Okay, nice. I think we have a new link

for YouTube. I'm just going to share it

with the people who are waiting on the

Luma.

>> Okay.

>> Yes.

Okay,

cool. I think we are live. I'm going to

share the link with people on

Relax.

Yes.

>> All right.

I'm almost done on my end here.

>> Cool. Awesome. I think we can get it

started. The YouTube link was I don't

know. There was some problem with

YouTube. So, I changed the link. I

updated on Luma as well. So, you might

get a notification like everyone who's

watching. But we are already live on

Twitter and LinkedIn as well.

>> Nice.

>> Of course, why not? Yeah. So there are

like 100 people watching right now. I'm

assuming a lot of people will join

later. But hey everyone, thanks for

finally uh joining. I'm really excited

about today's stream because I think

after once I'm so excited about the new

protocol that is coming in there was so

many good features and you know uh and

today we are going to discuss Tempo. So

if you don't know Tempo is the new EVML1

the blockchain designed for payments and

um yeah we are just going to wait for a

few more seconds. Actually not we can

just get started. So today our special

guest is Andrew and Andrew is the if you

don't know he's the co-founder of Herd

and before Herd he was at Dune and he is

a very famous and my favorite technical

content creator in the space. So if you

don't follow him on Twitter go and give

him a follow. Uh thank you Andrew for

joining. How's it going?

>> Thank you. Thank you. I'm back back at

my childhood home in LA for for this

week. So that's why my Wi-Fi Wi-Fi might

be a bit slow. Video's off so people get

to see my PFP. Uh but I'm normally I'm

not trying to be an or anything. My face

is already my face is already out there.

>> Yeah. Yeah. If you want to see Andrew,

go to his Twitter and, you know, listen

to his talks. He gave a good talk this

year at Dune Con, so go and listen to

that.

>> Yeah. Yeah. Yeah. Yeah. Yeah. Uh

>> okay. Yeah. Let's get started. Uh so

just the agenda for everyone. Um we are

initially I'm just going to start with

the quick overview of the features of

like the tempo and what they are trying

to build and then we will play with it

and try to like deploy a token and you

know a stable coin because of course it

is a stable coin focused chain. We will

explore their new features like native

account abstraction. Tempo has their own

explorer like they built their whole new

explorer. So we are also going to check

that out and we will try to see what

breaks what is working. It's a EVM chain

but there are few differences as

compared to current EVM chains right and

the chain is very much focused on stable

coin payments and it is engineer it is

engineered in such a way and these

features that I'm talking about like

native account extraction and new uh

token standard they are available at the

protocol level so that you know

developers can focus on building the

application instead of you know building

around the infrastructure like as of now

if you're building an EVM you will need

a different account obstruction provider

and a different guest sponsorship person

like infrastructure provider but

everything is built in in the in the

protocol itself with the tempo so Andrew

you want to take this stage and share

your screen yeah sounds good let me just

make sure I don't

share anything I'm not supposed to All

right cool so yeah tempo So, testNet

was, you know, I think it was a week

ago, two weeks ago. Um, and they have

great docs, but there are

there's a lot of things here in protocol

specs um that are somewhat different

from how normal normal not normal EVMs

work, but extends off of the normal

EVMs. Um, and so anyone who's I think

Tom mentioned try to follow all the

different account abstraction standards

or token standards or all that stuff.

Um, you know, it's usually pretty

confusing to navigate. Um, so I'm just

going to set the stage first by actually

just starting with the new transaction

type. I think this is the not the

easiest place to start, but I feel like

a lot of times when you have like 15 or

20 new concepts, you kind of need like

one concept to anchor to um as you learn

the other ones. Otherwise, it's kind of

like, oh, how does like tip 20 play with

fees, play with like the fee decks or

stablecoin decks or or block space and

all of that, right? So, I think if

you're reading about tempo, I would

honestly start start here. Um, and so

this main change is I'm just going to

specification. Um, is that there's this

new um, tempo transaction which has

types, right? And Ethereum transactions

always have types. Um, I think, uh, I'm

going to bung it up. I think we have

three Ethereum types right now. Uh,

Ethereum transaction types. I think

there's three. Legacy. Yeah, there's

legacy. There's Oh, I'm very wrong.

There's five. um I guess existing types

right so a lot of these types basically

introduce some changes to how uh you

submit transactions right which has

implications on wallets as well as on um

RPCs uh because there has to be

standards for how something how data is

structured when it's submitted um and so

yeah you see legacy types are here um

and then 209030 which has access list

has has a different type uh EIP559

which was from 20 what is this 2021 now

I think um this has type two um uh this

one's less important to us right now but

it's important to roll up people um and

then 7702 transactions was the newest I

think that was May this year which

allowed you to delegate your um your

wallet to a contract right so you can

turn any EOA into a smart a smart

contract account essentially without the

4337 approach. Um, and so yeah, this is

all a lot harder to follow when you're

just um reading, right? Uh, so I

normally always like to explore by

pulling up an explorer and then also um

pulling up the the alchemy dashboard for

for testing things out. This is like

anytime someone's a comes in as an

analyst or data engineer and they're

like I want to learn crypto data or like

I want to learn how to build crypto

apps. I'm just like okay do not start

with like GitHub or Dune or like

something you saw on Twitter. literally

come to the sandbox and get familiar

with the RPCs because if you don't

understand this, you'll have no idea

what you're building on top of in and

like it doesn't matter if you're a data

engineer, app dev, um maybe first lady

dev is a bit different, but you know, um

I've I've used this dashboard for five

years. It's changed a lot, but it's it's

just been mainly the same where you

select your your your chain and then

you'll have uh you know probably a

hundred or so different um RPCs here,

right? And these are all standards.

Yeah. Um

so let's actually I mean we could just

start with the tempo explorer actually.

>> Um so if I just

go to here, copy this hash in we can

start simple. We can do a get

transaction by hash.

And if I put this in here and I send the

request, you'll see uh this type is

actually legacy

um based off of here.

And let's actually this this now gets

into the the fun part of I am not

expecting everything to work

immediately. Uh I think part of the

things that happen when you're creating

a new standard uh when you have a new

transaction type is it has to be

supported across obviously the chain the

R the uh the explorers the RPCs etc. U

so I think for today's stream we're very

much like we're testing things out and

seeing what is expected or unexpected.

Um so I would have expected that this

type was 076.

Um, but I guess that maybe for Tempo,

especially from their docs, they maybe

they default. Oh, yeah. They actually

say it says legacy here. I'm done. Um,

so it's interesting. So I think I can't

tell if this is from their docs or not.

So let's actually The nice thing about

Tempo Docs is that you can actually

deploy or interact with the chain from

the docs. Um, so I'm trying to remember

where

let's see if I can find it. They let you

create a stable coin straight from the

docs. Here we go. All right. So, I can

sign in. Please don't dox me. All right.

Uh great.

Now everyone knows my password length.

Very fun. Um,

let's just call this Am I allowed to

call this alchemy? Probably not.

Call it all good. Um,

so I'm deploying a a transaction from

their docs because I'm curious. Yes.

Okay. So from their docs, the

transaction type is 076. So if I plug

this in to here,

uh, I should be able to see. Yes. Now we

have this new this new type here. All

right. Um, and again, you could you can

always see it defined in here already.

This is this is Rust in case you're

curious what what the hell pubstruct is.

I think this is Rust, right?

>> Yeah, this must be Rust.

>> Um they love Rust. But I always prefer

to go from docs to onchain as fast as I

can. So I'm now looking here and I can

see all right, let me side by side put

this transaction with um what I see here

because there might always be things

that just aren't shown in the explorer.

Um, so something else that I might want

to do is I actually want to um

put these side by side. So this was a

legacy transaction, right? Yeah. So I'm

actually going to go and put this one in

as well. Um, and there's a few ways of

going through this. One might be I just

copy both of these into chatbt and have

it explain to me the differences. Um,

but since I've already read the docs,

there are two key things I'm interested

in. Um,

the key authorization doesn't matter to

me right now that much. There were

changes in how you can sign things that

extend thing the capability of the

chain, but we're not going to talk about

that now. The main difference is this.

Um, you now have this calls array,

right? And for calls, you can see two

value, input, and data. Um I actually

didn't know about this data field

before. Was that in the docs?

Uh

vec call. Ah their docs are out of date.

So there's two value input. Um I see

data here. Uh my immediate question is

now is this a duplicate from the RPC

where input it was named data at some

point and then it got replaced to be

called input because that is what input

used to be called or input is called on

legacy transactions and data was not

deleted from the RPC or is this

something where data does exist here but

it's just not updated in the docs. Um,

so that's a question from my end and I

don't see a link to the repo for this

transaction type. I'm sure there's

>> right a Rust implementation. There it

is. It'd be nice if that was just here.

Like as as as someone who's been a

Devril, there are two things you learn

about documentation. One is no one reads

the overview page. Everyone skips to

subpage. And rule two is if there's more

than three or four things here, no one

reads it. Um, so anyways, so this is

temp transaction. Um, so they really

need like GitHub needs a chat just right

here. The co-pilot should be here. Uh,

calls call

>> uh pubstruct call

>> value input. Okay, so it's not in here

as well. So it's not that they forgot it

from the docs. This just seems like an

RPC problem which might be from from

refu uh or or from alchemy, who knows?

Um but anyways, point being this

shouldn't be here. Um and now if I go

back to what's different. Um so

transactions typically always have you

call a specific two address uh with some

value where value is like the ETH or the

native chain value uh or the native

chain token value. Um, and then there's

input, which this is how you actually

call a contract where the first four

bytes are the function signature, right?

So, there might be I don't know, this is

an uh an ERC20. I know it's a TIP20, but

it's like ERC20 token that has like a a

transfer function. This doesn't look

like the transfer um function signature,

but anyways, you have a transfer

function, and then there's bytes here,

32 bytes per argument. Um so it might be

like oh I'm transferring from my address

to another address. Um right and so big

difference now is

with this new transaction type you can

actually bundle multiple calls. Um which

we will actually do in a second. I'm not

going to do it right now. Uh but now

that actually is a structure that comes

from not comes from but is native to

Salana. Um so

let's see if the Salana documentation

has improved at all.

Okay. Yeah. So in Salana they call the

transactions messages and when you send

a message you have this vector of

instructions um which can I find it

here? Yes. So you can see for for Salana

uh there's a set of instructions that

has a program ID uh and accounts which

you can think of as

as the two basically and then you have

data which is input for us right um so

on Salana all transactions have always

been like multicall batched um but on

Ethereum if you wanted to batch

transactions you always had to use like

a multiall 3 um like a contract that has

um a contract that has an execute

function on it or multi call whatever

what oh I guess they have an aggregate

here uh but normally it's you see

execute or aggregate or something this

is common on like 772 or 4337 contracts

for 4337 it's called handle user ops but

it's essentially an array of call data

um and so this is one of the biggest

changes that they've made is instead of

expecting

people to use like account abstraction

where you have to create a smart account

wallet which is another contract or

delegate your current wallet to a

contract which is 7702 you now natively

get this batch execution um here right

and there's a lot of benefits to this

one of the main ones is for things like

the the classic I want to uh swap a

token but I have to approve before I

swap and that requires the user to sign

twice uh now this is only signing once

Right? And you know, there's stuff like

5792 for getting around signing multiple

transactions, but this is a much cleaner

way of of solving this problem. Um, and

so I'd say this is one of the main

differences that will affect devs,

probably affects RPCs. Um, and then

also,

um, just just the devx of and UX of

apps. Um, the other big thing that's

changed. All right. Sorry, we don't I've

not let you talk at all. I've just been

rambling. No, no, it's fine. You can

just go ahead. I was also talking about

there is one more method which is the

ETH get balance one which might break a

lot of changes because you know Tempo

doesn't have like a native token, right?

>> That's a good call. Let's see what

happens.

>> Yeah. Yeah. Actually, yeah. Yeah. If

you're going to call it, uh, as far as

my research, you're going to get like a

big hex number. So, let's do it.

>> Yeah. Let's see. So, the one that my

smart account

is tied to,

oh, UX thing. You should always be able

to copy any of these just with a click.

Um,

so now if I go here, if I go here, get

my balance, I see this, which I can just

take into

this place here. I'll swap these around.

convert.

>> What the hell?

>> Yep.

>> Yeah. So that they did it on purpose

just because they wanted to like make

sure that there is something there like

a large number, right? So this will like

break a lot of wallets cuz the entire

logic of wallet is that it checks the

accounts that if they have the native

balance or not, right? And if you don't

have it and if there is nothing, if

there is a big number and it's it's

returning a hex decimal, right? So,

>> y

>> the wallets are going to break,

>> I think it would actually have been

better to just set this to zero. Um,

setting it to a high number breaks more

things than setting it to zero. Um,

because I'm just thinking from

everything from like simulations to like

balances. It's like a lot of apps are

built around like oh if their balance is

zero um

there's probably some warnings and stuff

that come up.

>> Uh fair but if you if you set it to zero

then if if I'm if my account is trying

to do a transaction and it it checks

that my native balance is zero then how

am I going to perform a transaction?

>> Right. Right. Right. Right. I mean that

that also gets into I'm very curious how

let's see let's let's start getting more

advanced now. So I used AlphaUSD

for creating this token. If I go to my

balances I'm guessing was this

sponsored.

Can I see if who paid this?

How do I see who paid

>> in the explorer?

>> Yeah.

>> I don't I don't think I saw it here.

transaction fee. It doesn't tell me. It

doesn't tell me who paid my fee because

I'm looking here and then I'm looking at

my I was looking at the hash. If I look

at my account,

okay, I have some I formal alpha USD.

So, I have AlphaUSD, Path USD, which is

their native routing token, Beta USD,

Theta USD. Um,

okay. And I'm guessing this is

>> by the way for for the users who are

joining now, Tempo has this feature

because they don't have a native token.

So you can literally choose your token

that you want to pay fees in. And most

of them are going to be like stable

coin. You can go to their docs and you

can check out like how the gas system

works. But mostly just to give you a

TLDDR, they don't have native stable

coin. Uh you can choose your uh stable

coin of what you want to choose for

paying the fees. Yeah. Yeah. Um, and

that's one of the things I want to play

with is So, also this should show me

address or something on hover. Um, so I

have AlphaUSD. It'd be interesting to

see what happens or how we can pay with

Beta USD. Um, because then that should

go through their fee decks. Um, which

I think you always have to pay with a

stable coin. Do you have to pay with a

stable coin? Must be a USD native tip. I

actually don't remember if for FMM

the the token you're paying the fee with

if it has to be a stable coin or not. I

remember because one of their

>> does it it doesn't say here.

>> Yeah.

>> Fe specification

>> a tip is are tip stable coins or they

just

are they

>> no tip is like a standard like ERC

standard. I know that but because in

here in the docs all they say is you can

pay with any tip token right

>> um

>> but whose currency is USD so which means

a stable coin now

>> well I think it's just quoted in USD it

doesn't say exactly

>> interesting

>> that it's a st as long as that stable

coin who's currently as long as that

stable coin uh so that that grammar here

suggests that every tip is a stable

coin. Um,

which this gets into we we should write

a running list. But here, this is a good

time for us to actually run into code

now that we've uh been yapping enough.

Uh, I already have.

Um,

and yeah, I have no idea how long we're

going to go, but we have a lot to test

out. So, we already tested seeing a 076

transaction and compare it to legacy.

Um, some things we want to be able to

see is like tip20, how to see quote and

liquidity. Um, because the dexes are

um

the dexes are in pre-ompiles. So, you

don't have read functions. I think in

the same way of like if I was on unis

swap I'd have a read function for

getting the square root price or um you

don't have a read function for well yeah

you do have a read function for getting

like reserves and stuff u but I actually

don't know how we see that in um tip20s

with fmsmms

um so that'll be interesting to see are

tip20s stable coins always

um so we have that we want to trust test

sending a transaction with two transfers

and using uh

what was the other token? I have beta

USD. Let's try using beta USD.

Um which

which is

this one.

Um and then test reading traces

from this transaction.

All right.

Um and if we have time test calling the

pre-ompiles directly. Um so the way

these dexes work is there are

pre-ompiles. I don't know if I can find

a transfer here. Let's see if we can

actually find the the pre-ompile

address. Do you do you have the

pre-ompile addresses

account keychain? No.

They must have like some page that's

like hey here are all like deployments

page that's like here are all of the

pre-ompile addresses

right

>> it's not in the docs should be

>> I don't I don't see it this

>> actually show a page

>> I see this I fe Oh this is fee manager

is that the same as the feed dex

no feem contract is different

from

>> FedEx.

>> Feedex.

Let's see. I need to open my phone with

the with the um

with the chat with the tempo devs.

Yeah let's

>> where are the pre-ompile

addresses for the fem and stable?

Well,

need a

compile and

list. Uh

>> yeah for for also the people joining

joining now and also if you don't know

what tip20 is so tempo is as I said it's

a new EVML1 so they launched their own

token standard so it is extending the

current ERC20 token standard so Ethereum

has this ERC token 20 standard right and

which according to us is also like very

old according to me at least and there

are some problems with it and tempo is

like extending that standard and also

adding some new features, some new

methods to make sure that it is very

stable coined.

How do I say that? Very stable coined

aligned or

>> friendly.

>> Yeah, stable coin featured. Yeah. So,

there are like multiple features.

>> Sorry.

>> Sorry. Go on. No, I was just making sure

I was going to leak one of my cues.

Tempo test.

>> Oh,

>> I'm just going to start.

>> Yeah,

>> I'm going to just use it. Sorry. Go on.

>> No, no. Maybe I can just also share my

screen on the side.

Go for it. Um,

I'm going to need both of these.

All right,

we're going to set this in here.

uh

doesn't really matter that I'm doing

this because uh

this is all this will be deleted

anyways. Um, and now we can use cursor

and say, "Hey, can you try and set up a

Oh, how am I going to do this?" All

right. Well, we'll see.

Submit transaction to the tempo rece.

Um

or I'm just going to give it

this

with type USX2. Let's just try a basic

transfer of one token of

uh

let's just try transferring alpha USD

to

and let's just transfer it to oursel for

now. Where's my address? This is my

address.

Um,

by the way, we are V coding and trying

to do some transactions and see what

breaks.

>> I'm realizing one

>> one problem I have is

I can't export

I don't think I can export this this

this key.

And so, and I also, if I generate a

private key,

>> I don't know if they have a faucet yet.

Oh, get faucet funds. Here we go.

>> Yeah, there is a faucet. Yeah.

>> Okay. Uh,

>> and why don't you why can't you get a

private key? It's your metam mask,

right? Or you just don't want to show

>> Oh, no. like from I meant from the so

they use pass keys I think with Porto in

the docs

>> and I was I was like oh I want to just

use that wallet but I actually don't

know how I don't even know how to

yeah

I don't I don't even know how to pop up

Porto to to see that stuff. Okay. If I

add funds

to here.

Um, and I'm going to just tell cursor

spin up a new you

wallet for me. Just

>> till the time you do that, I'm just

going to take the stage from you. This

>> go for it. Go for it.

>> Okay. Uh, remove Andrew from stage. Add

my screen. So as I was talking about

TIP20 standard uh they have extended the

typical ERC20 standard just so that it

is more stable coin aligned and added

some really good new features. So one of

the features is transfer memos where is

that this one right so with that you can

like enable a 32 byte memo to the

transfer. So imagine like you're doing a

transfer on the blockchain right now

with ERC20 standard there is no way to

like drop a note for the payment

references right any invoice ID or any

note that you want to drop and actually

I've been talking to a lot of devs and

they told me that they need this feature

even on Ethereum because it's very hard

for them to like have a reference for a

particular transaction because if they

want to pass on that transaction to

another service right it it gets very

hard to like customize it Right. So

that's a really good new feature. And

another feature was that usually when

you are deploying a stable coin um like

let's let's say tether right they have

this pause function. So initially like

in the native like the original ERC20

standard there is no such thing as pause

right so you can't pause a particular

address or you can't block a particular

address to do the transaction. So for

that if you want to do that you have to

use something like open zap plane like

an extension to add that particular

function. But with tempo they have

already built in the block function like

the pause function inside the protocol

itself in the inside the standard

itself. So you know you can just

directly use it. Let me go to the code

and show that to you guys. Okay. So this

is the tip uh 20 standard the ERC20

standard again the same and all the

functions are already there and then if

you go down you will see the TIP20

extended function. So here you can see

the memo that we talked about like for

references and I think this is very very

important when you know when you're when

you are especially focusing on stable

coin and you know being probably like

stripe is going to be

I don't know taking in charge and tempo

is going to power stripes transaction so

it's going to be like a very important

feature for them and then the next one

the pause function that I was talking

about is here so you can like pause the

contract and you know block the

transfer. So if you go to any major

stable coin right now on Ethereum or any

chain they have this boss function but

it's not native. They have to use like

an extension to do that and there are

multiple like other features as well

like new features that I think you can

explore. Of course we are not going to

go through all of that. Uh but yeah,

this is like really great and uh you can

explore all of that. Yeah, I think

multiple features of of course we can't

talk about everything but they do have

still they have approval function. I

mean they could have removed that

because approval is such a pain in the

ass. I mean because

>> it's too it's too ingrained.

>> Yeah, it's just such a pain in the ass.

But the problem is what I think so that

they want it to be EVM compatible and as

they wanted to extend the ERC20 standard

they can't get rid of it. So but anyways

they they have native account

abstraction of course 7702

um you know on Ethereum as well. So you

know you can do that but again uh yes

overall I really like this uh tip 20

standard. Uh yes and if there is any new

method that you think is awesome just

maybe comment and we will explore that

too. Uh yes. Are you ready Andrew?

Should I pass on the stage?

>> Yeah, that was that was great. Great

timing for me to fix things up.

>> Okay.

>> Um

>> Okay. You have sharing the telegrams.

You might want to remove that.

>> Yeah. Yeah. Yeah. Yeah. No, I'm closing

this side. So this is a chat. So Alexi

says fee address is the same stable. I

thought that

>> Oh wow. You literally posted in the

Telegram group. Okay, nice.

>> Oh yeah, I will always bother people.

It's what it's their job.

I was pre-compile

not a contract.

Uh stable decks is at zero XDC. All

right. So, we're juggling a few things

at once, but I just want to see this

address is an account.

Interesting. Wait, so is Stableex is a

contract

and VM is a contract? I could have sort

I could have sworn at some point they

said that they were pre-ompiles.

Uh maybe there's pre-ompiles within them

>> or I'm just like

>> or I'm just losing my losing my [ __ ] Um

anyways so

I I added funds to test net to my

account. Um

I think the docs RPC actually lost track

of things. Um, but this transaction

hash, while it says it timed out here, I

do actually have it

here.

Um, so it says that that they minted me

1 mil ALUSD. And so in in my um repo

here, and we'll share this repo after,

uh, I have a generate wallet. This is

literally just for testing. Never

generate your wallet and store it in a

in in any file in a repo. Do not do

this. I'm just doing this for testing.

This is just test net. No real funds

will ever go on here. Please do not do

this. Uh but for testing, it's fine. So,

uh my public address is here. And so now

that I have this alpha,

um yes, yes, yes, yes. Made a mask. I

know you have new features. Okay. Um,

I'm going to Oh, I don't pick I have to

pick the asset before,

dude. Wallet UX

is painful to me. All right, so I need

to send I need to pick

tempo test net.

No token matching your filters.

Excuse me.

Can I not?

Hello. account one.

All right, let's see what's going on

here. Um,

so here sender I need to view the full

transaction. It is sending to me, right?

And I do have 1 mil of alpha USD now

which is at

this address.

However, when I go to transfer

on tempo,

oh, was it just not show? Oh, hey, this

is your this is our friend that you

mentioned our uh giant balance.

>> Oh, yeah.

>> So, it shows up as USD.

>> That's that's fun. It shows up like

that. I can't search the token for some

reason. Um, so

Can I Can I send it to Zerion?

How do I overview explore?

Send.

All right, let's

do they support tempo test net? No, they

don't support tempo test net. Great. All

right. Um,

I don't think I have any way of sending

this.

I don't think I have any way of sending

this to my

uh new generated account unless I export

my Metagmas private key. Um,

so I'm going to stop sharing.

>> Okay.

I think I'm going to take the stage till

now. So you

>> Yeah. Yeah. Take take the stage for a

second

>> while it is not exploited.

>> Yeah. Yeah. Yeah. Yeah. Yeah. I'm not

sharing my screen. Right.

>> Yeah.

>> Okay. All right.

>> Uh

>> that is this is cursed. This is

absolutely cursed.

>> Oh, by the way, okay, no worries. I can

I can take this stage for now. Okay.

Another new feature that I'm excited

about it is the block space. Right. So,

Tempo has this dedicated block space for

stable coins. So, and this is not a new

concept. Uh there is a project called

world chain. World coin if you have

heard of them. Solana also has this

concept of dedicated block space where

the space is reserved for few

transaction in in case of tempo that is

reserved just for staple coin

transaction because they want to focus

on staple coin. uh in comparison to

world chain for example world chain

>> I'll be back in like two minutes a

second

>> okay cool world chain is trying to

create this human proof of human

transaction where you know once you get

identified by their orb you prove that

you are a human with that they are you

know reserving the block space for the

humans right so that is fun right and

similarly tempo is trying to do the same

and uh They have this function. Wait,

wait a minute. Am I on the right do

talk? No. Yes, they have this reserved

block space

for stable coins.

H yeah, this is interesting. And then

another feature was the fees. I think we

have already shared about the FMM Fmm.

So there is this native FMM that

converts automatically converts your

token to the validator's preferred fee

token and eliminating the need for users

to hold any other separate gas token

which is really huge. Of course if you

have been coming from the Ethereum you

know that's a again pain in the ass for

developers. Um and with native account

abstraction you can like literally pay

network fees in any stable coins and um

batch and concurrent transactions and

then also schedule payments and you know

you can do modern authentication with

like pass keys that is a good one like

face ids and with your biometric. So

that is really huge. Uh

yes these are the few new features and

what else did I miss?

I think that is it, right, Andrew?

>> Sorry, I just got back. I'm asking the

team, someone from the team to send me

funds.

>> We're literally live debugging. Okay.

>> Yeah.

H the tip 20 rewards. Okay. Yeah, they

do have this DIP 20 rewards as well,

which is like built-in mechanism that

allows for distribution of rewards to

token holders. Oh, wow. Interesting.

Oh, I see. Staked in a separate

contract. Oh, wow. This is really

interesting. What did you find

>> the tip 20 rewards one? Right. So,

initially like right now you have to

like stake your contra stake your tokens

to like get rewards and stuff, but they

already have like this tip 20 rewards

thing which is like a built-in

mechanism. Um, that's interesting.

So, you don't have to stake any tokens.

Oh, yes. Yeah. Yeah. They have like auto

I guess auto rewards. Auto re It's not

rebate. I think it's just paying out. I

don't think it's actually increasing

supply, right? So, it's not like a

rebase.

>> No, no, no.

>> I think uh someone did that, too, right?

There was an L2. What was their name?

Blast. Native like native

>> Oh, yeah. They had native native yield.

Yeah. Yeah. Yeah.

>> Yeah.

>> Wow, that's a fun one. All right. I

think I'm going to download another

>> um wallet. I'm gonna download Phantom

just to generate a key for me to safely

export.

Uh,

create a new wallet. Create a seed

phrase.

Yes. Yes. Yes. All right.

>> Oh, nice. Yeah. So also by the way there

is no me on the temper blockchain which

is very interesting. I can create a

whole separate video and do a different

live stream on this but they're

architecture is designed in such a way

they have like this native VMM. So you

know there's no probabilistic ME.

>> Oh

>> there will be ME though there will be

ME.

>> There's there's there's always there's

always MV me.

>> Yeah

>> there's always MV.

>> All right. How do I manage accounts?

Developer settings. No. Uh experimental

features security.

>> I think we should do another session on

like a whiteboard session on the

architecture. I will try to invite

someone from tempo.

>> Yeah. Yeah. Yeah. Yeah. No, we

definitely should. Um

Oh my god.

Why is this?

>> You got it?

No,

I'm dealing with I'm dealing with crypto

UX right now.

>> Uh,

>> well, let me tell you, we don't have

much time. I mean, we were supposed to

end

>> minutes, so you got to hurry.

>> Yeah. Yeah, I'm I'm working. Let me see

if anyone

Oh, interesting. The the docs are really

interesting. So, you can actually launch

stable coin from their docs itself. So,

you can just sign in, create account.

Let's actually do that, you know. Let me

sign in. Uh check the prompt. Okay. It

is asking me to do it without iCloud

keychain. Okay.

I might have to create pass keys. I

don't have any.

Use phone or tablet.

Oh,

>> okay.

>> The team sent me some funds. All right.

>> Oh, nice. Okay.

>> Um, should I pass on the screen to you?

>> Yeah. Give me a second to now uninstall

Phantom.

All right. Remove Phantom.

Um

Oh, they minted me path USD. All right.

Uh yes. Yeah, you can can share my

screen again.

>> Yeah, you got to share your screen

first.

>> Okay. Okay.

Uh,

all right. So, they sent

my address. Oh, is it sharing?

>> Yes.

>> All right. They sent my address uh this

path USD. Um,

>> which is token 000. So, I'm just going

to update this vibe coded script in

here.

>> Okay. And once you do that, can you also

give us like a overview like what we are

trying to do here?

>> Yeah, essentially all we're trying to do

is actually send a tempo transaction.

Um, and what we want is ideally, well,

this is going to be with path USD. Um,

we want to send with two transfers in

calls. Um, and use path USDC USD as fee

token. Um because I want to I'm guessing

that fees have to be paid in AlphaUSD.

So if we pay with path USD, we should

see the fee AMM in action.

Um so I think it's trying to kill two

birds with one stone, which is literally

just like how do we see multicol and how

do we see path USD in an example

transaction? Because I think that going

through the explorer, there's no way

we're going to find that. I also gave

them the feedback of like, hey, on on

here for the faucet, you should really

just let people paste an address because

MetaMask does not support transferring

the token. Um, anyway,

>> so or close some of this.

>> So, all I told cursor here was to use VM

and to use the new transaction type. Um,

and you have to load the RPC private

address and private key into here, which

you can just generate by running bun bun

run generate wallet. Um, and

I have low faith that this will work

first try.

Token address value. Let's do value one.

um

call data hex.

>> Try it out.

>> This is literally encoding using the AI.

Um I'm assuming that we're 18 decimals.

So let's just

let's just

break.

>> Let's see.

>> All right.

RPC request failed. Failed to decode

sign transaction.

It failed.

>> Send params.

All right. There's a few ways around

this. One of these is

what happens if we just use ETH call. Do

you guys support calls yet in here?

>> Ah, this is still legacy. I think

>> if I look here, I can't input calls

here,

right? This is still the normal. This

assumes that the trans this new

transaction type doesn't exist yet.

>> Yesum.

>> Which makes sense because again this is

like when you have this new type there's

like thousands of interfaces that now

need to be updated to to support it. So

I can't use this to generate the

transaction. Um, they should actually,

it would be very nice if Tempo created

like a simple template transaction UI

that let you build a transaction, see

what the request was and send it, right?

Like imagine you had that like left side

is the template transaction, right side

is the um

the actual like code using VM because I

think they tweeted something about VM

now has tempo transaction support but

uh use tempo actions.

>> Let's see what are tempo actions. You

can use tempo where what is this VM

actions? You can also use tempo specific

actions. Oh, this is their own SDK that

they rolled up into VM because the the

the Weim or VM team is part of Ithaca,

which is now part of Tempo. Uh, shout

out to to Tom and Jake. So, it looks

like these are SDK things, but this is

all higher level. I'm like, where is my

low where's my [ __ ] submit

transaction,

guys? Come on. Tempo, where is it? Where

is it? Chains. Use tempo actions. That's

not what we want. Um, use VM actions.

Nope. All right. Can't find it there.

Um, cursor could not figure it out. I'll

say try to debug this and keep trying to

send the transaction until it works.

Um, it doesn't actually look like we're

specifying the fee token in here either.

Let me see. How do I if we go back to

the all the way back to the spec fee

token is optional here. So we should be

able to specify fee token.

>> Oh interesting. I mean

>> if it is optional then I mean if you

don't specify then what it is going what

it is going to use. It must default then

to um whatever the validator is

validator addresses but then it's like

>> I haven't looked enough at the validator

because I remember each validator can

specify what token they want fees in.

>> I see. I see. And then there is the

>> sorry I don't know if they're expecting

the wallets to be able to check and know

like oh this is what the current

validator's fee token is and set that as

it when it's sent or if it's Yeah.

Actually, I don't

I don't know cuz there has to be some

default.

>> There must be some default your wallet

uses when it sends the transaction and

this isn't specified. Um

>> yeah, exactly. Cuz I know if you choose

a token and if the validator doesn't

have that token, then using the native

PMM, it does the swap like based on what

the validator is using.

>> Yeah. Uh,

>> but it's like how does it like how does

it even figure out what it's running?

All right. Um,

>> valid question. I'm taking the notes.

>> It's also using I also realized I was

using ETH call earlier not

um ETH sign transaction.

So maybe this is actually

Nope, I lied. Uh sign transaction.

All right, I think we have to actually

read this access list

encode transaction payload. Let me just

see say

uh see if you can just sign this the

normal VM way, not with the tempo

specific stuff.

So I I think the error is partially

coming from

how this is being signed or it thinks

that it has to sign it a different way.

>> Create tempo transaction hash.

Oh wait no we were already using the

right signing, weren't we? Signature

accounts sign.

All right. Um this is kind of classic.

It's too wordy. So, I'm going to say,

hey, function and do it in line.

Um,

it looks like it's trying to do a bunch

of other stuff as well.

All right. This might be one where we

need their help, but it's failing

because it's failing to decode from the

RPC, which means somewhere along the

line, I'm guessing in this create tempo

payload,

I think it's trying to do way too much

because we have this new transaction

type

and it's trying to create and pass all

these new things.

I also see is it any um

okay

this is something we're going to have to

do another session. I'm going to have to

ask their team about this. Uh but one

thing I do want to just see before we go

is we do have it's not a multicol but

this one does have a single call and I

am curious how the traces look. Um, just

going to close a bunch of these other

tabs for now.

Um, so if I go debug trace transaction,

this is how you get stuff like uh and

I'll show off herd for a second. Um,

when you do like call deploy token on

clanker, there's all these traces for

everything that happens after it's

deployed or after this first call

happens because contracts call other

contracts. This is not exposed on their

explorer. There's just events. Oh,

actually, no. They do have a trace. I

lied. Um,

it looks like it's call.

Wait, sorry. Is this the one I was

looking at before? No, this is minting.

I want to see

the crate token. This is call's trace.

Why is Why is call coming from the null

address? That is worrying me.

Let's see what happens if I put in this

transaction hash. I'm looking um

I'm looking at the decoded trace of this

crate token and it looks like the first

call is to the null address before it

calls the tip factory.

So, I'm guessing that

>> this is somewhat similar to 7702 where

you call your own address in 7702.

Technically, if you look at the trace,

you're calling your own address and then

that does a delegate call, I believe, to

everything else,

>> but it looks like for them it's doing a

call to the zero address. Anyways, we

can verify this by going to this debug

tra transaction

and I see it's failed. Why is it failed?

Failed. True. Okay. Very very helpful.

Uh

any any ideas here?

Let's see. I did copy the right

transaction, right? Yeah.

Like this. If I do this, this works,

right?

Yeah. So, I can get this. But if I go to

debug trace transaction with the same

thing,

it's failed. I'm guessing maybe that's

just not

supported yet. Maybe we can replay the

transaction. Maybe this works. Oh, okay.

All right. So, replaying works. Debug

trace does not work yet. Um, again, this

is a classic like it's hard to support

all the RPCs. Um,

those are both pretty popular tracing

endpoints. Um, so for anyone who's

curious, it's now good to know that

>> that doesn't work. Um, so I can see

>> what endpoint was it again?

>> Debug debug trace transaction.

>> Debug trace. Yeah, I'm going to check

like I'm taking notes by the way

everyone. I'm just going to drop the

notes of con like what worked, what

didn't because we actually here to like

test.

>> So the first call in the trace.

>> Oh, this actually looks different. VM

trace null. Oh, that's bec. Oh god, I

hate this. Um,

debug trace returns this trace object

that's like your trace, your calls, and

then within each calls the traces. I

think replay transaction. Yeah, it

already basically unnests it. So you

have trace address where each depth is a

child call. So yeah, it is calling to

two with zero gas its own call type. The

from is zero as well. Oh, the from isn't

even the wallet. Um I would have

expected that the from here was the

wallet address. Um but it's not

>> interesting.

>> Which now makes it harder.

>> But what exactly is happening here? Like

why is this called from zero address to

zero address? It's a it's a good that's

a good question. So what happened here

is I'm calling it from my E7 FFF, right?

Um and typically for traces um like if I

go to this this trace on herd, it's the

this caller this from calling this

contract. And so when this contract is

called, it can only call one other thing

at a time, right? Um versus when I have

many calls like an execute technically

>> like

there becomes like multiple top level

traces. There become multiple trace

zeros u which breaks the model of how

traces work in um

in Ethereum because Ethereum has always

assumed you call a contract any other

calls are a child of that first call.

You see trace address here is zero

right? So like if I if I shrink this to

call depth one,

>> you can see there's a zero trace here.

Um and then everything else is like 001

02 03 04. These are all child traces.

>> You can't have multiple zero traces. Um

where like it's like hey yeah this

wallet called five contracts um that

kind of breaks

>> all of the data engineering. Um

>> and so to get around that they've added

uh this zero this null call um at the

empty trace which basically makes it so

anyone who's done doing typical data

engineering is going to be like oh the

null address is now calling this address

right because logically the way traces

work today it's if there's a child trace

this is being called by this uh which is

incorrect. So anyone who's building on

top of tempo has to ignore

and not ignore but has to kind of switch

around

how they're showing this. Um

>> interesting.

So what does this breaks? Like if there

are any dashboards right now are they

going to break?

>> Yeah. So anything that's like tracking

like who who's actually calling this

contract? Um

>> it depends on how they're joining and

representing the data. U but this this

from being correct is is good. Uh so I

think for the most part we'll be able to

just ignore this. This just kind of

breaks um anyone who's using this first

trace as like what was the contract

called? Um this is now technically wrong

where it's like oh it's you're you're

going to show that the zero address was

called. um which like this technically

does not match what's in here where the

first call is to here, right? So it's

kind of two changes. One is like that

two input value is now gone because it's

nested within calls and then that first

call doesn't actually match this first

two here, right? So you have to have now

this custom mapping of like oh uh if I'm

joining traces on transactions per se I

have to join on the second subtrace

so that like if there were more multiple

calls there'd be like trace address zero

trace address one trace address two um I

would have to join that to calls one

calls two calls three and ignore this

top level call here um So

it's a slight change. It's it's hard to

say exactly what it breaks per se. Um

but both of these are well this is a

very big change because this change is

like anyone who is decoding transactions

before and assume that each transaction

makes only one call. Uh now that breaks

because you have to unnst this array

basically which was like a big Salana

problem which is unnsting as soon as you

have to unnst or basically expand an

array to flatten it. Uh that makes every

single SQL query

10 times more expensive. Um so that's

going to be a big impact on teams like

Dune Alium etc. U because if you keep it

as a single single row of calls when you

join it then you have to unnst this. All

right. And unsting with a join is also

expensive. U and yeah, for anyone

working with traces data, this just

makes things a little more complex. Um I

don't know. Do do you know if Tenderly

supports

um tempo test net yet? I don't think it

does.

>> I don't think so. No, I thought it

>> Yeah, it'd be interesting to see.

No, I don't think they support it. No,

they don't. No,

>> it would be interesting to see it. Um,

but yeah, this would be a better example

if I could send this transaction that

had the multiple calls in it and we

could also see the fee being swapped

because I'm also very curious of like

for that fee swap is that bundled into

calls or does that actually happen

outside of calls? Right? U because at

some point if you're swapping it has to

call that contract, right? So then it's

like what happens if

>> like the fee swap

does not actually show up in calls but

it shows up in traces, right? And now

it's like oh who's actually calling a

swap contract? Well, you can't get it

from here. You have to get it from here,

right? Um, and so yeah, there's a lot of

nuances here about like standardizing

across the transactions RPC, the traces

RPC. Um, and especially once you

introduce pre-ompiles, um, then like

technically you can't trace within the

pre-ompile. So like if for example, EMM

was doing a multihop or like or the path

decks or the stable decks was doing like

a multihop swap, you might not be able

to see which paths it's routing through,

right? And that's important data for

people. Um,

but you can't see it if it's in a

pre-ompile, but they're saying that it's

in um

actual contracts and not pre-ompiles.

So, that's also something we'll have to

look into more. But, a lot to dive into.

You know, this is just this is classic

kind of like the dev X for anytime um

you know, anyone's trying to learn how

to build on a new chain, especially when

so many things change. So, you know,

>> this is it's never it's never it's never

pretty. Yeah, it's never pretty, but

this is always good for people to start

getting familiar with like ah these are

some of the changes, these are some of

the integrations like MetaMask not being

able to transfer or find that token, you

know, uh things that need to be

>> Why was that happening again? Why was

MetaMask not able to transfer the

>> I don't know it because it requires me

to add like search the asset before I

can choose to send it. Um and it just

couldn't find that asset. Um so then it

just didn't let me send it, right? I

would expect I mean I guess for safety

reasons they don't let you just paste

any address in there but it's like they

should be able to be like hey you've

pasted an address check if it has like

the typical ERC20 like functions on it

and let you transfer. Um I don't know

it's very weird that the wallet could

not detect the token. Um but anyways

yeah stuff like that stuff like the

faucet should be allow you to paste an

address. stuff like the fact that cursor

can't just oneshot um sending of uh uh

this new transaction type with VM

because the actions are abstracted with

tempo actions, but I need I need support

on the the send calls. Um so, you know,

a lot of little things. I'm sure the

Tempo team will hopefully go through

this at 3x speed and and make some

improvements, but um yeah, we

>> I'm probably taking notes. Like I'm

probably taking notes. I'm going to tag

the tempo team on Twitter that hey, this

is the

>> Yeah, these are things we ran into.

Yeah.

>> So, the feedback is Metamas token. It

didn't direct the didn't like

>> going through Yeah. going through the

faucet and stuff is like the connected

wallets maybe don't all support

transferring the token, so you should

let them paste.

>> Yes. Paste the wallet. And then another

one was of course cursor wasn't able to

do a tempo transaction in one shot. Is

that is that correct?

>> Yeah.

>> Um and then another one was of course

the tempo transaction UI. So like we

would love something like you know you

can see we can send request and you know

left side there's tempo transaction and

then kind of like a sandbox.

>> Yeah.

>> Yeah. And few sorry

>> oh no I just realized the tempo docs do

let you fund another wallet like by

pasting. It's just hidden at the bottom

of the screen.

>> Jeez.

>> So in the So here I was looking where is

it? I was looking at this faucet and I

had to scroll down to here

>> and they're sending of funds to another

address. U

>> Oh man, we wasted like 15 minutes.

>> Yeah. Yeah. Yeah. Yeah. Yeah. But like

classic doc stuff. If you don't put it

up front, no one's going to [ __ ]

scroll to it. So

>> Well, fair. Okay. That's sad.

>> Yeah.

out. Okay. And then uh another one was

the swap. So when you did the

transaction, you weren't able to see the

swap transaction, right?

>> When I when I was looking at the the RPC

data for or what what I was talking

about?

>> Yes. For the RPC data when you were

looking

to

>> Yeah. I mean, there wasn't I think the

ones that that the transactions we were

looking at were all paid with AlphaUSD,

so there wasn't a swap in there. I don't

know. I haven't we we'd have to test out

I think ourselves putting a different

fee token to see if it see how the swap

works but we couldn't get that far

because we couldn't figure out how to

actually encode the transaction

correctly. Uh

>> okay

>> that's fair.

>> Yeah

>> I think it was it was a good shot. Um

and also the cursor couldn't do it in

one shot is like a great feedback

because of course like the docs who goes

Yeah. Exactly. Yeah.

>> Um and then yeah RPC wise the replay

trans the debug trace transaction didn't

work even though replay did. Um so

that's just

>> did right. Replay worked.

>> Replay did. Yes. Um but a lot of teams I

think still use debug

>> trace. We do as we do as well actually.

Um

>> yeah, I will have to check with the team

if the debug trace method is supported

or not.

>> Yeah. Yeah. Yeah. Um

>> okay. Yeah. I think that is it. We

covered almost everything and we do have

a question regarding the default fee

like if we don't set up maybe we can try

on our own or we can just ask the team

>> for what?

>> For the default fee. So for example, if

you are not putting anything

>> what how does it figure out what to pay

with? Yeah. I don't know.

>> Yes. Yes.

>> Um and then

we can add this in later of um I'll send

this to you. I I just published the

repo. Obviously it doesn't work yet.

We're getting help, but we'll we'll make

it work. Um I put in the chat. It's just

tempo test

on my on my GitHub. So, uh, as we fix

it, hopefully people can refer to it and

test with it.

>> Yep. We should do, I think, another one

and like part two.

>> Yeah, we'll do a part two. We'll get

we'll get more of the tempo team's

actual support and then um yeah,

hopefully get one of them to join us

next time so that we're not in the dark

as much.

>> Oh, nice. Okay, almost like 900 people

now. Actually, 850 people are watching,

which is massive. I wasn't expecting.

>> Yeah. I mean, I think everyone's very

excited about it, but also very like,

you know, I want to say worried, but

it's it's it's hard it's hard to learn

these new things, especially when you're

used to your scripts and tools and apps

working a certain way. Um, it's it's

very easy for devs uh and users to churn

when it's suddenly like, oh, I just

don't understand this or I can't get it

to work. Um, so you know, Utam and I

will keep trying to do the leg work to

>> to make the stuff easier and I'm sure

that you know they will have their own

Tempo should have their own devs as

well. Uh, so

>> actually they don't. Yeah

to like join us in the next stream or

probably Georgius. Let's see if I can

get them on board. But okay, I think uh

we are at the end of our stream. So

thank you for joining. Hope this was

helpful and we might do another part two

uh during I don't know holiday. So if

you're logged in we are also logged in

and we are probably coming to you again

and if you're building anything on tempo

alme go to elme.com we have support for

tempo you can start building there if

you need any help you can ping me and

also check out Andrew's profile and also

check out herd which is a really great

tool.

>> Great.

No, that's that that's it. This was this

was this was fun. Thank you, Tom. Thanks

everyone for watching.

>> Thank you everyone. Bye. Take care.