this is physics I cannot go past this okay and I think once you've already added all these things to take advantage of the existing engine now if you change the engine you have to change all these dependencies again to work with the engine that's insanely complex and brittle and and fragile right and it can work right it certainly can work but if if if if we are building this industry for the next you know many many decades then why not not just optimize the engine as much as you can first and then work up from there right because it's going to happen at some point anyways and that's kind of like maybe what salana folks uh uh uh that's the philosophy right we I'm like so for uh uh for example uh uh people think I'm like super anti- l2s or something that is not the case I'm actually not against l2s at all I have invested in eclipse for example uh what I am against is neglecting the L1 and then going to l2s and then L3 that also depending on these l2s that also depend on the l1's right that's just it makes me uncomfortable and I'm just the the philosophy here is optimize the L1 as much as you can first uh uh make it make it robust and then once you run start running into uh uh issues there okay then you know if you need the al2 go ahead add the al2 it's permissionless blockchain right I can deploy an al2 on salana right now today nobody can stop me uh what are they going to do right like uh and and so in fact um a few teams were actually debating using Sal as a DA layer and I actually uh I was I was helping them and and so um there this this this notion of of of like Sal people against or at least technical Sal people being against al2 is just not really true it's more so there against the idea of moving scaling off of the L1 before the L1 is optimized right it's kind of just like an engineering nitpick in a sense and we can see that vitalic actually at least sort of start to agreee with the direction of that and I don't think the rollup Centric road map is false because like eum does need it based on how they architect to certain things but I do think the L1 still needs a much more work before you just offload everything to the l2s uh and then now there's going to be because now with like things like uh Ras people doing RS of service once you start launching all these rollups that already have certain assumptions they make of the L1 it's going to be super like I like a car right like you you you you release a car and then oh wait there's a part that we need to actually now rechange what are we going to do well you either release a new car or you do a recall there's no easy solutions there and one like we already don't have any like you know that useful products in fact probably the most useful one is farcaster which barely even used as a chain uh right and while that's already the case why not just just just bite the bullet and and do the hard work of scaling from the bottom up or at least building from the bottom up and then we can obviously get to the l2s like nobody's saying don't do rollups um obviously the problem with that is incentives and VCS and tokens and all this all the stuff which is the fun side of crypto is just uh uh excuse everything but yeah that's that's those are kind my few cents on it yeah yeah and specifically in the case of ethereum I think anat toally put it best when you he he frequently says they're steer they're steering a multi-billion dollar ship like you can't just like overhaul everything at this point so you know part of it is like carrying Tech debt I don't know if you totally like people definitely disagree calling what they have today Tech dead and I think that's totally reasonable um but it's just like more so to the point there is because they were first they had to go kind of do a lot of you know uh a cave exploration and and Paving the way for everybody to come after them and yeah people can now look at okay well because they went this way they're dealing with these externalities but if we just put a fork in the road from three years ago start there we can create our own path and there's just like so much to be learned there um when it comes with the l2s I think the stuff that polygon is doing with the aggregation layer is probably what the end state of a lot of this looks like but we are still so far away from that I uh I bridged over to Bas there was some interesting stuff going on there the other day related to farcaster I was playing around with it had enough and was like all right I'm going back to to arbitrum getting from B to arbitrum took me two days a series of failed transactions four different Bridges across um or sorry it was uh synapse and Wormhole portal gave me failed transactions because my rabby wallet couldn't read the current usdc balance I had only the Explorer could um and then uh rabby doesn't connect to cctp domoney so I couldn't use circle's offering to get across Ross and the across Bridge doesn't support native usdc from base to arbitrum so that is like you know I I I work in this industry every day I'm I'm boots on the ground I tried four different services and I couldn't get there it took me two days eventually my transaction just worked on on one of the liquidity Bridges um but my point really is like we when we look out at the landscape today you mentioned Rass and these conduits and calderas cranking out chains left and right it's very impressive what they're doing but we just keep making more chains without any way to connect between the two of them and that that's I think that my weird non-consensus take around all of this is every new rollup is introducing more friction and exposing more users to the Pains of this very modular world that it's actually negative EV for each new rule up to towards the modular thesis which is is kind of counterintuitive you'd think oh more chains more rollups that's Prov proving that this thing is the right way to go and that it works and people want it but the reality is the second you're on that chain and you want to go somewhere else you got to go through a bridging experience and stripe Amazon Shopify these like kings of consumer interaction have all put out plenty of research that says you add one process one more web page load one more click one more second to the checkout flow and you're killing conversion users give up uh and now we're adding this like pain painful infuriating experience of bridging between two chains and not only is that an extra process but it sucks and yes sometimes it works perfectly and I'm sure I'm going to get dunked on because I'm like oh you're an idiot like you didn't use XYZ Bridge like when I have to go tell my normal friends of hey come check this thing out on chain like it'll be fun and then they go through this experience it's embarrassing uh and so if I could like go back in time and try to like change the course of the future that is probably what I would try to change is like pushing inop before we go creating all these chains and like the IBC over in the cosmos ecosystem does a pretty good job you know there it's not perfect uh like for example Rel layers are not incentivized so if you want to run a relayer for your chain uh you know you kind of have to do it out of the kindness of your heart or usually like the core teams or some of the the uh you know longstanding members of the cosmos ecosystem will do this uh for chains but like you know I joked about bridging to say earlier like they had a relayer issue and like getting there was impossible so you know IBC still has some Growing Pains but broadly that's probably one of the better standards I know Avalanche uh and their WP messaging or awm is quite similar and try good that's like based for subnet to subnet communication um but eth doesn't really have that today and I I'm I'm like worried about how the evolution of that is going to play out yeah I mean you know you're you're a researcher you you you you are on the absolute Forefront of the industry and if if you find the bridging ux complicated then it's complicated there's no other way to put that like is there like another solution that somebody knows that might be better okay sure but that's you know you can't blame the market for uh Your solution is not surfacing properly and um yeah I think I tweeted something like the number of L2 is increasing like as as n increases the probability of success or salot goes up or something like that um be because um yeah like PE people think you you can just abstract all of this away from like all everyone but it's like I'm extremely skeptical of that because of the sheer number of permutations and security assumptions and Trust assumptions like with web 2 you can abstract this much easier because you don't care if something is centralized in web 2 or decentralized or has whatever trust assumption but in crypto you do right the second you add a weaker link your entire trust model is different right like if I'm abstracting over like all these rollups all these uh Bridges but that maybe one of these Bridges has like one multisig like now that's totally different and that's a that's a big risk or or Central uh uh point of risk and I do think you can obviously improve the state of abstraction and and aggregation as well but I think fully eliminating it like the friction is I don't think that happens uh I just because just as an engineer like the amount of things you need to abstract over first of all that's crazy enough with the amount of rollups but but also the amount of like those things are always changing we're in an early industry right it's not like TCP or or like IP where these are like neutral protocols these are companies that have raised VC funding with tokens and they're going to move at their Pace they have their dependen IES and you have to abstract over all these different things with different dependencies different incentives in an aders environment where everyone's trying to steal your money and hack literally everything I mean bridges are where all the hacks happen um while we don't have any products okay so you put all those things together and I'm like you know maybe I do think it gets there at some point but I think it's much longer than people actually think personally uh either that or what you need to do which some more reasonable people that I agree with at least say is you have arbitrum and you have another L2 and like what 99% of all activity goes through those massive Roll-Ups okay that's really the only other approach in my view right so like there's the approach of par distribution of like two major l2s or three major l2s and okay that's fine there's the other approach of thousands of rollups that's I I just cannot see that working um at least within a reasonable time frame um and then the other approaches obviously L ones and maybe the L ones are kind of interconnected in some way um I think the first and last points there like multiple massive Roll-Ups and then multiple massive all ones work I think that's where the industry goes the thousands of rollups thesis I cannot possibly see that working um within a reasonable time frame again yeah I I tend to agree with you it'll be really like there there the Thousand rollups thesis will the million rollup thesis we're going to see it in full force this cycle there is no doubt about it there will there will be a lot of rollups um but again my my base case is that that introduces the frictions associated with that outcome to users and it'll be it's up in the air on how the users respond to that but uh last thing I want to hit on here is actually a specific L2 because I've kind of had a uh a I don't know if a come to Jesus moment's the right term but I had a change of tune let's say uh so blast horrible launch strategy vile marketing they called people pre-rich for depositing into their one-way Bridge contract that didn't have an L2 attached to it you know there was no product um that really frustrated me you they had like MLM style images of the referral scheme not illegal but questionable morally and it's just like it was frustrating and it was very hard for me to chew on that you know this was heat of the bare Market uh definitely the bare Market Blues were at play here we don't have apps we don't have products and like now we just have this scam L2 launching but yeah then i' I've been thinking about it I've been thinking about it first I thought it was a scam then I was like damn now it's A2 billion dollar scam but my my realization now is like this is just a casino and casinos don't have like morals that I necessarily align with they you know they advertise that you're going to come in here and make a bunch of money and games that are like largely rigged against you and you know strip clubs put attractive women on their signs like it's not like there is moral advertising or questionably moral advertising exists everywhere in the world um so I think I kind of felt guilty to just being pissed at the time uh because if you so they just did a builder competition they had 47 winners from this competition about people that are you know pitching apps to build on the platform and they accumulated over $2 billion in deposits deposits and they they used this as Leverage to you know show off why it's exciting to go build on this chain right they in their in their Builder competition they had this quote which is this is your chance to get in front of all of our tvl and our depositors right that's not I guess it's more of a paraphrase but nonetheless they they were using this as like hey come build here look we have these users and at the end of the day you know they're rewarding users for depositing into their you know one-way Bridge contract that doesn't have an L2 attached to it yet they're rewarding you they're like there there will be an airdrop allocation for you um and like that that's all anybody is doing right now in defi so to me it's like okay it really wasn't that different from everything else sure I don't like agree with it but love it or hate it it worked and there's more winning applications in this competition than exist or are that are listed on defi llama on tronu Manta Aptos SE Stark net and near like that's that is proof of the pudding that this thing is going to work now are all these going to be revolutionary applications absolutely not but they uh they're probably going to get used and again like everybody is kind of launching these applications in this like gamified context incentives Galore like it's just going to be like a defi summer farming casino and sure that's not going to revolutionize the world and put crypto in front in in more people's hands and like that's you know that's not the greatest thing outcome in the world but like it's gonna it's going to be used it's going to be active and again like as an analyst I got to pay attention to it whether I want to or not like there will be activity there and it will be worth looking at what people people are doing and there will probably be some lessons to be learned uh in the incentive side of this and what works and what doesn't now I think the last piece on this that I was definitely guilty of being very mad was there's no way they're doing anything useful on the tech side and like revolutionizing how to build a rollup but they don't need to like that's not the world we live in we just talked about rups as a service providers that like can build chains in a matter of Clicks in in seconds right uh so something like a standard op stack chain with Celestia underneath is going to do the here and uh it's just very interesting to me how I was super upset about this at first but I've kind of realized like they had a Playbook they executed its Perfection sure I don't morally agree with a lot of the things that happened but at the end of the day probably nothing was illegal and it's it worked they have two billion dollars in deposits that's like launching at definitely in the top five of of l2s by tvl but maybe even the top three I haven't looked at the numbers currently but it's it's impressive work like Okay so look at kind of maybe two different approaches to this one is that what happened just now which is all these deposits were were processed and there's not Builder competitions and everything is going fine okay the second one of this is that it actually blew up and because due to some weird Anon multisig scam rug whatever um now obviously we don't see that aspect because it do happen but the risk of that was very real and you know just because I didn't die from Russian roll at once doesn't mean Russian R it's a good game to play um and I'm not saying like blast is even that bad I think like the ethics are certainly questionable I I don't think it would have been a scame uh I just thought it was morally super weird uh and but but you know I I never thought that they would just like Rob people uh there would probably be like some um trust assumptions that were maybe uh uh like for example like Bridge get hacked right like uh maybe the multisig people we don't know who they are that's a weird assumption that you're asking people to make and but but I want to be like very very clear in in how I think personally and it's like very inspired by let's say TB which is that like just because something didn't happen does not mean it's it's a good idea is kind of how I think about it and that's super important in crypto where people are like oh you know I I made money from this therefore I will AP on the shitcoin as well and then now you lost everything okay um okay the second part is um yeah it works obviously I think that's that's undeniable right like you look at the data it works um they they got a good amount of deposits they got uh good developer interests good mind share as well and I don't agree with the ethics of it obviously but who who gives a [ __ ] about that uh right like you know I care about it but I there's a whole Market that that doesn't and so you can't really like get mad at the market that's just what it is obviously those people exist and blast exists to follow that segment so it is what it is um the one thing I would say though is um like for example when people say like oh well like here's this TBL that you get access to all these users for your for your products if you are actually a product Builder you'll kind of be like wait a minute what do you mean who are these users well the segment of those users are just people who want to make money that's a huge market and that's probably not who your product is targeted towards so like the other thing that's a kind of an on tangent is in crypto people like are super super bad at segmenting their markets right like webu like normal Silicon Valley startups are extremely obsessed with segmenting the market properly for their ideal customer personas whereas in crypto it's like oh well if they want to make money they're in my market it's like no no that's why you don't have pmf you need to like be like wait a minute farcaster does a good job of this right like foraster targeted the right people like Builders over a long period of time um who want to jam on crypto thoughts and and ideas like it's a decentralized social platform and they start with a very narrow specific specific segment of that market they just say oh everybody wants decentralized social our Tam is three gazillion dollars and if we get even 1% of that will be sad it's like no no no you need to just Define that segment very well first um but but like you know uh I do think you know hopefully uh like blast did did uh obviously work empirically in terms of what they've uh said they were going to do uh I still think it's pretty morally reprehensible uh I I think like Pac-Man actually followed me and then I I was like [ __ ] blastic I just started [ __ ] on and just like instantly unfollowed me uh and um I I think I'm still I'm still not a fan um I think things like arbitrum are much better right like arbitrum has proofs uh they have a known multisig like public you can know who the people are it's not a on like optimism as well but you know whatever uh and I I think like you're already kind of cutting some sort of introducing additional risk with an al2 like as V alic said earlier in the episode right like where where it's not clearly better or clear where putting on the al2 is less risk it actually might be more risk but then if you now make the al2s even riskier where you don't have proofs you don't have a functioning L2 you just have a bridge and you don't know who controls it and it's like some MLM scheme now we're getting a little weird like I think if I were to use an al2 or build out an al2 I would almost certainly use arbitrum as opposed to literally anything else out there right now yeah yeah no I I I definitely agree with your sentiment there and I think you brought up the user base here which is a great point this is probably the most reflexive user base you could ever come across so if you're a long-term Builder like that's not who you want using your product right you want Power users that will be there forever that will fall in love with your product and can you know be consumers for for for life um that is probably not what the BL average blast dep even the upper cortile of blast depositors are going to be like they uh they are there to farm and that's like a like that's kind of why my my framework on this is just a casino like that's what it is people are going to go there they're going to try to build fun games that like gamified things mostly probably related to farming um kind of like honestly what def5 summer was like I think that you're probably going to see a lot of the same stuff get run back round to um so is that going to change world again know and back to our previous discussion at the beginning of this podcast we need more like meaningful applications on the consumer side don't expect those to to kind of be there uh but I changed my tune on it and I am paying attention um so that's kind of that's kind of where I'm at but uh we covered a lot today man anything else you want to you want to jam on before we leave it for the audience no I think that that was good this was an awesome SC discussion today Mt and uh thank you to the audience for listening to my first light speed Roundup and uh really excited to keep jamming these out uh until next time yes sir thank you for [Music] [Music] [Music] listening