NFTs in the Bitcoin blockchain - Ordinal Theory

[Wind_FURY]

There's been some strife and turmoil within the Ordinals community recently. Turns out Rodarmor hates BRC20 and is looking for ways to sabotage them.

https://twitter.com/rodarmor/status/1705369258426036526

Can't you grief BRC-20 holders by spamming their addresses with transfer inscriptions, locking their balance, which they would then need to send back to themselves in order to unlock their balance?

I believe part of the issue is the debate about Casey Rodarmor's proposal to change the numbering system of inscriptions. I'll not pretend that I understood everything technically to have a strong opinion, and I would welcome anyone who understands to post an ELI-5. BUT, I hate admitting this, franky1 is right. Because Casey Rodarmor invented the arbitrary numbering system that isn't truly backed by anything, he can also "change the rules" arbitrarily.

[vjudeu]

and I would welcome anyone who understands to post an ELI-5

It is quite simple: the main goal of Ordinals was to push data on-chain. And it was 100% delivered. You have that famous, almost 4 MB block, as one of the first inscriptions. It is like the Genesis Block, but for Ordinals: it shows you exactly, what this system is really about.

And then, on top of that, you have a fee-like ownership system, that was artificially added, just before release. Because well, if your only goal is to just push some data, then you don't need any proper ownership at all! But then, you realize that people will want some kind of ownership, if your protocol will evolve over time.

So, you have a perfect cloud-storage-like system, with no ownership enforced. How to solve that? Well, you can look at transaction fees, as your inspiration. You can easily note, that miners don't need to explicitly be listed as one of the recipients of transaction fees. Everything is counted implicitly. So, maybe you can enforce ownership, if you just count single satoshis, and assign ownership to each of them?

And that's how Ordinals ended up with the system we know today. The main idea behind ownership, was to just map each satoshi to its owner. But because an earlier version was cloud-storage-only, there was no ownership. Old signatures didn't contain any information about ownership. The creator had no idea about sighashes, and thought that the whole transaction is always signed. Also, the creator thought that all amounts will be non-zero, maybe because of the dust limit (that is a standardness rule, but not a consensus rule).

Then, the first version of the code was created. It just counted all satoshis on your inputs, and all satoshis on your outputs. And the whole idea behind tracing satoshis is very simple: if you have any transaction, you can sum all amounts in all inputs, this is your entry value. Then, you sum all amounts in all outputs, this is your exit value. If your outputs are smaller than your inputs, then you have some fee, so you place it at the end. Then, you have this picture:

Code:

+--------------------------------+
| Alice    1.23 -> Daniel   6.50 |
| Bob      4.56    Elaine   3.10 |
| Charlie  7.89    Frank    4.05 |
+--------------------------------+
| inputs  13.68 -> outputs 13.65 |
|                  fees     0.03 |
+--------------------------------+

And then, you can try to map the ownership. If Alice owns some satoshi with index "0", then it goes to Daniel. However, because signatures are not checked at all, then what if all inputs used SIGHASH_ANYONECANPAY? Or if some wallet will sort your inputs differently? Then, this thing can happen:

Code:

+--------------------------------+
| Bob      4.56 -> Daniel   6.50 |
| Alice    1.23    Elaine   3.10 |
| Charlie  7.89    Frank    4.05 |
+--------------------------------+
| inputs  13.68 -> outputs 13.65 |
|                  fees     0.03 |
+--------------------------------+

Then, guess what, your precious satoshi with index "0" is no longer owned by Alice, but by Bob instead! The situation is even worse than that: if you assume that all of those satoshis are indexed from "0" to "13,68,000,000" (exclusive, like zero-indexed arrays), then you can own satoshi number "13,67,999,999", if you index things from zero. And what happens then? Of course, this particular satoshi is then sent as fees, to some random miner, and then it can fly between a lot of different addresses, if people are unaware, that they "own" some Ordinals!

And there is more: imagine that people decided to include some message to their transaction:

Code:

+---------------------------------+
| Bob      4.56 -> OP_RETURN 0.00 |
| Alice    1.23    Daniel    6.50 |
| Charlie  7.89    Elaine    3.10 |
|                  Frank     4.05 |
+---------------------------------+
| inputs  13.68 -> outputs  13.65 |
|                  fees      0.03 |
+---------------------------------+

What happens then with the precious satoshi with index "0"? Well, the first input is OP_RETURN with zero amount. And then, if you code it in a wrong way, this transaction may mess up your global satoshi counter, when you start from "0", and you plan to end with something around 21 millions.

[nutildah]

I believe part of the issue is the debate about Casey Rodarmor's proposal to change the numbering system of inscriptions.

Except that's not what happened, or was even being discussed.

https://twitter.com/rodarmor/status/1706440775998062655


What happened was a self-described "Ordinals Influencer" is using his clout on social media to make sure any changes that are made will help the value of his own bag of Ordinals.

I definitely believe the entire thing could have been better thought-out before being executed, but one thing that's not happening is Rodarmor trying to change the numbering system.

[Wind_FURY]

and I would welcome anyone who understands to post an ELI-5

It is quite simple: the main goal of Ordinals was to push data on-chain. And it was 100% delivered. You have that famous, almost 4 MB block, as one of the first inscriptions. It is like the Genesis Block, but for Ordinals: it shows you exactly, what this system is really about.

And then, on top of that, you have a fee-like ownership system, that was artificially added, just before release. Because well, if your only goal is to just push some data, then you don't need any proper ownership at all! But then, you realize that people will want some kind of ownership, if your protocol will evolve over time.

So, you have a perfect cloud-storage-like system, with no ownership enforced. How to solve that? Well, you can look at transaction fees, as your inspiration. You can easily note, that miners don't need to explicitly be listed as one of the recipients of transaction fees. Everything is counted implicitly. So, maybe you can enforce ownership, if you just count single satoshis, and assign ownership to each of them?

And that's how Ordinals ended up with the system we know today. The main idea behind ownership, was to just map each satoshi to its owner. But because an earlier version was cloud-storage-only, there was no ownership. Old signatures didn't contain any information about ownership. The creator had no idea about sighashes, and thought that the whole transaction is always signed. Also, the creator thought that all amounts will be non-zero, maybe because of the dust limit (that is a standardness rule, but not a consensus rule).

Then, the first version of the code was created. It just counted all satoshis on your inputs, and all satoshis on your outputs. And the whole idea behind tracing satoshis is very simple: if you have any transaction, you can sum all amounts in all inputs, this is your entry value. Then, you sum all amounts in all outputs, this is your exit value. If your outputs are smaller than your inputs, then you have some fee, so you place it at the end. Then, you have this picture:

Code:

+--------------------------------+
| Alice    1.23 -> Daniel   6.50 |
| Bob      4.56    Elaine   3.10 |
| Charlie  7.89    Frank    4.05 |
+--------------------------------+
| inputs  13.68 -> outputs 13.65 |
|                  fees     0.03 |
+--------------------------------+

And then, you can try to map the ownership. If Alice owns some satoshi with index "0", then it goes to Daniel. However, because signatures are not checked at all, then what if all inputs used SIGHASH_ANYONECANPAY? Or if some wallet will sort your inputs differently? Then, this thing can happen:

Code:

+--------------------------------+
| Bob      4.56 -> Daniel   6.50 |
| Alice    1.23    Elaine   3.10 |
| Charlie  7.89    Frank    4.05 |
+--------------------------------+
| inputs  13.68 -> outputs 13.65 |
|                  fees     0.03 |
+--------------------------------+

Then, guess what, your precious satoshi with index "0" is no longer owned by Alice, but by Bob instead! The situation is even worse than that: if you assume that all of those satoshis are indexed from "0" to "13,68,000,000" (exclusive, like zero-indexed arrays), then you can own satoshi number "13,67,999,999", if you index things from zero. And what happens then? Of course, this particular satoshi is then sent as fees, to some random miner, and then it can fly between a lot of different addresses, if people are unaware, that they "own" some Ordinals!

And there is more: imagine that people decided to include some message to their transaction:

Code:

+---------------------------------+
| Bob      4.56 -> OP_RETURN 0.00 |
| Alice    1.23    Daniel    6.50 |
| Charlie  7.89    Elaine    3.10 |
|                  Frank     4.05 |
+---------------------------------+
| inputs  13.68 -> outputs  13.65 |
|                  fees      0.03 |
+---------------------------------+

What happens then with the precious satoshi with index "0"? Well, the first input is OP_RETURN with zero amount. And then, if you code it in a wrong way, this transaction may mess up your global satoshi counter, when you start from "0", and you plan to end with something around 21 millions.

Ser, I asked for ELI-5. Haha.

Thanks for the well-illustrated explanation. Yeah, Ordinals is merely a made-up numbering system from the mind of Casey Rodarmor, which from the viewpoint of the blockchain the numbering/indexing is fluid that needs to resequenced? Did I get it right?

I believe part of the issue is the debate about Casey Rodarmor's proposal to change the numbering system of inscriptions.

Except that's not what happened, or was even being discussed.

https://twitter.com/rodarmor/status/1706440775998062655


What happened was a self-described "Ordinals Influencer" is using his clout on social media to make sure any changes that are made will help the value of his own bag of Ordinals.

I definitely believe the entire thing could have been better thought-out before being executed, but one thing that's not happening is Rodarmor trying to change the numbering system.

Are the numbers supposed to be so important? Isn't the ART, and having to proof that you own it, that's supposed to be the most important?

[stompix]

Has something happened these days that led to this drop in BRC-20 tx?
https://dune.com/queries/2432736/3996424
I checked to see if it's an error but no, the mempool keeps going down so definitely there is a pause in those, in terms of raw volume yesterday was the 6th lowest from April!!

Doesn't look like the fad ending so abruptly so maybe a large service related to those experiencing issues?
Is this the result virtual monkey zoo going offline or whatever the congregation point of those decentralized assets is?

[ABCbits]

--snip--

I believe part of the issue is the debate about Casey Rodarmor's proposal to change the numbering system of inscriptions. I'll not pretend that I understood everything technically to have a strong opinion, and I would welcome anyone who understands to post an ELI-5. BUT, I hate admitting this, franky1 is right. Because Casey Rodarmor invented the arbitrary numbering system that isn't truly backed by anything, he can also "change the rules" arbitrarily.

While @nutildah already clarify what Casey said, i'd like to point out Ordinals community doesn't have to follow changes made by him either. In fact, supporting new rule require code change and software update.

--snip--

Are the numbers supposed to be so important? Isn't the ART, and having to proof that you own it, that's supposed to be the most important?

Don't forget Ordinal handbook mention 6 type of satoshi rarity. I expect people who like lucky/good number also find it's important.

Current Supply

    common: 1.9 quadrillion
    uncommon: 808,262
    rare: 369
    epic: 3
    legendary: 0
    mythic: 1

[nutildah]

Are the numbers supposed to be so important? Isn't the ART, and having to proof that you own it, that's supposed to be the most important?

Don't forget Ordinal handbook mention 6 type of satoshi rarity. I expect people who like lucky/good number also find it's important.

It is a way of creating "artificial scarcity" for the sake of assigning extra value to individual sats... Although there is some logic involved in the way the system has been created, most people don't recognize this value and they'd need to be onboarded -- similar to the way they are onboarded into NFTs or other collectibles, I suppose. Its just that the value proposition is particularly ridiculous because it didn't even exist a year ago.

Maybe if this type of Ordinals numbering system for satoshis sticks around in its current form for a decade, then it will be taken a bit more seriously... But its highly doubtful.

[NotATether]

Don't forget Ordinal handbook mention 6 type of satoshi rarity. I expect people who like lucky/good number also find it's important.

Current Supply

    common: 1.9 quadrillion
    uncommon: 808,262
    rare: 369
    epic: 3
    legendary: 0
    mythic: 1

Looks like during the entire minting timespan, there were 800k uncommon,  369 rare and 3 epic sats created (I'm willing to bet the one mythic is some kind of genesis transaction so it doesn't count).

It also seems that people are getting impatient with minting though, because the mempool size has been steadily deflating.

[d5000]

Has something happened these days that led to this drop in BRC-20 tx?

I've noticed that too, as I've also created a related query dashboard

Apart of the community "turmoil" mentioned by nutildah above, maybe the "Runes" proposal by Casey Rodarmor has to do something with that. See the proposal here. I've not looked really at it still, at a very first glance it looks like ... coloured coins with another name?

On the other hand, Ordi token is doing relatively well, it has lost 85% of its value from ATH but has stabilized already some weeks ago and even had a bounce. AFAIK this is one of the most popular Ordinals/BRC-20 tokens.

So I'm currently quite clueless, but your suspicion about most of the Ordinals activity being maintained by a few centralized actors (regardless of their intentions) could be true actually.

[stompix]

Has something happened these days that led to this drop in BRC-20 tx?

I've noticed that too, as I've also created a related query dashboard

Yeah, the more time goes by the clearer it becomes this is an abrupt halt, even if it were a bubble going off or finally the hype dying it wouldn't have happened just like that.

So I'm currently quite clueless, but your suspicion about most of the Ordinals activity being maintained by a few centralized actors (regardless of their intentions) could be true actually.

To be honest I don't plan to dig deeper and waste time on something that I'm not sure I want to know or have knowledge of but my initial thought on this decline was to blame it on something like that
- a large ordinal minting service going either offline or having troubles
- as you said, this whole activity might be actually faked by some groups who as you said are in control and faking both auction and trading activity to trick buyers into it and those who have taken a break, are probably not worth it anymore.

More than thinking his tweets and position have started this slowdown in minting one might have to ask if he and his business friends were not actually responsible for the entire activity, so rather than him making others stop it was just him deciding to stop.

[ABCbits]

Has anyone read Rodarmor blog about Runes (https://rodarmor.com/blog/runes/) which describe different ways to create NFT with less bloat?

--snip--

It is a way of creating "artificial scarcity" for the sake of assigning extra value to individual sats... Although there is some logic involved in the way the system has been created, most people don't recognize this value and they'd need to be onboarded -- similar to the way they are onboarded into NFTs or other collectibles, I suppose. Its just that the value proposition is particularly ridiculous because it didn't even exist a year ago.

Maybe if this type of Ordinals numbering system for satoshis sticks around in its current form for a decade, then it will be taken a bit more seriously... But its highly doubtful.

Yeah, so far most people don't about this "artificial scarcity". Otherwise, we might see miner/pool offer "uncommon" and "rare" rarity satoshi.

--snip--

Looks like during the entire minting timespan, there were 800k uncommon,  369 rare and 3 epic sats created (I'm willing to bet the one mythic is some kind of genesis transaction so it doesn't count).

It also seems that people are getting impatient with minting though, because the mempool size has been steadily deflating.

There's no need for bet or guess when it's already stated on the docs.

This gives us the following rarity levels:

    common: Any sat that is not the first sat of its block
    uncommon: The first sat of each block
    rare: The first sat of each difficulty adjustment period
    epic: The first sat of each halving epoch
    legendary: The first sat of each cycle
    mythic: The first sat of the genesis block

[nutildah]

Has anyone read Rodarmor blog about Runes (https://rodarmor.com/blog/runes/) which describe different ways to create NFT with less bloat?

Yeah this is all pretty dumb, but even he kind of admitted it by stating it was designed from a "worse is better" approach. He is basically re-creating Counterparty (XCP), but minus proper infrastructure and 9+ years of development. Its especially irritating because he called XCP a "shitcoin" earlier this year, but now he is trying to make a shitcoin factory on Bitcoin.

Ordinals sure was a weird thing. Mempool suggests the fad seems to have been completed now. Time to move on to Ethscriptions, lol.

Yeah, so far most people don't about this "artificial scarcity". Otherwise, we might see miner/pool offer "uncommon" and "rare" rarity satoshi.

I did see a couple of inscriptions minted on "uncommon sats" go for sale on OpenSea. This is made possible through a wrapping service called Emblem Vault that vaults an NFT on the bitcoin network and produces an Ethereum NFT, for the purpose of trading on markets like OpenSea. Only the NFT owner has access to the private key on the Bitcoin side of things. I looked for the sales on OpenSea but for the life of me I can't find them.

[Wind_FURY]

--snip--

Are the numbers supposed to be so important? Isn't the ART, and having to proof that you own it, that's supposed to be the most important?

Don't forget Ordinal handbook mention 6 type of satoshi rarity. I expect people who like lucky/good number also find it's important.

Current Supply

    common: 1.9 quadrillion
    uncommon: 808,262
    rare: 369
    epic: 3
    legendary: 0
    mythic: 1

That's precisely the point I was making, Casey Rodarmor's "6 types of rarity" came from a "hand book" that was merely made up from his own mind. It's not backed/secured by anything except the belief that it's real and therefore it must have value.

I found a blog that gives an ELI-5 about the numbering-system, why Casey wants to change it to a more unstable numbering system, and the drama.

https://trustmachines.co/blog/proposed-changes-to-the-bitcoin-ordinals-numbering-system/

[larry_vw_1955]

I found a blog that gives an ELI-5 about the numbering-system, why Casey wants to change it to a more unstable numbering system, and the drama.

sounds like a disaster for ordinals. a good example of poor planning in software engineering if you can call it software engineering at all. its more like a hack with some type of bug.

[Wind_FURY]

I found a blog that gives an ELI-5 about the numbering-system, why Casey wants to change it to a more unstable numbering system, and the drama.

sounds like a disaster for ordinals. a good example of poor planning in software engineering if you can call it software engineering at all. its more like a hack with some type of bug.

Probably, but Bitcoin also stumbled during its early days of development too, so perhaps it couldn't truly be considered a "disaster". Other developers could also learn from it, OR if Ordinals dies, Casey Rodarmor showed that developers don't need permission to build and develop something on top of Bitcoin. He just went to work and built something.
 
Plus another good thing about Ordinals or CounterParty being built on top of Bitcoin is as long as the Bitcoin network is running, you could download a client, input your keys, and your "artifacts" of dick pics and fart sounds will still be there.

[ABCbits]

Has anyone read Rodarmor blog about Runes (https://rodarmor.com/blog/runes/) which describe different ways to create NFT with less bloat?

Yeah this is all pretty dumb, but even he kind of admitted it by stating it was designed from a "worse is better" approach. He is basically re-creating Counterparty (XCP), but minus proper infrastructure and 9+ years of development. Its especially irritating because he called XCP a "shitcoin" earlier this year, but now he is trying to make a shitcoin factory on Bitcoin.

Doesn't Counterparty use address/proof of burn though? It seems Runes a bit better since it use OP_RETURN instead which doesn't bloat total UTXO.

Ordinals sure was a weird thing. Mempool suggests the fad seems to have been completed now. Time to move on to Ethscriptions, lol.

On technical level, it's weird to see Ethscriptions become popular since it doesn't even use ERC-721 (or similar ERC). Bitcoin doesn't support NFT/token by itself, while Ethereum user create it's own specification when it's already supported by the coin.

[larry_vw_1955]

Probably, but Bitcoin also stumbled during its early days of development too, so perhaps it couldn't truly be considered a "disaster". Other developers could also learn from it, OR if Ordinals dies, Casey Rodarmor showed that developers don't need permission to build and develop something on top of Bitcoin. He just went to work and built something.

i'm not sure i fully understand the technical issue that caused the numbering system to go wrong but the numbering system was kind of a fundamental way of identifying a particular ordinal most people probably thought of it as identifying their particular ordinal 100%. casey should have been smart enough to anticipate something like this though. if he designed it from the ground up...but i guess maybe he wasn't smart enough. and an unforeseen bug crept in. a shame they don't have a program that can catch those kinds of bugs.
 

Plus another good thing about Ordinals or CounterParty being built on top of Bitcoin is as long as the Bitcoin network is running, you could download a client, input your keys, and your "artifacts" of dick pics and fart sounds will still be there.

so how exactly is an ordinal identified now if not by inscription #. and are inscription #s still static or can they change with time? 

[vjudeu]

Doesn't Counterparty use address/proof of burn though?

I would rather say "Proof of Trap", because coins are not "burned". They are "trapped" on address 1CounterpartyXXXXXXXXXXXXXXXUWLpVr, that is unlikely to be spent, but mathematically, there should exist around 2^96 matching public keys. Just nobody knows them, as long as RIPEMD-160 is not broken.

However, if that hash function will ever be broken in the future, then guess what: those coins will move. When it comes to the real Proof of Burn, it should use OP_RETURN, or even better, it could be done by miners, by taking less coinbase reward. That last approach is the most space-efficient, because it is all about changing some uint64 into smaller value, so there is no additional output needed, as it is the case in OP_RETURN, where you can explicitly say "I had to create this txid:vout explicitly to burn those coins".

[nutildah]

Doesn't Counterparty use address/proof of burn though? It seems Runes a bit better since it use OP_RETURN instead which doesn't bloat total UTXO.

Proof of Burn refers to how XCP was created. During the burn period in Jan 2014, peeps who sent BTC to the "burn address" were rewarded with a proportional amount of XCP. And most Counterparty transaction data is stored in OP_RETURN, but occasionally it is also stored in other fields and multisig txs.

I would rather say "Proof of Trap", because coins are not "burned". They are "trapped" on address 1CounterpartyXXXXXXXXXXXXXXXUWLpVr

Its a terminology thing that has been around for almost 10 years now... You can't actually remove (circulating) BTC from the total supply, but by sending it to an address for which nobody knows the private key, it is just as good as "burned."

In 2018 or so, Counterparty implemented a "destroy" feature where tokens could be completely removed from circulation. Thus the distinction of the two terms, "burn" and "destroy."

However, if that hash function will ever be broken in the future, then guess what: those coins will move.

Bitcoin will have much bigger problems if/when this ever happens.

[ABCbits]

Doesn't Counterparty use address/proof of burn though?

I would rather say "Proof of Trap", because coins are not "burned". They are "trapped" on address
--snip--

You're correct. I simply use term that most people use.

Doesn't Counterparty use address/proof of burn though? It seems Runes a bit better since it use OP_RETURN instead which doesn't bloat total UTXO.

Proof of Burn refers to how XCP was created. During the burn period in Jan 2014, peeps who sent BTC to the "burn address" were rewarded with a proportional amount of XCP. And most Counterparty transaction data is stored in OP_RETURN, but occasionally it is also stored in other fields and multisig txs.

You're right, should've re-read the protocol specification[1] earlier. It's rather ugly to store the data on address and multisig public key though.

[1] https://github.com/CounterpartyXCP/Documentation/blob/master/Developers/protocol_specification.md