---
title: "HOW TO BUILD A CAT | Twitter Space Recording | AlphaGrowth"
description: "Twitter Space Recording from 2024-01-23 with saudiverse 🇸🇦, Quantum Cats, Rijndael, Dan Held 🧙‍♂️, Eric Wall, Casey, huuep.eth (🍌,🍌), Tyler Whittle, Dev Aloft"
url: "https://alphagrowth.io/spaces/how-to-build-a-cat"
---

# HOW TO BUILD A CAT

Recorded: Jan 23, 2024 Duration: 1:23:52

## Space Recording

[Open Original Space](https://twitter.com/i/spaces/1OdKrjmnlEzKX)

0:00 1:23:52

### Full Transcription

Oh

Oh

Oh

Oh

Oh

Oh

Jimmy ow, I love this song. I

Sometimes I sing this in the shower. I don't know if I should be telling you this

No, I wanted to know that well, you know, we're all about the truth not not just the easy truth, but

Sometimes the hard truth to

Pleasure to have everyone here

We've got

We've got the rainbow wizard Eric wall

I never know if we I never know if you ever will unmute themselves and say nothing you something

You never know with this guy. We've got the one and only Casey order more

A lot of accomplishments for Casey. I think he has a Bitcoin meetup or something really nice guy

Danny

from on chain monkey

Rindell another Bitcoiner Dan held other Bitcoiner. Everyone here is an unaccomplished Bitcoiner on this space

We're gonna talk about how we're gonna attack Bitcoin

You know, we heard about Bitcoin two minutes ago. All of us meant this is actually I'm looking at this stage right now

Everyone on stage right now is a Bitcoin boomer

Who's who's been dealing with Bitcoin way too many years way too many way more than would be healthy or recommended and

Yeah, we're here to attack Bitcoin. I think there's a bunch of like cool

I'm like here is for entertainment basically the other people on stage are smart

So I'm gonna try to like shut up soon

but

we're gonna talk about I

Think there are two cool things that I want to talk about one is like, you know evolving inscriptions and presizing transactions to to to make

Inscriptions change over time

Which is something we leverage with quantum cats, but also Danny

did with some of the OCM collections and and

And and and also maybe we can compare that to stuff like

The sad endpoint which is different and maybe we can talk about the difference there

But also I want to talk about op cat and how it's an attack on Bitcoin. Maybe

Maybe we'll start there Casey. I don't know if you remember I think we first met in

Stanford

late last year and

And

And I don't know we were at the time we didn't tell you we were already scheming

About breaking Bitcoin with op cat and I think we asked you about op cat briefly and and and I don't remember what you said

I don't do you remember that occurrence? I

Do I do I remember I was you know standing across the room and I noticed, you know, very very handsome Bitcoin developer

So I wanted to talk to him and then you were standing next to him. So that's that's how we met

That sounds very accurate

Yeah, I'm I'm I'm definitely not here to be either the developer or the handsome person. So yeah

But yeah, no, I totally remember did we talk about op cat at the time

Very briefly I the reason I asked you was because I don't remember what I said and I don't remember what you said

So no, it's a blur. It's a blur to me too, but you guys have been scheming for a while

I'm always impressed. You guys managed to like keep the lid on things and then

doing

really cool stuff

Thank you, sir. Thank you, sir. No, it means a lot

so, okay, we'll be back to attacking Bitcoin with op cat soon, but Randall, do you want to like maybe

Hit us off with a bit about like what?

Evolving descriptions are and what other ways there are to do this stuff and let's go from there

Yeah, sure, so

And you know, I know that Danny and I have talked about this on other spaces before

I don't think I've talked publicly about it with Casey, but I think we traded some DMS at some point. So

What what we decided to do for the quantum cats collection. We had this idea a couple months ago

it's funny the

That the original prototype for this was just like a giant pile of shell scripts

But but what we decided to do was we thought it'd be really cool if

you know so inscriptions have this recursion feature where if you have like an HTML or a

Like an SVG inscription

You can refer to the contents of another inscription by using the slash content endpoint on the

unlike the word server and

You know people have been using this for a little while to do things like you

Inscribe all the traits for some

You know

Inscription at one time and then you inscribe a whole bunch of inscriptions that like stitch them together

So you might have you know a hat a head and a body in different inscriptions and then you you stitch them together

using recursion and the idea that we had was you know

What if you had like a PFP that referred to inscriptions that weren't actually on chain yet?

And when you go to view them, it would you know 404, but you could quietly handle that in the background and then

You know at some point you reveal those pre signs but unbroadcasted

inscriptions and then you know the request

can complete successfully and the trait just sort of appears and like the the user experience of this is like and

Like the the very first prototype of this that I showed to Udi

I had a very crappily drawn like stick figure and then I mine a block and all of a sudden

He has a wizard hat and then I mine another block and now he's holding like a potion and then I mine another block

And he's wearing like a wizard robe and we thought that it would be a cool way to have

Inscriptions sort of evolve over time

And as we started kind of gaming out

Like all of the concepts behind quantum cats

The other things that we really liked about that were that you know in in a covenant system

whether it's cat or CTV

generally how that's gonna work is

You have to ahead of time like pre-compute the future states of your Bitcoin smart contract

And commit to those transaction hashes in your script

And then as time goes on and you're like playing forward this this like smart contract

Whether it's a vault or a coin pool or something else

You're sort of like revealing these transactions on chain that you pre-computed and we really liked that

You know the method that we would have for revealing more traits of these cats over time

Had like this very

Similar mechanism to how you would actually use something like op cat to build more interesting smart contracts on top of Bitcoin

so we had this like nice sort of artistic symmetry between the technical implementation and

Like the the the concept for the collection

And you know like there's there's other ways that you could build sort of evolving inscriptions

like we paid a lot of attention to the Saturn point decided for

Sort of technical and artistic reasons that that we wanted to do with pre-signed transactions

Which you know led to a whole bunch of?

Fun

Implementation challenges that we had to deal with that like super happy to talk about on this space

But yeah, that's that's what we ended up doing

Can you talk about the trade-offs of using pre-signed transactions versus the SAT endpoint I'm super curious about that

Yeah, absolutely. So

You know with pre-signed transactions and to

I know Danny and I have talked about the details here. I don't know if other people looked into it

so the way that that the quantum cats work is

Each cat is a really teeny tiny inscription. It's 109 bytes

And so each cat is basically just a script tag that recursively pulls in a single piece of JavaScript

Internally, we call it the dispatcher. That's like what it's called in the code and

the dispatcher is this this blob of JavaScript that

Fills in like a whole HTML page with some style sheets and a whole bunch of JavaScript that then

tries to recursively fetch a bunch of other

Inscriptions each each one of those we call a layer connector and that's a

Little blob of JSON that maps a unique cat ID to a

Piece of art that's been inscribed

So if you look at the cats right now, you'll see that like each one of them has a background image

So there's like a layer connector that says okay cat number 17 should pull in this particular background image

Those images are also encrypted

And so the layer connector bit of JSON also has a decryption key in it. So, you know the kind of

Trace of the recursion here is a cat is basically just a script tag with an ID

That pulls in this thing called the dispatcher the dispatcher pulls in

this thing called a layer connector, which is just like a look up table and a decryption key and then the dispatcher

Pulls in the actual artwork decrypts it using just like browser crypto APIs and renders it to a canvas

So that's that's kind of the flow

So if you if you think of that of like we have these four

Resource types where you have like the cats the dispatcher the layer connectors and the artwork

Everything except for the actual cats had to be pre signed

So, you know all of the artwork we sort of pre signed all of the layer connectors are pre signed

the

Dispatcher is pre signed and then the cats were able to just sort of like inscribe normally and

Like one of the the bits of sort of operational complexity there is

We had to be really really careful around coin control because you know, if we

like let's say that we have a blue background and a black background if we

pre sign the

The the inscription transactions to you know

Either create or reveal both of those backgrounds using the same UT XO

then only one of them is possible to reveal on chain like ever and

you know the cats have a

Inscription ID baked into them for the dispatcher the dispatcher has

Inscription IDs baked into it for all the layer connectors and the layer connectors have inscription IDs baked into them for all of the artwork

so if we ever screw up and

Like use this like assign the same UT XO to multiple

pre signed inscriptions then

You know, we can't ever fix that again

Or if we screw up and like sweep a wallet and spend all those UT XOs then those pre signed transactions are never actually

Revealable and we can't fix that ever again. So that's like the

The the big like downside and challenge operationally with the pre signed transactions

if you have the recursive satin point you have sort of

Like that part is a lot easier because you know you say all right for the background

I'm gonna have all the cats point their background at this sat and then when we decide to reveal a trait

we're just going to use like

Inscription to re-inscribe that set and then the rendering code will look at whatever the last

Inscription on that set is and that's what it'll render

right, so so you you can kind of have like a late binding choice of what the artwork is going to be or

You know, it gives you more flexibility in the future. You're not like locked into it from the start. Does that make sense? I

Think I don't understand the flexibility part because can't you

assign like an arbitrary inscription to a sat so you can in the future you can change the

You can change the

You can like change you can decide differently what you want to sign to the sat whereas with the pre signed transaction

You're sort of like locked into the content. Yeah. No that that's exactly right. So with with the pre signed transaction

We are locked in from the start

with the

sat and point I

Think I know is cutting offers. It's just me. Hey, can you guys hear me? No, I yeah. Yeah you wrote

Yeah, but just like okay. Yeah, so you're exactly right. So with the pre signed transaction

everything has to be completely done before we inscribed any of the cats if we were using the sat-in point then we could sort of

be baking off the artwork as we go and we have more flexibility about sort of what the arts going to be in the future like if

Six months from now, you know

Some new meme is really hot and we say we want to put this meme on quantum cats

We'd be able to do that with the sat-in point. We would not be able to do that with freesign transactions

So you have a lot more future flexibility

Um

We we decided that you know the way some of these cats are going to evolve

we think that they're like we think the evolutions are cool and gonna be surprising and fun and whimsical and

we kind of liked the aesthetics of

when these

Evolutions happen people having to be sure that this was all planned from the start

Like we like that that was something that that we we wanted to really aim for

There might be ways that you could do that with like a commit and reveal scheme with the sat-in point, but we thought it would be

Just sort of like fun and obvious if people can go on mempool that space and see

You know particular out points that were committed two months ago all of a sudden be spent and reveal some trait

That's like really surprising or fun or you know, whatever

Yeah, I think it's I think it's a really nice aesthetic choice. Oh wait, I was gonna ask something. Oh, yeah

How do you deal with?

fees do you have like anchor outputs or like anyone can pay outputs to

Boost the fees of your pre signed transactions because you can't predict the fee environment in the future. Yes

Yeah, so

You know, I thought about adding an anchor output

Late in the process

But that was after I'd gone and built a whole bunch of tooling to do batch parallel CPFP because like

We still like when we reveal a particular trait connector in the future

We're gonna own that inscription. Like we're not like we're not, you know, air dropping it to somebody or something, right?

like we we own it and so

we built some tooling that

You know, we you like the the primitive here is we've got a little function that you

Give it an out point that you want to bump to a particular fee level and it goes and does all the CPFP math and says

All right. Here's how many SATs I need and it constructs a transaction to

you know, make sure that the inscription ends up in the first output and that

you know, the the second input sort of goes all to fees and

we

Built some automation that can build a whole tree of those transactions. So you you get like a single

Deposit address and an amount to pay and it'll split that deposit into a bunch of smaller UTXOs that are all

Right sized to CPFP a bunch of reveal transactions

to whatever fee level you want and so we actually

Got to use this the day that we revealed all the cats

So, you know, like when we first inscribed these what everybody saw was an image that just said Tapper Wizards presents

And then a few days later that you know

Transformed into the blue and red cat and it said Tapper Wizards presents quantum cats

When we actually revealed the current generation of the cats

You know classic Murphy's law like right before I hit the button to broadcast those transactions

Like this huge

atomicals mint popped off and

Spiked the fees and all of our reveal transactions were pre-signed at a particular fee rate and ended up just

Tumbling around and kind of the back of the mempool

And so we got to exercise this CPFP code and fee bump all these reveal transactions up to I think it was like 150

sats per v-byte

You know

I'm really interested in anchor outputs for transactions. Like generally it's something that I'd love to work on

For this collection. We don't need them anymore because we have sort of general-purpose CPFP tooling

But I think anchor outputs would be great for inscriptions. I

See some other hands on stage Danny you've had your hand up for a while then maybe I don't know devil off

I'm sure I'm sure yours were burning when I was talking about anchor outputs. I

Think this how a quantum cast how you guys used

Unsigned or pre-signed transaction is like it's performance art and it's you know

You had the choice of using this out on point or unsigned and you know it you use something

That's a lot harder to do right to show that you know

this is kind of the artistic way we want to do it and it fits in with quantum cats art very well because

Your art is about the op cat right op cat is about having you know predetermined

You know outcomes that you have to already, you know sign ahead of time, right?

So it actually fits I think for a Bitcoin people who already dig into this this type of Bitcoin art

It makes a lot of sense to to talk about you know

It's like quantum cats is not just this, you know cat image that people are excited by it

But it's actually the fact that you use, you know

Basically, it's about op cat and it's about you know, pre-signed transactions and and pointing out that you know

You can do it this way like and I think it goes back to also, you know

I think of or knows itself dorms program, right? That's part on Bitcoin. Also, there's a big performance art

It's whole last year, right? It's been that way

Like this idea of recursive inscriptions, right, you know Casey talked about that in 2022, right in 2022

When it came out, you know for us when we used the pre-signed transactions

We actually inscribed art collection. This is the dimensions those same dimensions that we did in February

So it's only almost a year now and that was basically

Before recursion was a thing, right? It was an idea, but it wasn't a thing yet

and so we inscribed it that basically that the 3js code that

Needed and also we can strike F late a decompression code that needed to actually

uncompressed to compress libraries because we've had to keep it small and so when recursion was available

I that was when we could then send the pre-signed transactions and that's so back in this was June

I think last year when we did the pre-signed front end as right as you were saying, right?

It's very stressful to hand these pre-signed transactions and and know if it'll work or not, right?

So for our case here, we didn't have a choice of the satin point because there wasn't available yet

So it was the only thing we could actually do

In terms of you know, we didn't have the trail but one of the benefits of pre-signed transaction

This it's a Bitcoin thing, right? So you have one less degree of

Uncertainty and in the or no protocol, right? You don't have to go into an or no

No to figure out that this is a satin point, right?

Like the Bitcoin though it knows about the transaction already. So so that's also a kind of a way of

People might choose to use pre-signed transactions because it is one less I guess degree above above Bitcoin as or no

It's kind of like closer to the metal right like I think maybe life of FIFO or maybe you said something like that

like it's it feels a little bit closer to the metal than

Then the satin point

Yeah, and also use a little more data

So I guess the satin point is just a few bytes smaller

So if you really want to optimize, you know for your inscription size, the satin point is actually better there

So so yeah, there are different trade-offs

Yeah, and I mean to what you're saying about performance art that I'm gonna keep like Eric Wall said this thing

internally when we were

working on some kind of other

Fun little stuff inside quantum cats

he said like this is skateboard tricks on Bitcoin and I think that really kind of

encapsulates like one of the things that I love so much about ordinals and inscriptions like I feel like

people figuring out

Cool ways of you know using Bitcoin and ordinals features together to do stuff

That's like brand new it feels like you're doing skateboard tricks on Bitcoin and like that's just a lot of fun

The other trick that you guys did I was really cool was adding the encryption to the pre-sign

So you actually combined so I don't you'll realize that so they did pre-sign and added

Encryption or their your decryption when it was revealed. So that's actually a really cool dynamic

You know what we did was we did the pre-sign adding decompression. So we had things compressed and then the when it's you know

revealed it's also

It decompresses the compressed data, but you know people can actually see the compressed data

But now we don't encrypt it. You can't even see it. So it's actually it's a cooler reveal

Yeah, that was actually really interesting. So that that was actually a

later

Decision and feature like in the project. So what happened there the original plan was

You know, we've got this like kind of

So so originally how it worked

So the way that I described this right is each cat points at this dispatcher layer and then a dispatcher points at this like trait

Connector layer and then the trait connector points at the artwork

Originally that that layer of indirection with the connector actually wasn't there

So originally how it worked was each cat actually referred directly to the artwork that was going to be inscribed

That was pre-signed and so we were going to inscribe all the cats and each cat would recursively pull in, you know

It's background its body its eyes like whatever whatever it has and we would you know?

Reveal the artwork as the cats would evolve and somebody on the team pointed out like even if you don't know

What the actual artwork is you can look at?

What are the inscription IDs that the cats are pulling in and start clustering them and saying alright for layer X like this?

inscription is pulled in by

More of the collection than that one

so this one's gonna be more rare and

you know people are gonna start like trading and speculating on these things before we've revealed the artwork and what we

Actually wanted to do is we we wanted people to be surprised as the cats evolved and we we didn't want people

Like like we didn't want to have information leaking about you know

What what cats were gonna be before they evolved into that so we added this layer of indirection where?

you know if you look at the cat all of the

Like every cat pulls in the same inscriptions directly

But then they also have a unique ID and so when they pull in this like layer connector

They use their unique ID to D reference like what's the actual piece of art?

So until we reveal this this sort of lookup map in the middle

You don't know if your cat is more or less rare than another cat for like a particular layer

So we went and built that out tested that a couple hundred times did a couple mainnet tests of that

We're really happy with it

and then we got into this like crazy period where fees were you know between like 200 and 500 stats per v-byte for a

couple weeks and

It was like okay. Well if fees

are gonna be high like this over the course of

the time period when we want to do evolutions like these are gonna be really expensive and

You know the way that the pre-signed inscriptions work is we have to commit to the fee rate at the time that we sort of bake the

Whole collection, and then we can CPFP them later on

But what we'd really love to be able to do is if there's a dip in fee rates

We'd love to be able to just put all the art on chain if we can

because you know if you look at the footprint of the artwork versus these lookup layers versus the JavaScript versus the HTML like the

The actual artwork was like 90% of the of the size right of the total size and so

You know we decided to add this

Encryption mechanism so that you know when fees got low we could

opportunistically

Publish all of the artwork nobody would be able to see it and then you know when we need to actually

Reveal an evolution the only thing that we're broadcasting is this sort of lookup table

Which is just a blob of JSON and a decryption key and so our sort of like per

Evolution on chain footprint like at the time that we have to evolve them is is minimized that way

So it's funny like a lot of people are really

Happy about the the encryption method like I've heard from other projects. They're like that's cool. We want to do that

But it's it's funny. It was actually like a fee management technique that we that we thought of

Yeah, that's what we do with karma to that we're gonna release soon we encrypt to some of the images so that

And fees are low now, so everyone else who was you know, thinking of it ascribing now is a good time to ascribe your stuff and encrypt it

Yeah, um it dev alofta came up on stage, and you've had your hand up for a little while. What's up?

Hey, how's it going? So great to hear from you. I I had a technical question

Yeah, so specifically regarding the CPFP mechanism

You described a mechanism that is very similar to what I do, which is if you have a bunch of revelation transactions

And you want to boost them, you know, whether or not you use anchor outputs. I do a very similar thing where I calculate

how much

Total fees you will need in order to spend from all of those revelation transactions such that the resulting fee of the package will exactly

boost everything to your

Boost everything to your to your desired fee rate

but one question that I had and something that i've been thinking about for a while is

Because you first calculate that and that seems very similar to my technique and you then fund that boost

If you wanted to then rbf that CPFP transaction later

There are a bunch of eviction rules that are a little tricky. So obviously the easy and obvious ones are

The total fee rate and the total fee has to be higher

And then you also have to have an incremental fee. That is the replacing or the evicting transaction

However, there's also this additional requirement saying that you cannot add new unspent utxos

So you have to use transaction outputs that have previously been spent but are not currently confirmed

If you are spending an output that already exists it likely will not

Facultate the necessary fee rate that you want to bump your CPFP package fee rate, too

so I was thinking about

Creating calculating the new input size that I need having users send money there

spend from there with a fairly low fee rate

and just send everything or most of the money to a

temporary output just to make sure that that utxo is no longer utxo but is spent and

And so doing enable it to then be immediately included in the actual CPFP

transaction replacement, but it seems like a bit of an awkward technique and might also

possibly not work with

Relay properly if you replace it too quickly and maybe some of the other bitcoin nodes in the network only see the replacing transaction

They might possibly reject it. So i've been thinking about either delaying or possibly

Waiting until we have proper package relay in place

And I also have been thinking about whether you want to just keep replacing and then refunding the user the original and now

unused utxo if you just want to keep adding more utxos on and on to the CPFP transaction, so

Just curious if you are doing anything like that and what your thoughts have been for rbfing CPFP boosts

Yeah, no, it's it's something. Um, I I don't have a super satisfying answer for uh, I was going down the path of

um

more or less not not not doing the sort of like, uh

I don't know poor man's ephemeral anchor is is maybe like the the the back of the napkin way that I would describe the technique

that you described but like, you know something like that where where we

um, sort of do like an inverse pin of of the of the

Point that we want to use to to do the rbf

um, I I was describing

a technique kind of like that to greg sanders who's you know doing a lot of work on like package relay and

uh

like he's he's my sherpa for for kind of like ephemeral anchor stuff and um,

you know

It got really complicated really really fast

um, so the

Current solution that we have which you know ended up kind of being a

Uh, so so this this whole uh encrypted traits thing ended up being a blessing in more ways than one

so one of them is

You know another relay policy that folks might not be aware of if you haven't uh had to fight with this

Is the total size of an unconfirmed package can't be more than 101 kilobytes

And so if you're trying to like repeatedly cpfp

artwork like large assets

Then um something that we ran into in testing was like yeah, I can do that, you know once or twice

But then I start running into like package relay limits

so

The fact that now all of our artwork is already on chain and the thing that we need to bump

Are json files that are sized on the order of like, you know, a couple tens of kilobytes

Um means that we we have a little bit more runway to just do

um, uh, like

two or three successive cpfp's

Um our strategy for the one time that we've had to do this in prod so far was you know, um

I'm gonna make up the numbers a little bit but like at the time

the going fee rate for the next block was like

110 sets per v-byte and we just went ahead and boosted up to 150

Right and we were like, you know, if we if we have to cpfp this again, like we we can but

Like realistically, it'll be okay. Like it'll it'll probably be fine. Um, you know, one of the things that

Like i'm really interested in working out

Is reliable rbf for that cpfp

Both for future projects and also like I would love to take some of this tooling that we've built and try to spin it out into

You know other tools or services or whatever for like the community use because I think um, just generally

Uh, like fee bumping of inscriptions has tended to be a foot gun

Like i've seen several inscriptions go to valhalla because people I think we lost them again

Going to sparrow and oh, can you guys hear me?

Am I here I I can't hear irondell, okay deviloft. Can you not hear him?

Can you hear me I can hear you deviloft and I can also hear irondell

Can you hear irondell? I I can hear myself. I can't hear deviloft

Um, I don't know casey or danny. Can you guys give me like a thumbs up if you can hear me?

Yeah, danny dan can deviloft can casey cool. I think deviloft maybe if you

Uh drop down and then request to speak again. It'll let you let you back in that should fix it

Yeah, well, we'll definitely bring you back up. I think like two of the things one of the things that was definitely working in our favor

here is that um

We own for the things that we need to fee bump we own the eventual inscriptions

So like if we were trying to inscribe something to

somebody's wallet

right, then

Then we have different options for fee bumping. But because we're both the sender and receiver for these things

We're able to use cpfp really effectively

Um casey another thing that I thought you might be interested in I need to finish getting a pr together for ord

um a feature that I I hacked in and it was sort of like

Uh violently untested for a little while but now has been very well tested

is um

when you uh

Presign or sign an inscription

I had the ord and scribe commands spit out

The um sat point that was used to do the inscription

And so then what we ended up doing was keeping track of

you know, so

uh

When we were pre-signing everything we would use like the sat point flag to say this is the

Uh the the utxo basically that I want you to use for this inscription

And then we would verify

That the command was like using the same one like from the command output and then we would actually

deserialize the um commit transaction and make sure that it only was spending its assigned utxo

All that goes into a database and then sort of at every step in the process before

Any keys get touched before anything gets inscribed

There's a whole series of sanity checks that runs that goes through like every pre-signed commit transaction

Make sure that it's only spending it's assigned utxo

Make sure that no transactions spend the same utxo

um

deserializes all the revealed transactions and make sure that what they're spending is actually the

Uh commit transaction output like just going through and doing all that sanity checking like every step along the way

Was actually um a decent amount of all the code that we wrote just to make sure that you know

When we go to do some future evolution, we don't all of a sudden just you know discover that we can't

Oh, and we lost casey

Nice

He was things happened he was like this is so boring i'm gonna get off the stage i'm just kidding

Yeah, maybe maybe uh eric was eric was so oh, I think he's back i'm back

I don't know. I i'm not sure I under I totally make sense to save the sad point or the yeah

The sad point output is the is the is the pr just to add the sad point output to the inscribed command

So, um, there's there's a few things and I need

To break them up into like multiple prs so that you know, uh

Raph and and anybody else who wants to review them can decide if you know

All of them want to go in or if none of them want to go in or if they want to go in piecemeal

But the stuff that I that I needed to add so one of them was exactly that like adding

the sad point output to

The the you know output struct that comes out of the inscribed command another one was um

I added a pre-signed flag to inscribe so that it would you know generate and sign the

commit and reveal transactions but not actually broadcast them and

Spit those out within, you know new fields and the output and then the third one. Um,

so

you know

in order to like the the workflow for this

was um

So we've got all this art that like needs to go on chain and all it has to be pre-signed

So I added a little uh estimate flag

And what the estimate flag is is you say, you know wallet inscribe dash dash estimate

You give a fee rate and you point it at a file

and it'll um go through and tell you

What the total fees will be to inscribe that it's it's like the same

Code path that that spits out the fees when you do a normal inscribe. It just doesn't actually inscribe anything

and um in order to make that work because like the way org normally works

Is you know, it it goes and like finds a utxo in your wallet and inscribes it

So I punch like a fake utxo into the wallet that has 21 million bitcoin in it

And then it basically does like a dry run of of uh inscribe and so

You know, we've got all this artwork

So we go through and say estimate on all of the artwork and get the total cost

for uh inscribing everything at a particular fee rate and then we build up a tree of

utxo splitting transactions

So what we end up with is like

Put this many bitcoin into this address and then the automation will split it

Into one utxo that's perfectly sized for each piece of artwork

So our wallet has, you know as many utxos at the correct value sizes for all the artwork that we need to presign

And then we just go through and we just line up like this utxo goes to this piece of artwork

We're going to presign it and we're going to mark that as used in our database and kind of like move on

So, you know estimating, uh cost, um spitting out the sap point and then um

Uh, uh pre-signing were the big things that

That we needed to add that'll end up getting upstream. Yeah, those all

Those those all sound like awesome additions. Those all sound great

Eric was telling me before

that uh

And i'm feeling the same way is it for at least for me as a cold victorian maxi cold survivor

I feel like i'm learning more about bitcoin

Into space than probably in the last 10 years

Eric is this uh, are you experiencing the same?

Yeah, I was uh, I was just casually listening in and I don't think that i've ever

Uh been in any bitcoin spaces that are as deeply technical as the ordinal

Uh bitcoin space is where you talk about

uh

pfping

Transactions in in this delicate way and like really understand like a deep technical understanding for how bitcoin works is required

when

There's so much at stake at stake as it can be for people that are issuing ordinals

so I think that uh

Like it's it's really adding to the body of knowledge of bitcoin

And I don't think that bitcoin has been like stress tested not only in like the fee mechanism

like the the fee stress testing, but

The tools that are available. I see randal occurring like running into

Speed bumps with bitcoin core all the time because he's using it in a way that the

Nobody's probably has ever used it in before

uh, so I think that you know for all the

bitcoin maxim lists that are like think that they are

Uh, like so expert on bitcoin

I'd be surprised if if they knew like

Two percent of what someone like danny has learned about bitcoin in the last year

Yeah, it turns out when you learn use bitcoin in super crazy ways you have to know a lot of super crazy things

Casey one of the things I ran into i'm sure danny's run into this too. I wouldn't be surprised if if you

had not run into this also is um

You know bitcoin core

when you add

A couple thousand utxos to the wallet it takes a while to rescan and so if you have automation around core

That's either doing lots of uh, like just lots of sends and receives or is for example doing batch inscribing

um

It causes the bitcoin core rpc interface to just hang so like there's there's like code

In our software to say like, all right. I just broadcast this uh, this transaction

Um, if I get any of these errors back from the rpc interface

Then like it's fine because it's just bitcoin core crapping out and we're gonna

You know back off and retry after a little bit if we get these other error codes

Then it actually means that we screwed something up. Um, and and it's it's been really funny because it's like

Um, you know i've i've done a decent amount of bitcoin development

With sort of more traditional use cases and it's always sort of like yeah, you know, you've got some automation to

Send and receive, you know a number of transactions that you can count on one hand and um

doing something where you all of a sudden have like

Thousands of new utxos and your wallet wants to go back and like rescan the last you know 120 blocks or something

In order to like look for new stuff is is not

The normal case for uh for core, which is really funny

Yeah, i've i've never run into that because i don't make any descriptions

In 2013 because I started a bitcoin exchange back then so we had

My core wallet issues. So we basically, you know distributed that back then

That's really funny, um

Yeah, maybe you nerds like found out cool things about bitcoin and stuff, but

Myself as a former member of the bitcoin laser eye

maximalist cult

Uh have very thorough knowledge about picking studio mics and recording equipment

And I know how to sign up for a youtube account for a new podcast and those are things that you guys cannot do

So well case you can i don't know if you've seen uh the whole money podcast

It's my favorite bitcoin podcast and uh, they've got great av equipment also

Bro, thanks for the thanks for the plug. Aaron was actually in my dns being like you must chill the podcast. But uh, yeah

Thanks for oh

I'll chill okay. We're gonna do a quick hell money spot here

So, uh hell money podcast is the best bitcoin podcast because it's not actually a bitcoin podcast

It's actually an astrology podcast

But erin fucked up and invited casey every time and somehow he ends up talking about ordinals the entire time

and uh, erin, you know just wants to talk about astrology, but the result of that is that it's like

a fun accessible

Podcast that you know happens to talk about bitcoin a whole lot but is also like

Um, you know with with people that are like fun and entertaining and have like interesting perspectives

And every once in a while casey will go on a rant about like how bitcoin, you know will help to

Dematerialize the state if only we get people to use it and dgens love uh jpeg

So here we are with inscriptions

And so it's it's simultaneously like the most accessible podcasts and also the most based one in bitcoin

So go subscribe to hell money podcast give them five stars on itunes

Yeah, it really it really lurches and veers between uh between two extremes and sometimes we wear costumes

I think we've done like three costume episodes

Highlight was probably me dressing up as trump and erin dressing up as as aoc

Yeah, that one was pretty good. I also liked the one where

You were the liver king and she was that weird dude who's trying to live forever. Whatever that guy's name is

Uh, that one was pretty good, too

But yeah, everybody's going brian johnson and brian johnson. They're brian johnson. They're brian johnson. Yeah serious

What is it like aries whatever energy? I don't actually know anything about astrology. I just larp so that erin doesn't unfollow me

That's a good strategy eric go ahead

No, oh, I think we lost eric. Um, so we've been talking about like here. Oh, you're here

Yeah, I think I accidentally just muted myself

I was going to say that the recent episode with germany ruben

I see germany popping into these various spaces and people are trying to squeeze him on like tell us about your

Experience of trying to activate ctv

So but he's been a little bit elusive on twitter space

People keep like asking him for this story, but he does go through the entirety of well

Maybe not the entirety

But he goes through a thorough account of what it was like from his perspective trying to get ctv activated

And the different curve balls that was thrown at them through that process

uh

So if anyone is like interested in that particular story, I think that maybe the hell money podcast episode with germany

Is maybe the best uh account of that story and one of the one of the most interesting stories in in in bitcoin

uh

I think

Well that that that's like a perfect segue. I was going to see if we want to talk about cat and covenants since we've been talking about

Quantum cat inscription stuff for a little bit. Um, I don't know if you guys want to talk about

the

The thing that quantum cats is actually about right which is

uh, you know bitcoin development process and covenants and how do we get like more

Expressive script construction so that we can scale bitcoin for more use cases in like non-custodial ways

So, um, you know for folks who uh might not know if your timeline isn't blowing up with op cat stuff

You're following the wrong people. Um, you know, the idea here is uh,

A lot of folks in bitcoin including, uh the taproot wizards team

We we really think that there's some new capabilities for bitcoin. That would be really helpful for people for

You know scaling functionality privacy and um, there's a bunch of options that people have proposed

They're all variants of uh, a general class of capabilities called a covenant

Which is sort of like how do you put restrictions on?

The ways that coins can be spent and that doesn't sound super useful

But what it lets you do is it lets you build?

You know protocols or um other bitcoin smart contracts that let more people

Interact with with bitcoin

Uh in a way where they can control sort of their part of a utxo

But they can't steal other people's money and that's really useful for like multi-party protocols

It's useful for some like interesting privacy designs scaling lightning, etc

Um, jeremy, uh, jeremy ruben came up with a really great idea several years ago called check template verify or ctv

and it's um, it's this like very clean very elegant way of uh constraining the ways that

Some coins can be spent but there's been some other

Proposals that have come up through the years and one of them is there's an old bitcoin opcode called op cat

That got deactivated by satoshi before he or she left and um, you know something that

People have said over the years. Um, sometimes in response to ctv

Sometimes on their own is like we don't need these other designs. You can do everything that you want to with cat

And um, you know, there's lots of people talking about ctv and very interested in ctv. I think we're all fans of ctv

um

But this conversation just hasn't really been able to to move forward for a bunch of reasons and um

You know, even though it's not moving forward people

hold up

These alternative covenant designs as reasons to not move the conversation forward like if if we don't know

Enough about all these different designs then we shouldn't talk about anything is is kind of like the way that this ends up going

So one of the things that we really wanted to do

With quantum cats was we wanted to see if we could

Use the the kind of like energy mobilization that nft

Collections are really really good at producing and see if we could get um, you know

More and different bitcoiners to engage in the conversation about covenants generally

Maybe through the lens of looking at op cat, but we we really want to to help people explore the whole design space and so

You know, we we've had our community through a quest

Be doing a whole lot of research on op cat what it's good for what it's bad for

Um reasons to activate it reasons not to activate it. I've been doing some dev work on

Actually using cat with snore signatures to do covenants in bitcoin

And uh, you know, we're trying to see if we can just move

The whole conversation forward a little bit like, you know, maybe maybe not all the way

But maybe we can we can help get it a little bit on stock

Um, and who made that uh, do you want to talk about a little bit more general background like, you know,

Danny I suspect you can hear kasey

I think because kasey was speaking too and I think you didn't notice

So is that okay so so danny if you can um drop down from the stage for a minute and you will bring you

Back up that should fix it

I was gonna say who made that website that like a bitcoin bip journey website

That's one of the best like bitcoin educational

Like documents that i've i've ever seen

So, uh, we just brought tyler up on stage tyler is on the tapper wizards team and he

Was responsible for bipland. Um, you know tyler

Did so much research to produce bipland. It's insane

like my

I think my my first or second week at the company

I was on the phone with tyler for like six hours just like giving him

My brain dump on like, uh some of this stuff

Uh, like, you know, my I I think like my my only two contributions to the whole thing were uh,

The covenants bike shed and the moving goal posts

Um, I don't know if anybody else came up with those but i'm gonna claim them because I love them but tyler talked to

Just like a huge array of people across the whole bitcoin dev ecosystem and got

So much feedback and synthesized all of it and like I don't know casey. I don't know if you've seen this but on the bipland

Like map each of the stops are actually links to like mailing lists posts or articles or other things

So it's it really is like this this, you know map to to guide you to the primary sources for some of these stops

But it's been amazing because we we put that out and it's you know

It's a very like tongue-in-cheek, but sort of also like tough love

Look at you know, what is objectively a convoluted process?

and the it's really funny because um, like the unanimous feedback that we've got is

Like even people that we thought would hate it like all love it like the day we put that out

I just got all these dms from, you know, dev friends of mine being like

This is incredible. Like all the in-jokes are hilarious. I'm sending this to everybody that I know it's uh, it's it's been really great to see

Randall there's just one one thing that i'd like to adjust with what you just said. Um

We didn't

To my uh memory we didn't

We didn't uh assign the responsibility of bipland to tyler tyler just produced it

I think in the yeah, it was his idea

Yeah, it was absolutely his idea. Here's bit. Here's bipland

Uh just came out of I don't know tyler, but did it like come out of your own?

uh

like

Quest of trying to understand the process of like how

Because from to my memory just came out of like one day there was bit planned

In the middle of all this and we're like all of us were like, holy shit

This should this is going to be like the best part of this of the whole

Collection. Yeah, I think it really came from

a couple things

one was

As I started talking to everyone

It literally was inspired from candy land because no matter who I talked to

They're all like you hit this point and then your bip gets killed and you're basically sent back to start and you you

you're done, you know, whether it was

Jeremy or whoever it was like there are every every step along this process. There are so many traps and pitfalls

That you can fall into and and you just get

shuttered and and sent back to the start and so

That just made me think of candy land as a kid

And I think candy land also is like it's a magical fun fairy tale place and I think it really

Resonated with the taproot wizards brand where I was just like yeah candy land is very magical. It's very whimsical

It's it's very on brand

for the wizards and then yeah, the second thing is

I

At one point I started doing a write-up on this and it it got so long where I was like this is going to be another

dissertation for me to explain the issues that the organization around bitcoin has and so

This is not the most effective medium to communicate this to people and I think we need to do it in

You know a light-hearted fun way that you don't have to read about, you know

the institutions that have been built up and

The history of bitcoin and the scars that are in the organization to understand what's going on

You know, we can just give you a fun little map to follow

And hopefully each one of those steps prompts a little rabbit hole that you fall down into

And you know through those bits and pieces you can eventually

Learn a bit about how bitcoin as an organization has functioned in the past

Um, so that yeah, that was really where it came from is just uh, that plus udi had bought me an ipad and we were starting

Uh starting to do technically bitcoin with yon and I was like, well

I might as well get some use out of this thing and and do some drawings

So a little bit of serendipity, but that's kind of the backstory as to where it came from

Yeah, all that work, I mean paid off like so big I was so impressed by it and it also is

Like not only is it a good educational resource, but it's a good educational resource for something. That's

So mysterious, you know, like when you're

Thinking about like, okay

Like how do changes in bitcoin happen like generally the conversation happens in so many different places in so many different ways and

the

like the

The the venues and what you're there for is so confusing you're like I want to change bitcoin like what do I do?

Do I open an issue on?

Like bitcoin cores github do I like open a pr just making the change?

Do I you know go to?

the

Like the dev mailing list like people don't even know that the dev mailing list exists

Like how do I even post the dev mailing list? So it's all this like super confusing

like tacit information

Uh, so it's it's amazing that it that it works so well and like yeah, it's just I mean like i'm just gushing now

It's a it's a great resource. I really appreciate it

It's kind of funny, right because like, you know the the aesthetics of it

It's this thing where it's like, you know, uh, these these squiggly hand-drawn things

That you know, like it looks messy and raw and like, you know, tyler scribbled it out on his ipad the first week

He ever owned an ipad, right and um, you know people

people who aren't familiar with

bitcoin history

I've heard this feedback a couple times. They're like, yeah, I looked at this and I thought this is a big joke

Like there's no way that the process is anything like this

And then they go and talk to like other devs or other people who've been around bitcoin

They're like, no, this is actually really funny because this is the most accurate representation. We've ever seen like it's it's

hilarious

Ooh, do you do you remember that moment when we were in Palo Alto during the uh,

um

the stanford, uh

conference and we had like uh

A meeting with a lot of zk developers. We've had like a

A bunch of different academics and companies wanted to get together and discuss like how do we get zero knowledge proofs into bitcoin?

And I remember ellie ben

Sasson from stark where was there and ellie has been around he's the ceo of stark when he's been around

Uh bitcoin since ages since like he has been having presentations about it

since 2013 so it's not like a rookie in the bitcoin world by any means and

uh, like the starkware team has

is of the opinion that obviously the

The ser knowledge proof execution environment that's suitable for bitcoin is the chiral

Uh language that and the chiral vm that they have that they're using on starknet and they were like

So who do we go to now to like what is the who are the entities and who are the people that we're supposed to go?

And make this proposal like where where is where is it and a bunch of people were sitting around that table like around the room

some of them were more senior and

veteran people of the bitcoin consensus process and

there's sort of like

It's kind of difficult to explain like so if you don't know it's going to take a very long time to explain

where you go to convince people of

like where you go to make the the

Proposal for the specific bitcoin improvement proposal. So people just like looking at each other and sort of laughing

Well, that's actually a difficult question to ask

Like how do you even answer that question?

And I think that bit plan is like that's the link that i'm gonna send to elie l.i

Elie now that now he has a resource that he can can look at

Uh, maybe maybe that's like the best explanation that he can get

Yeah, I had a really similar experience I was working on a um

Uh non-custodial or what I would say like a assisted custodial bitcoin wallet

before I I started full-time at taproot wizards and and like I was talking to somebody who was new to bitcoin and

We were talking about like a proposed opcode that would be

Really useful for for certain classes of recovery and they were like, oh cool

So so like when's that going to get turned on and I just laughed and they're like

Oh, no, is it going to be like in a year and I like laughed more and it took them a while to to understand

Why I was laughing at them

But

Yeah, yeah, I think in some ways I got lucky in that. I

I'm I didn't know anything about bitcoin this time last year

Like my knowledge of bitcoin was I was reading a white paper about a zk rollup being built on bitcoin

and that was the extent of my knowledge and so I think

I I think i'd benefited a lot coming into bitcoin with a fresh perspective

And just being like well i'm new here

Let's figure this out

Surely surely there's something that exists and bipland was just kind of the the emergent thing that came out of

Let's go do research and talk to talk to as many does as we can about how this process really unfolds

Is

Yeah, so that was that was like one part of it was bipland and then the other part of it to go with it is

We we launched this quest

for people who are interested in the quantum cats collection and

There's you know multiple parts of the quest but one of them was we asked people to go out

and learn about and and here we we made it a little bit choose your own adventure, but

You know, the the prompt was something the effect of like, um

You know go learn about like op cat learn about the history of it the use cases

Um, you know like what you can do with it why people would find it interesting, but also, you know

if if you find reasons why

We shouldn't activate it or if there's alternatives that you think are better

Then, you know find some way of of explaining those two and like use your creativity. So it could be

A twitter thread it could be a video

Uh, like whatever you want. And so over the last couple days, we've had all this content just blowing up all of our timelines

Where people have been recording

videos

One person made this like really cool interactive website where it looks like an old like windows 95 installer

And it shows you what op cat does and then actually shows you

How without the stack size limits that were introduced in taproot how like op cat can just like

Exhaust memory by doing duping cat over and over again

Um, we had another team of people

Um start playing with uh, like putting together a little bitcoin script that used op cat to uh, to like

Smush two halves of the signature together and do check sig which was kind of fun to see

Um, we've had a few people, you know, write up stuff about like ctv and like why it might be more space efficient for certain applications

So, um, so, you know, we we've got this

Uh group of people that are are digging into op cat and you know

like there's there's a lot of variance amongst uh, like the quality

of some of the submissions but

You know like the the goal here is like you get a lot of people excited

A subset of them are going to do like casual research a subset of them are going to do serious research

And sort of at the bottom of the funnel you're going to have um,

You know some people making like really high quality content

That is about either cat or covenants more generally

And that just becomes like community resources and and you know when somebody else wants to learn about these

These different options then there's like more content that's available for them

And and you know, like i've taken the time that quest four has been going to go and uh, there's this old

Andrew polstra blog post about how you could um

Use cat along with a specially crafted snore signature to do a basic covenant

Uh sort of like check sig from stack, but but not quite all the way there

And uh, andrew polstra wrote about this years ago and for the last couple years

Um, you know anytime somebody says, you know, we don't need ctv and apo. We can just use cat

Like they point to something that points to something that points to something and it always like bottoms out at this blog post

Um, I did some research and as far as I can tell nobody actually ever implemented what was in that blog post

So I I like went and did it and so, you know, i'm currently working on um

A whole bunch of little covenant applications using op cat

I've got like a basic output constraining covenant

um, I think i'm working towards a vault but uh, like, you know, it's it's going to be sort of like, uh,

The the tapper wizard's like first party contribution to the it's like hat conversation is going to be some

You know working code that people can run and play with that actually uses cat for covenants

And then we can put that next to other proposals like l enhance or ctv

Or op vault and we can have like a better apples to apples comparison

Of you know, which of these covenant proposals achieves?

What we want out of bitcoin and and maybe we can you know move a little bit past the bike shedding a little bit sooner

I wanted to make sort of like a

Meta point or or like a meta comment, which I think is really interesting. So

It kind of feels to me like um bitcoin script

Is just on the edge of being incredibly expressive?

And that there are these op codes

That they sort of have their specialty they're sort of designed for like one thing, you know, like op cat

Well, I mean op cat is like a weird beast. It does a lot of things but there's also apo which was you know

sort of designed to

Make a better version of the lightning network. There's op ctv, which is designed for

um, you know making these like pre created trees and transactions, but

they

have a lot of

overlap in the things that they enable and it's a really common part of the discussion where

Somebody's talking about apo and like what they want to do with apo and somebody else is like

Oh, yeah, you can do that with op cat or ctv

And the other thing

Like the other way around also happens where you're like, oh talking about apo and you're like, oh you can kind of do that

With ctv or you can do that with op cat and it kind of reminds me of the situation

um

Of Turing completeness in a programming language, right? This is sort of the

concept that all programming languages above a certain amount of expressivity

are equally

Capable of sort of computing all the same things now. There might be efficiency trade-offs between programming languages

but generally they can

All calculate the same things or run the same programs

And there's this interesting thing when you have a programming language where or even a non-programming language where very often it

It accidentally becomes Turing complete because of some you know thing you added to it

So I think the two components that you need for Turing completeness is you need

conditional

Branches you need to be able to take two forks in the road

And you need to be able to jump backwards

So you can do like essentially like iteration or something like recursion

And like once those two things are added

Then you get this explosion in the amount of expressivity of the language and and these things

Will not make bitcoin Turing complete, but it feels like bitcoin is just on the cusp of

a new layer of

Expressivity and these sort of like these these op codes

Let you let you do that and they all have their specialties

And yeah, that overlap isn't perfect

But a lot of them kind of get you to the same place in terms of expressivity. I don't know reindahl. Have you sort of

Note notice that or thought about that

yeah, well and like

Like absolutely, and I think that that's good and also bad at the same time, right?

Because I think one of the things that's really challenging

About the whole covenants discussion is that there is this sort of like reducibility of covenants, right?

So it's like one of the reason why some people like cat is because in theory

A bunch of other covenants proposals sort of like reduced down to cat or like a constrained form of cat, right?

So like you can use cat plus snorzing jurors to emulate ctv. So

You know like the things that you can do with cat are a strict superset of like what you can do with ctv

So sometimes when people say like oh we should do ctv because it's like really constrained, you know

It's not gonna unleash drive chains on the world or whatever

Um, some people will say oh, well, you know, you can emulate ctv with cat

So why don't we just do cat and then we get sort of ctv plus other stuff and the trade-off there, right is like, um

In order to emulate ctv with cat it takes, uh much larger scripts

Um, and like i'm happy to get into what the construction is

But the short version is like you have to build up a snore signature on the stack

And then you call check sig and that like validates it that that ends up being

More code and more data than just having a 32 byte hash and calling, uh, you know op ctv

Um, so it's sort of like you you can um, you can kind of like

It's almost like a risk versus cisc thing, right? Like you you have a more primitive operation

But you can do it more like like a higher number of times in order to emulate the more complicated operation

Um, the thing that I think is missing from a lot of those discussions though

Is like what's the actual feasibility or practicality of that?

um, like i'll give you an example like one of the

Things that a lot of people are excited about i'm excited about I think it's cool

Is this like bitvm idea that's been talked about lately and like bitvm is really cool because in theory

It would let us do sort of like Turing complete

um

Smart contracts that can resolve or that can be verified on chain like much more cheaply

So you're not doing the contract execution on chain. You're doing verification of a fraud proof on chain

and um, that's like a really exciting primitive but

like

Economically and in terms of the size of the transaction graph

Is that actually going to be practical to use?

Right because like with with Turing completion, right like or Turing completeness, right?

Like there's you know, you can build a Turing complete computer out of pebbles on the beach

But you're not gonna you know

Like do a lot with that, you know, just because something is is theoretically possible doesn't mean that it's actually practical

and I feel like a lot of these covenant proposals fall into that trap a little bit where um,

You know if if something is going to be used a lot

By the network all the time then we probably do want to have an op code that specializes in it

but if we if we you know don't think that something's going to be used all the time or if we

You know aren't sure then maybe you could make an argument that we want sort of a lower level primitive

That then people can compose into the higher level functionality

And so, you know one of the things I think so so that distinction of like do we want

The more specialized but more limited instruction or do we want the less specialized but more verbose instruction?

I think that that

argument is is one of the things that kind of stops progress on this discussion

because um, you can't really

You know figure out what the obvious right thing is to do when you're sort of arguing it from first principles

And so one of the things that I think we're hoping to produce out of this effort is

You know can

Us and the broader because right it'll become up with like others too or just me

I think I think maybe just you I can hear him. Okay

Yeah, so like one of the things that that we're we're hoping comes out of this whole quest and this whole effort

Is maybe we can have some more concrete examples of like here's what ctv looks like with a dedicated instruction

Here's what ctv looks like with cat emulation

Let's put them next to each other and see like does this actually make sense and maybe it doesn't maybe it doesn't actually move

The conversation forward, but it's it's something that we can do that's more concrete than just uh, sort of arguing about it

I think these concrete samples are actually really important. Yeah

There's so many ways that the discussion like gets gets bogged down. Um, which is really unfortunate

If you sort of look at op cat and its implementation, it's so so so so or not up not up cat

Not up cat. Well, op cat is marginally more complicated. But if you look at op ctv in particular, it's so simple

And it's like, okay, we add this one op code

I mean, I think it's like a hundred or two hundred lines of code maybe even less

That actually implements the opcode in bitcoin core not not counting tests

But it just kind of gets bogged down in this, you know, like bike shedding hell

Where everybody is like talking about like oh like maybe it should include different fields or like oh, here's this like alternative proposal

But that nobody's worked out or maybe we could do this with a with a with a different thing. Um, and I think it's really

Unfortunate for op ctv like I mean, I think op ctv is one of those things like if you don't like it don't use it

It's like no, no skin off your back

And it's so simple and it's it's

I don't as far as uh covenant or uh, yeah covenant proposals go. I don't think anybody's proposed a simpler

Covenant so it has this really appealing property that like, okay, maybe it's not exactly the absolute most optimal thing

But it's certainly the simplest thing and it opens up

Experimentation and so yeah, we should probably just probably just do it op cat is not super complicated

But op cat is is more complicated, you know, you have to think about like bitcoin

Script size or bitcoin script like stack size limits and stuff like that

But it's it's it's it's only like marginally more complicated. But yeah op ctv

And so and so it's it's it's funny like one of the things that i'm so uh, you know

One of the things i'm discovering as i'm actually working with cat

um, so two things one is like

I I really was worried about the stack size limits

And what i'm finding is that in practice i'm actually not running into them yet

And maybe when I get a little bit closer to having a vault done I will but um, you know, I

Like what the the data that you actually need to manipulate on the stack in order to build this like special snore

Snore signature ended up actually being more compact than I was expecting because um

I I naively thought that I would need like a like a fully serialized transaction on the stack

and then I would hash it and then that's what I would use because like the the

The thing that everybody sort of knows or that I I thought I knew was you know

Okay

The way that you use cat to do a covenant is you build a serialized transaction on the stack?

You hash it and then you use some kind of magic

In order to assert that that hash matches the current transaction that you're inside of because like

If you if if I told you that you could emulate checksig from stack

Like that's that's what you would or that's what I would assume would happen

It turns out what you actually do is you build up like the um

Like the sig hash which doesn't have the fully serialized transaction. It has

Like commitments to all the inputs it has commitments the output, but those commitments are like you

concatenate

the uh

Like consensus serialization of all the outputs and then you hash them and you include that

In the uh in like the message that then you hash and becomes into the sig hash, right?

And so, you know, you you have a bunch of pieces that you're moving around

But what you end up with is like you have a bunch of hashes that are all attached to each other and those are all fixed size

right, so

Um, i'm running into the the stack element size limit way more than I thought I would

um

And then the other thing that i've discovered is uh, I don't think you actually can do full-blown checksig from stack

discovering that cat is simultaneously like

Uh simpler to use but also more limited than maybe the the hype

Um, and and I think it's because like nobody's nobody's actually like gone and done it yet

So i'm really looking forward to having like a fully

Worked out example of something real that people can play with and and maybe we can have some better

reasoning around

What this thing can and can't do and whether this is better or worse than something like

You know ctv plus checksig from stack or ctv plus op vault or whatever. It's gonna be

I switched to uh to another account

I have two phones here and uh, sorry about uh, oh my god, udi. I didn't know that you were the quantum cats account

I'm the quantum cat i'm both dead and udi

Is the way it works? Yeah

Yeah, yeah, that's that's my that's my big quantum cat joke for today

Are we are we decided though? If like if like op cat is is good or bad for bitcoin

It's like if it's an attack for bitcoin or if like we part

Did we figure that out?

Well, you know, I I don't think we figured it out yet

And you know what i've what i've learned from uh, my twitter mentions is that trying to figure it out is an attack on bitcoin

So, you know, like I don't I don't know if if cat is good or bad. Uh, I don't know if

ctv plus vault is good or bad, but I know that uh trying to figure it out, um in public with like research and

Implementation is is an attack on bitcoin

And if you have fun while doing that rindall then then like I may god have mercy on your soul

Yeah, that's definitely an attack on bitcoin

So

Well sounds like we have this resolved

I'm really glad that we figured it out

That's really cool. Um, yeah, I look I don't know about you guys. I learned a lot

You probably didn't because you just said things that you already know

But I learned a lot

So it was great for me. I hope it was great for everyone else. Um, I hope you enjoyed bitcoin in general because

We're breaking it. So it's over but

But it was a fun time it was a fun 15 years

And

Yeah, I hope that before we break it. There's something else fun to do

No, but seriously though, I at least like i'm a troll, but I think for everyone else on stage

It should be obvious that this whole thing is really like it's like it's a love letter to bitcoin

It's people really care about bitcoin a lot and it's just a way I think that's how I see it. So I had to express like

Why

We are so obsessed with it, you know

And it's just just a tool to express it. So I hope some people can see that maybe

Anyways, thank you everyone

For uh sharing this space with us. This was great

Uh, if you're in the audience go ahead and follow everyone on stage. These are all visionaries

You don't have to follow me but everyone else and um

And yeah, thank you for joining us

Uh, thank you for everyone who participated in the quest by the way for quantum cats. It was we just

Wrapped it up a couple hours ago. Um, we're gonna have the results

Uh very soon ready for you all but really thank you for the participation. It was all amazing

I mean I especially today there's so many submissions today

And they're really good. So thanks for that. Uh, thanks for being with us and uh, yeah jim. Yeah. Have a great day

### Host

- [saudiverse 🇸🇦](https://alphagrowth.io/spaces/participant/udiwertheimer.md)
  [X profile](https://twitter.com/udiWertheimer)
- [Quantum Cats](https://alphagrowth.io/spaces/participant/quantumcatsxyz.md)
  [X profile](https://twitter.com/QuantumCatsXYZ)
- [Rijndael](https://alphagrowth.io/spaces/participant/rot13maxi.md)
  [X profile](https://twitter.com/rot13maxi)

### Speaker

- [Dan Held 🧙‍♂️](https://alphagrowth.io/spaces/participant/danheld.md)
  [X profile](https://twitter.com/danheld)
- [Eric Wall](https://alphagrowth.io/spaces/participant/ercwl.md)
  [X profile](https://twitter.com/ercwl)
- [Casey](https://alphagrowth.io/spaces/participant/rodarmor.md)
  [X profile](https://twitter.com/rodarmor)
- [huuep.eth (🍌,🍌)](https://alphagrowth.io/spaces/participant/huuep.md)
  [X profile](https://twitter.com/huuep)
- [Tyler Whittle](https://alphagrowth.io/spaces/participant/dr_dao.md)
  [X profile](https://twitter.com/Dr_DAO_)
- [Dev Aloft](https://alphagrowth.io/spaces/participant/devaloft.md)
  [X profile](https://twitter.com/devaloft)

## Want more TVL for your protocol or chain?

AlphaGrowth grows TVL for protocols and chains through asset acquisition and integration, grants, incentives and LP deals. We built this data and keep it free.

[How we grow TVL](https://alphagrowth.io/defi/index.md?utm_source=markdown&utm_medium=ai_agent&utm_campaign=aeo) [Talk to AlphaGrowth](https://alphagrowth.io/contact-us/index.md?utm_source=markdown&utm_medium=ai_agent&utm_campaign=aeo)
