Johnnatan Messias

Transcript

thank you so [Music] much hey everyone thank you so much for being here I really appreciate it today I'm going to talk about uh our work on inscriptions um yeah so but just yeah um I was already introduced but just let uh just to let you know I'm a computer Scientist by Nature um I have vast academic experience with a lot of Publications and also experience in the IND industry working at Chic Labs uh metr labs and Kumi and I also have taught taught uh lecturers uh for Masters undergrad and PhD students so my research interest they rely on social

Computing so basically I have a bunch of Publications on this space machine learning and of course the centralized Ledger Technologies like blockchains but then whenever someone ask me uh okay so what actually you like doing and then I usually say yeah you know socially disruptive techn olog is my thing so that's why social Computing machine learning and decentralized Ledger Technologies so in the blockchain space I've done uh a lot of uh interesting work for instance uh with regarding decentralized governance as well so we have these papers where we show um token concentrations um voting costs for um on decentralized autonomous organizations we also have worked on airdrop as well

so basically whether uh the platform conducting the air drop is actually or not achieving their their goals so all these papers are available on my website feel free to check it out and of course data paper so um this is essentially uh we released um last year uh actually this year no last year sorry um uh one year of Z syn data we wrote um a paper about it describing some uh interesting analysis that you could do and also um regarding inscriptions which is the based of this top other than that I've done a lot of work on the Defi and VA space as well with Publications on on interesting place let's say Financial cryptography and data security as well

as IMC uh but today of course we are going to focus on this specific paper uh regarding inscriptions so let's start so let me first then start with this plot so this plot shows the number of transactions issued um on those five chains per time which means that as you can see here there are some spikes so for instance around November 23 on December 2023 and even in January around January 15 so these spikes they were really interesting but the thing is yeah you know for instance here in November uh 26 it was basically around the time where

Dev connect was being run in instan and I was there so guess what I was attending a venue and then something happened in the e um sorry in znc blockchain at the time I was working uh at metalab so yeah I left the venue and just went to the hotel to get my computer and and start analyzing what is actually going on right so because of these spikes which is basically I mean uh for instance around uh December we had more than five million transactions being issued in just one day so there were a lot of down times on regarding these rollups for instance arbitrum Sur from this um zync as well because zync block explorer was lagging behind so

users they went crazy uh as well as regarding arbitrum so and there are many more other news regarding this but then yeah so there are a lot of this transactions so we can ask some questions what are they doing right what are these transactions about also uh who are actually issuing these transactions and why so for instance uh is this coming want an end Zoom because of I mean you know think about it uh rops they have a b they have an infrastructure and they need to accommodate the pro the execution and process of this transaction so as much as you have you need more uh it demands more computer resource which means that probably it can also install the roll up

Network and how much fees these users were paying right but then okay I started Gathering some of these transactions just a sample of them and then here basically we have this a kync egg block Explorer I I hope you can see this and then we can see some uh interesting things so the sender and receiver are the same H interesting why is that so also they are transferring zero eer again why is that um so let's check other transactions and see if they have some uh similarities again same thing um same iser and receiver zero transfers it's strange but then one thing I noticed is that they have the

same input call data so what is this about so yeah we could basically uh convert this hex code back to a key and then surprise we have kind of a Json format file rating to the blockchain and we are talking about a lot of transactions essentially with this approach we can also filter all these transactions following that pattern which is this 0x 64 and so on and so forth that basically thr translates to data double column which means that we can filter among all the transactions available those that match this pattern and this uh I I basically after some digging on some websites I I go out to know that these are essentially

inscriptions transactions but what are inscriptions so some definition here basically this trend originated uh as ordinals in Bitcoin right and of course it basically reached evm chains they are essentially inscribed data or arbitr data that is written um that that people are writing to the blockchain right typically of course they are written in the transaction input call data and um so you could basically have um Json format files in the case of token inscriptions which is the base of this work as well as image name it you can encode everything there right and they are typically encoded in the transaction

put call data in Hax uh format of of course and um they do not rely on any smart contract remember those two transactions I showed there was no smart contract been called right and then some um key operations that they have we you uh can basically for instance deploy um an inscriptions uh uh an inscription on the network so you need just to write this deploy command that's what uh these uh inscription companies were doing and here are some examples that you can see uh for instance here we have the protocol zc20 remind I mean it reminds us some erc20 right um the deploy operation the T sync which is kind of the token's name

as well as the total Supply and the maximum amount of inscriptions you can claim per transaction and by claiming I mean of course there is an operation for that so whenever you as an user wants to claim some inscriptions on the blockchain you could basically follow this PT in here which is again you sub you inscribe this data to the transaction um input call data with the operation mint the token protocol specification and the number of inscriptions you are getting or the tokens um similarly you can also transfer to users so there are some examples there uh on top of the screen you can list on some marketplaces for

instance on the K sync you uh you have Z KS Market as well as Bing in and you can also buy or sell through these marketplaces as well uh so basically when you list in a Marketplace you have to spfy of course the protocol according to the definition that on the deploy the operation list TI um as well as the amount of inscriptions you want to list and the price in way okay um then of course in any data paper we need data right so we were evaluating all these five transactions here and uh in order to do that we relied on two data sources mainly for instance for ZK sync we relied on an archive node or external

node where we could gather all the transactions blocks uh transaction receipts anything and here is basically our pipeline we could futter out these transactions using the tag that I showed before and we used we St these um files in park format and those are the beautiful tools we use to uh during our data science uh Analytics so here notice that we also use poers which allows uh a lot of beautiful things U so I can discuss this offline if you if you want um also we ried on nons query to get um data regarding Arbitron based Ean optimism of course anyone could deploy all the archive nodes related to these chains but think about the computer

infrastructure right and data space and whatever that you need so sometimes it's just good to rely on some external tools as well right given that you can prove that the the the data that they provide is actually accurate in terms of what you find on the Chain so with that we obtained uh here there are some statistics for instance um we had um we could identify 160 million uh inscriptions transactions transactions related to descriptions for Arbitron 70 million for Z case syn era era and then here we have the unique number of features the number of blocks in which they appeared clear clear that were a lot right so let's review is it the first plot I showed you about the transactions distribution when is this

coming to on an end right keep in mind that now we are in the future in compar to before so it's easy to spot on the plot but think about in real time right it can get scary sometimes so when one thing we noticed here is basically on this plot I show uh the fraction of or we can basically translate it to the percentage of inscriptions transaction in compared to the total transactions in which chain clearly here we can see that for instance around December 96% of all Arbitron as well as the kync transactions were essentially inscriptions 96% and here 50% uh eon of course we have some spike here for Aron

for zyn zy again so clearly speaking Yeah people were engaged in a lot of inscriptions right and when let's then investigate what they were actually doing right so basically regarding the top protocols that basically were that that we saw on the the the blockchain most of the chains here they have their top protocol but as the cas s they have um two mainly protocols which is zc20 and layer 220 right um and when we look at the type of operations that people use of course we would expect mint right because basically user Bo meeting where deploying uh just one per token and protocol right U but then as you can see

there are um maintain claim here keep in mind that they are kind of the same right it's basically um the way that the protocol defin some of them def find hey if you want to claim inscriptions you just issue um a Json with an operation list and the other oh yeah we use me instead but they are the same um and then regarding tokens tokens are essentially a com combination right um of the um token protocol as well as the ti which is the the name because in each what they call protocol uh can you can have many tokens or ticks there so here we see that again uh there are um top leading um inscriptions tokens here for each of these chains in ethereum um it's

a bit more uh balanced but of course we have this um earc any nodes here that basically uh has a spike and there are all the others uh they have less than 1.52% uh of all the inscriptions issued to etherum ZK Sync It's really interesting here because zc20 sync it basically had almost 42% while or layer 220 um uh L2 uh 33% so yeah those two were the line one in um on the cas syc but then uh what about the issue there we see that uh for instance uh who are they right are they Bots are they normal

users one thing we noticed is that when we see here uh basically here we see a table that shows uh the inscription transactions per issuer which means that um here we report the average as you can see the standard deviation is pretty high which means that the the data is really skilled and the maximum amount is basically there was one issuer in that issued for instance in AR 38,000 transactions to claim the descriptions on etherum 67,000 while in casing for 40K um which means that just small just a few accounts were basically claiming everything um and then of course it seems that they are Bots right but yeah

we we did not went through we did not go through to identify whether they are actually Bots or not but come on these are basically automat uh they use some code to make some automations right um what uh and about the amount of fees they were actually spending on this look how interesting it is here basically we report the uh distribution of fees uh per token chain operation but clearly here we see uh on the K sync only Z c2n sync with the operation mint users or accounts I guess it's better to call it than accounts right they spent alone uh 88 uh 89 eer in fees on a rollup which

is quite cheap to issue transactions and when if we s up all those Zing uh these stop operations and the for these tokens for Zing here clearly they have spent more than 100 ether in fees which is a lot of money right um and then here we also report the median um um price that uh we had for each uh chain here uh one thing we noticed is that yeah due to the nature of a ZK roll up uh basically uh ZK s saw a few reduction because as we had a lot of transactions be included in in in in in a single batch for instance um then yeah to compute the proofs so

basically for the end user in the end they ended up paying less fees okay but in the end of course they together paid more than 100 fees only on the cas sync um then what about the future of inscriptions uh the thing is is it good or bad actually right uh people might say hey this is basically um I don't know not not not I mean this is maybe something that's not useful to the to the blockchain but you know uh but we also have to say that this for the first time was a real uh stress test for all the rollups and if we want blockchain to escape

to basically have Global adoption we have to deal with this high flux of transactions and we need for instance to on board our parents right we need these things to to do a better job in terms of scalability of course there are a lot of solutions uh coming um and this is really awesome right and regarding the opportuni so yeah the thing is you cannot basically exchange these token inscriptions for usdc right so because you don't have smart contract of course you can propose some solution for that ownership of these descriptions is kind of unsure um so the thing is although these things are on the Chain you rely on some uh application or the encryption

inscriptions deployer to basically keep track of the accounts balances so one thing I can can mention is that okay so if I issue a transaction with that pattern and issue another one from the web uh interface application of the in inscriptions deployer which one is actually going to be content the one that the application actually uh uh counted based because I I issued through their platform or the one that I basically issued on the Chain so for instance here um there is something that I tested regarding uh some inscriptions that were listed on one of these marketplaces and basically I issued a buy um operation there and the the

basically I bought some inscriptions and uh so one thing I did I tried to buy it again the same uh listing by the transaction and then guess what of course uh I I expect that but it's just for a fun fact is that you know uh I could I I basically sent the money uh that the seller was asking twice uh which means that of course my transaction wouldn't be reverted because there is no smart contract to check these conditions right so it was quite interesting interesting then hope hype so yeah you know um sometimes we have uh some new Tendencies happening on the blockchain uh there are for instance pump fun on Solana which is interesting

as well to analyze uh but then people get engaged on social media telegram Discord group groups right and one thing we noticed is that there was um really high interest for users during the beginning of the trends but it's it seems that uh it phased over time and um of of course as I mentioned there is no mechanism that could um impede or or avoid you uh repurchasing uh previously sold inscriptions right um and of course Bots as usual they can basically take the lead and me most often so for today that's all I have to tell you but of course feel free to reach out to me I'll

check out my other works uh available on my website there are some talk available as well on YouTube and of course feel free to ask me anything even offline regarding the other works as well with that thank you so much for being here guys I really appreciate thank you very much Jonathan I'm afraid we run out of time so if you can you if you have any questions if you can take them outside thank you very much thank you [Music]