On Ordinals: Where do you stand?

[Hispo]

I am just afraid of the potential attack vector to the whole blockchain this may be.
I am not talking about some random guy minting 5 NFTs for fun. A government or even a hedge fund could use their economical power to spam the blockchain in some sort of denial of service attack, specially halting honest transactions of people who just wanted to send some satoshis at fair price...

Granted, there will be miners or pools that could see this as a problem and may ignore those NFTs transactions, but are we just going to leave the destiny of this project on the assumption of good will of some pools?

Are we going to wait until some hedge fund shorts Bitcoin and proceeds to spam us for days, while the FUD commences?

[1714489779]

1714489779

[1714489779]

1714489779

[1714489779]

1714489779

[1714489779]

1714489779

[pooya87]

Those parasites are earning money selling NFTs or whatever their spam business is, at the expense of everybody else's livelihood. Yes i take it very personal what Ordinals and their accomplices are doing.

This is the biggest problem with this type of spam. Unlike the 2017 massive spam attack campaign that had to be funded by the attackers and from their pockets (eg. DASH foundation paying for spams), as I said before this time they are providing an incentive for regular users/spammers to spam the blockchain. It is basically bitcoin infected by cryptokitties.

[franky1]

the only way to move an ordinal is to when spending value. copy/paste the deadweight data from one witness. and put it into a new witness in the next transaction

franky if that's true then it's even worse than we thought.   you would think the guy that designed this thing would have tried to prevent duplicating the image every time someone sold/transferred it. but i guess not?  

the claim that an ordinal/inscription is sitting in 1sat is wrong..

a sat(first sat/1sat) does not have >4mb of binary bits..
at binary
1 sat is "1" (1bit),  2sat is "10" (2bit),  3sat is "11" (2bit),
4sat is "100" (3bit),  5sat is "101" (3bit),  6sat is "110" (3bit),
7sat is "111" (3bit), 8sat is "1000" (4bit), 9sat is "1001" (4bit)
and so on

where by one byte(8bit) would be upto 256 sats
and just 4bytes of data meme would be 32bit which is upto 4294967295sat
(near 43btc)
no way in hell is anyone burning/using 43btc to represent a 4byte file
and there is not enough sats in circulation to represent a 4mb file

even if you were to lock up all 19m btc
1930000000000000sats
 110110110110101001100100100011010101010000000000000
|             |              |             |             |             |              |
is only 6bytes of file data

thus
ordinals/inscriptions are not sitting in the "value" pocket of a output
ordinals/inscriptions are not sitting as a 'sat'

they are in the signature/script/witness area
(choose the buzzword for whatever era of recognition you got into bitcoin)
it certainly isnt in the "output value" area(where sats sit)

[nutildah]

no nutilda

yes frank

I will spell out for you exactly what role UTXOs play in ordinal ownership since you refuse to do the research on your own:

https://docs.ordinals.com/guides/collecting.html

Ordinals can be sent to any bitcoin address, and are safe as long as the UTXO that contains them is not spent. However, if that wallet is then used to send bitcoin, it may select the UTXO containing the ordinal as an input, and send the inscription or spend it to fees.

Saying ordinal data is "not saved in UTXOs" is simply false as ordinal-containing UTXOs are the singular thing that is required to be in an address to transfer the ordinal from it.

and where the witness data of a tx.. is separate in the parent, does not cross over to the child in wallets like core or other native bitcoin wallet
... (followed by more arguing against points that were never made) ...

Nobody ever said that it does, nobody said any of these things are true and they are not part of how the system functions. If you would have actually read my first post in this thread instead of knee-jerk reacting to it, you would have learned the difference between ORDINALS and INSCRIPTIONS. Inscriptions make use of witness data whereas ordinals do not. Its not a matter of opinion, its a matter of understanding how the protocol works.

You're now just arguing for the sake of arguing. There's no point of continuing this conversation any further because every time you are proven wrong you just double down in your wrongness while acting like a dismissive dickhead, regardless of the subject being discussed.

I encourage you to actually LOOK at the components of the ordinal send transaction I outlined in my previous post before responding. You won't do it because if you were capable of honest comprehension you would have to admit you don't have the first clue what you're talking about.

[takuma sato]

Any blockspace that isn't used to prove how X BTC moved from A to B is a waste of space and thus an attack on Bitcoin by the various iterations of peoples that have tried to turn Bitcoin into some sort of Ethereum clusterfuck.

[franky1]

a UTXO is not txdata/deadweight data

a UTXO is TXID+index

a utxo does not mention sat amount when its put in as a input of a spending tx
a utxo does not mention the witness data of the previous transaction

please read raw tx data as to what goes into a child transaction when a parents contains a ordinal meme image

that meme is not in the child. just a txid and a index

..
also
ordinals are the meme images
inscriptions is a identifier

read your own links
am inscription is the script of describing what is to be put into the witness data. the ordinal is the image/meme

again both ordinal and inscription of ordinal. are not included in the UTXO set and dont get put into an input of the child spending the UTXO

PLEASE forget what you read on some promotional website.. and read real actual transaction data on the bitcoin network.

read the bits where he says incompatible wallets
as that is bitcoin and bitcoin wallets dont do as website says.
his software will do other things to GUI present information to appear linked by using his softwares tricks of copying one piece of data to another piece outside the default bitcoin rules

learn the difference between what you see at GUI (user interface) level. and what is actually seen at binary level of blockchain data and what difference for instance core does to a utxo. vs what caseys software does with utxo + deadweight +meta date

you are not simply spending sats/utxo to move a ordinal

if it were true then just using core you would move an ordinal by just spending a p2TR utxo
but it doesnt work that way

it requires caseys special software where the software doesnt just put in a utxo as input.. from the utxoset
it then seeks out the txid in the blockchain data files(.blk) finds the tx looks at the witness data, copies the witness data and puts it as witness data into the child tx

learn the difference

[franky1]

ok lets take things back to basics
to help you get unstuck

cases ordinal theory feature is a build up of 3 main sub deatures cmbined.
il just call them version 1ord,version 2ord,version 3ord
for simplicity
v1ord

Code:

# subsidy of block at given height
def subsidy(height):
  return 50 * 100_000_000 >> height // 210_000

# first ordinal of subsidy of block at given height
def first_ordinal(height):
  start = 0
  for height in range(height):
    start += subsidy(height)
  return start

this is before any bloat data concept(v3ord)

lets say he started the 'first ordinal height' of block 750,000(for easy math)

block 750,000 offering 6.25btc means the reward is the
first 625000000 sats
when the mining pool coinbase winner maakes its first transaction

bc1qblocreward 6.25 ->bc1qminera 0.00001000
                                   bc1qminerb 0.00001000
                                   bc1qminerc 0.00001000
                                   bc1qminerd 0.00001000
                                   (and so on)
bc1qminera is now ranked as owning the first 1000sat of ordinal rank
bc1qminera is now ranked as owning the 1001-2000sat of ordinal rank
(and so on)

bc1qminera 0.00001000->bc1qminera 0.00000001
                                       bc1qretailerX 0.00000375
                                       bc1qretailerY 0.00000375
                                       (249sat lost to fee)
bc1qminera is now ranked as owning the first 1sat of ordinal rank
bc1qretailerX is now ranked as owning the 2-376sat of ordinal rank
bc1qretailery is now ranked as owning the 377-751sat of ordinal rank                        

whereby its simply about ranking each sat
in caseys algo of ranking sats

where the sat in bv1qminera is rated as being the rarest sat as its the first.
(its a side show of rarity)
its not about putting in data into sats nor anything else its just his algo.
declaring certain sats as rare

v2ord
now to transfer the sat (knowing utxo dont retain sat data or transfer sats but destroy txid from utxoset to create new txid into utxoset) involves careful "utxo selection" to pick the rare one as the first input and also the first output
to keep its "rarity claim"
so whomevers address is in first poutputof child tx then owns that rare sat

EG if someone wants 2000sat from minera and minera wants to give them 1999 less rare +1 rare sat
bc1qminera 0.00000001->bc1qsomeone 0.00000001
bc1qminera 0.00002249    bc1qsomeone 0.00001999
                                       (250sat lost to fee)


whereby the first output being 1sat then is IN CASEYS THEORY AND SOFTWARE
treated as the same "first sat"

emphasis. this again is not about putting in any special data into blockchain
this is about the sat. being ranked as first by caseys ranking system.. not bitcoin protocol
which to then transfer the deadweight meme. is not just having the dead weight move by normal bitcoin code of making a transaction, but via caseys software uploading memes into a raw TX witness area

v3ord
now we get to the deadweight of memes

these memes are not put into the sat area of "value" of a raw tx
they are put into the "script/signature/witness area (pick the buzzword that appeals to how old school you want to interpret the area talked about)

the dead weight meme is put into the signature area(witness) of a tx. and if that tx has a rare sat. then that makes that deadweight meme rare

having an ordinal meme(deadweight image file) in the first sat does not mean its having an image in the bytes where sats sit in a tx
its having dead weight in the witness area of a transaction that also contains a rare sat(aka first sat)

GET IT YET

..
deadweight (memes) are not about sats..
sats are a separate algo, feature about ranking system of rarity

..
now thats cleared up
to transfer the meme requires putting the meme file into the witness of the next tx to tag it to the output that has been deemed by caseys algo as being the one that has the rare sat to ensure they stay linked as a rare meme

otherwise you dont own the meme you just own the rare sat if you were in position 1 of a taint of position 1's according to his theory

but outside his theory

the utxo in is a different txid as an output thus not the same

a true transfer would be where a file is hashed and then appended
with say (made up) bc1N[hash]<checksum>
which becomes an identifier of a file. or satoshi amount
which if output 1 this address and its designated output 2 is someones address that gets claimed ownership. then they can use that said address..
(but that requires changing bitcoin to allow that)
which core cant/wont

this someones utxo 5 childs away(unless dead data is added) has no reference in the tx to call it the owner.

apart from caseys software which is not a bitcoin rule but some guys theory

[Artemis3]

Every month i have been doing small transactions at 1 sat/b to pay for monthly utility bills, food, etc. The usual "waiting line" (memblock size) to wait was usually around 6mb which was like a couple of hours of wait at most. Today I'm trying a transaction, guess what? 61mb waiting line...

Those parasites are earning money selling NFTs or whatever their spam business is, at the expense of everybody else's livelihood. Yes i take it very personal what Ordinals and their accomplices are doing.

24 hours later and the transaction remains unconfirmed... Yes i could rise the fee, but this is not urgent and its a clear demonstration of whats going on... Of course everyone can rise the fee and then the spammers will do it too. It would be fine if those were normal transactions but this is pure spam, there is no other way to call non tx content in the blockchain, don't call it NFT or memes, the content itself doesn't matter, they are spammers pure and simple and that's how they should be called.

[d5000]

I am just afraid of the potential attack vector to the whole blockchain this may be.
[...]A government or even a hedge fund could use their economical power to spam the blockchain in some sort of denial of service attack, specially halting honest transactions of people who just wanted to send some satoshis at fair price...

I don't get why they shouldn't have spammed the Bitcoin blockchain in other ways. Data spamming is possible since Bitcoin exists (with "fake" addresses, later OP_RETURN etc.). But it's not even necessary if spamming is their main interest. They can simply send lots of "financial" transactions without sense, like the big blockers did (presumedly) in mid-2017. The "witness discount" exists since 2017, so all they have to do is to create big Segwit transactions with lots of inputs and outputs. I recently read on Twitter that with Taproot it's possible to achieve a 15% lower fee cost per data byte than with earlier methods, so it was not really a game-changer for pure spammers.

A long continuation of the Ordinals hype would be more concerning. But I expect the NFT hype in general to be over soon (it is already much less of an hype than 2021), and the Ordinals hype will also follow. People will see that if they inscribe more than a few pixels (i.e. higher resolution images or even videos) their transactions will be possibly more expensive than what they can expect to get as a price.

So I generally think people are over-reacting to Ordinals Inscriptions, even if I absolutely don't mind NFTs (at least those of the "pixel art" category, maybe there could be some valuable NFT use cases in the "legal contract" category). The Ethereum blockchain is also full of bloat and spam since at least the big ICO/DEX hype ~2017 ("people using decentralized exchanges to trade decentralized exchange tokens" is still my favourite   ) and it continues to work.

By the way, I was able to transact BTC these days for the same fee as I normally do (for convenience reasons I almost always use a little bit more than 1 sat/vbyte, between 1,5 and 3 sat/vbyte). I think expecting 1 sat/vbyte could not be realistic anymore in the near future, even without Ordinals.

PS: I have a slight hope that Ordinals could speed up the development of sidechain solutions on Bitcoin. There is for example a proposal for Rollups (a special kind of sidechain known from Ethereum).

[paid2]

I agree with you Artemis3. The fees have become really hard and it's very frustrating to think that it's because of this Ordinal mess.
We can still trade with small fees, fortunately, but you have to look at the mempool more than you used to, and I must admit that it's annoying some days.

Is it a will of the big mining pools? Do they just want heavier blocks with more fees and therefore more profits for them? One could imagine something like that.

In any case, if it goes on like this, I think that the Lightning Network will experience a new boom.  The Bitcoin blockchain is becoming less and less efficient nowadays, I hope like you that this NFT fad will pass. Developers should find a way to deal with this.
I can't even imagine the transaction fees for users of some CEXs that charge excessive withdrawal fees. If already when transactions at 4 sat/vB pass they exaggerate, I dare not imagine currently 

[franky1]

I am just afraid of the potential attack vector to the whole blockchain this may be.
[...]A government or even a hedge fund could use their economical power to spam the blockchain in some sort of denial of service attack, specially halting honest transactions of people who just wanted to send some satoshis at fair price...

I don't get why they shouldn't have spammed the Bitcoin blockchain in other ways. Data spamming is possible since Bitcoin exists (with "fake" addresses, later OP_RETURN etc.).

PS: I have a slight hope that Ordinals could speed up the development of sidechain solutions on Bitcoin. There is for example a proposal for Rollups (a special kind of sidechain known from Ethereum).

op return only allows 80 bytes
and thats not enough to put anything useful outside of a payment signature into that space
so no one used if for much else but a hash to then reference that hash on other networks

to put say 3000kb on the block chain was not possible pre tr
to put say 30kb on the block chain was not possible pre segwit
from memory the most it could have been was 10kb in segwit days and 80bytes per opcode in legacy days

forget for now about sidechains. as thats different teams and corporate business product stuff
bitcoin devs need to work on bitcoin stuff of plugging the holes that were made when the rules were softened over the years

make consensus hard again

..
current subnetworks (those beginning with L) are not functionally viable to handle peoples hoards, they are for the penny spenders. not the normal people that have 1st world country monthly-yearly salary amounts

yes new subnetworks will take their place that actually have real utility and are functional. but  for the moment bitcoin devs need to plug the holes of bitcoin and concentrate on bitcoin

[Artemis3]

By the way, I was able to transact BTC these days for the same fee as I normally do (for convenience reasons I almost always use a little bit more than 1 sat/vbyte, between 1,5 and 3 sat/vbyte). I think expecting 1 sat/vbyte could not be realistic anymore in the near future, even without Ordinals.

PS: I have a slight hope that Ordinals could speed up the development of sidechain solutions on Bitcoin. There is for example a proposal for Rollups (a special kind of sidechain known from Ethereum).

Yes, for now, but then 2 won't be enough. 3 won't be enough, and its a repeat of Jan 2018 (two weeks for 1 sat/b) all over again. Still waiting for that transaction to have an exact count of how many hours is taking, where the same until last week was just about two hours. What if Ordinals leaves and comes another and does the same? What if they don't leave and more join? Where does it end? "Use lightning, on chain sucks, use altcoin, bitcoin sucks" etc...

Is it a will of the big mining pools? Do they just want heavier blocks with more fees and therefore more profits for them? One could imagine something like that.

One name has been mentioned and its very involved with altcoins and some other finance activities, so it may not be coincidence.

[d5000]

Yes, for now, but then 2 won't be enough. 3 won't be enough, and its a repeat of Jan 2018 (two weeks for 1 sat/b) all over again.

Then also the Inscription/Ordinal guys will have to pay more for their trash to be included. With transactions of well over 1 kB, paying 2 (or worse, 3 or 4) instead of 1 sat/vByte can already be a big difference. Most of these NFTs probably won't sell for more than $10 or $20, so if you have to pay $5 in fees (and well over $10-20 for anything closer to "high-resolution") this will already make many people think if it's worth it.

And the 2017/2018 blockchain bloat event was much, much worse than now. 2 weeks for an 1 sat/byte tx? This was illusory then. I think it was about 2 months (look at the All view of Johoe's mempool graph, I see there no possibility for an 1 sat/vByte tx for most of the early quarter of 2018. We're not even close to the situation in the first half of 2021, when the bloat was simply caused by a bullish hype.)

I'm following the evolution of Inscription transactions and the worst may be already over. Februrary 8/9 was the peak with over 20000 inscriptions in 24 hs., now the number has lowered to 6000-7000. It's a novelty, so people are trying it out, but I'm almost sure it will go down to less than 1000 inscriptions per day soon, and also with a focus on less "heavy" transactions.

Anyway I've proposed the Namecoin guys to promote Ordinals on their chain, as it's Bitcoin merged-mined (and thus it can be described as an "extension" of Bitcoin), and it's focussed on data/non-financial transactions anyway. They seem to be not uninterested, but Taproot is still not live on their chain.

What if Ordinals leaves and comes another and does the same? What if they don't leave and more join? Where does it end? "Use lightning, on chain sucks, use altcoin, bitcoin sucks" etc...

Again, it's a free market, if the hype continues the "inscribers" will have also to pay more fees. 2017/18 was also not the end of Bitcoin at all

[Artemis3]

And yet my current transaction is still pending. I did one such transaction in early Jan 2018 which is why i remember, it may have come to two months by late Jan/Feb like you say.

And don't forget, if those things start selling higher to account for the fee, they won't mind increasing it. this is a foreign competition to actual transactions within the blockchain. People were already complaining about Bitcoin capacity even before this, and while you think it "may be over", i think we have seen nothing yet, and leaving the door open for spam abuse is a mistake.

Sure if they go to Namecoin its fine for now, but this can still be abused by others in the future, and they did for us a perfect demonstration of it.

[BADecker]

Bitcoin is somewhat of a dinosaur. Sure, it's been upgraded here and there, through forks and internal controls by the Devs. But it's a dinosaur because it isn't easily understood or used by the general public.

The thing that Ordinals might do is break Bitcoin just enough so that somebody will make something that is practical for public use.

[d5000]

And yet my current transaction is still pending.

Did you use RBF? I think 1 sat/vbyte without RBF is quite risky, at all times. At any moment, a surge in activity can happen and delay the tx for weeks or even months, like it happened not only 2017/18 but also in mid-2019, late 2020 and the whole first half of 2021.

Edit, having read an earlier post of yours in this thread: Fees are now slightly below 2 sat/vbyte. So my tip for you would be to RBF (or if not possible, double spend) the transaction ASAP to 2 sat/vbyte. Tomorrow (working day) this fee level will be difficult to achieve.

leaving the door open for spam abuse is a mistake.

How would you "close the door"? If there was a convincing idea which does not rely on centralization in any way, I'd be happy to support it ... But I think there is no trivial solution.

One could make Taproot more restrictive again, lowering the 520-bytes limit for data pushes, and limit these to one per output, but that's not wanted by the devs because they want to allow complex smart contracts. And anyway: without Taproot, with older methods, like Peter Todd demonstrated, data packages like the Ordinals inscriptions would pay only 15% more than with the current method.

A solution discussed by the devs in their mailing list is to allow higher byte counts in OP_RETURN outputs to make it the most convenient method. An OP_RETURN-based NFT technique like the Rare Pepes on Counterparty, would have some slight advantages to Ordinals.

Trying to make miners and nodes not relaying transactions with inscriptions, like I read earlier in this thread? That's not fault-proof at all. It could make waiting times for them a little bit longer, but there will always be miners happy to include these transactions to harvest the fees.

At the end, I think rollups/sidechains (like drivechains) are the best long term solutions, but they will take a while to materialize. For the time being I think the best strategy is to support a "relocation" of the Ordinals project to another chain like NMC, LTC, Doge or GRS ...

[franky1]

@d5000
the main promise of taproot was that sigspam of TR would only appear as 1 signature length

so why need an open door of >3.99mb full weight limit

as for op_return
to have an identifier you only need space for a 256bit hash (32bytes) so why extend past 80bytes

I'm following the evolution of Inscription transactions and the worst may be already over. Februrary 8/9 was the peak with over 20000 inscriptions in 24 hs., now the number has lowered to 6000-7000. It's a novelty, so people are trying it out, but I'm almost sure it will go down to less than 1000 inscriptions per day soon, and also with a focus on less "heavy" transactions.

even 1000 a day of just 30kb small memes is alot

there are only 144 blocks a day average so you are saying 1000 is acceptable even at a low 30kb each
that means you expect ~7 per block meaning 0.2mb per block of bloat
(turning a average 1.3mb block of 2000tx into a 1.5mb block of 2007)

which could have been utilised by instead another ~325 average p2p payment tx's
(turning a average 1.5mb block of 2325tx )
also
if consensus was hardened with proper opcode byte limits and the cludge of the base vs weight code changed to just being a true legacy 4mb block where segwit and tx are side by side as legacy sharing the same space with no special premiumising of certain formats and miscounting of others
where bytes are counted as bytes again,,
then that could allow ~6000 normal average p2p payments for the 4mb
at the good utility expectation. rather then the current soft consensus that allows 1tx for 4mb extreme at the other end of the scale
..
i do find it funny how people thought straight forward p2p payment formats like legacy(nothing special) need to be charged 4x but silly new formats that do not solely function as p2p payments but do bloaty non p2p payment stuff are miscounted and not 4x thus get to bloat for cheap

[larry_vw_1955]

In any case, if it goes on like this, I think that the Lightning Network will experience a new boom.  The Bitcoin blockchain is becoming less and less efficient nowadays, I hope like you that this NFT fad will pass.

they already got over 60,000 "nfts" stored on bitcoin now. seems like every day thousands more get added.   at some point the millionth one will be added and so on...

i do find it funny how people thought straight forward p2p payment formats like legacy(nothing special) need to be charged 4x but silly new formats that do not solely function as p2p payments but do bloaty non p2p payment stuff are miscounted and not 4x thus get to bloat for cheap

that's why i always said franky that if you mess with something simple and try and add more complexity to it (aka features) this is the exactly the type of thing that happens. unintended consequences. you may even call them BUGS. unfortunately fixing bugs is alot harder and inconvenient to do once you've put your code into production and the world is using it. especially with bitcoin. 

[d5000]

the main promise of taproot was that sigspam of TR would only appear as 1 signature length

so why need an open door of >3.99mb full weight limit

I also don't know the reasons, and my understanding of Taproot is currently limited. I would also be more comfortable with a stricter limit, but for some reason the Core team didn't want that.

as for op_return
to have an identifier you only need space for a 256bit hash (32bytes) so why extend past 80bytes

OP_RETURN afaik would allow to create a NFT system similar to the current Ordinals Inscriptions but with less UTXO set bloat, for example based on Counterparty, because there would be no need to store any UTXOs for the ownership management (you basically can avoid one additional UTXO per NFT being created). OP_RETURN can also be fully pruned, while afaik the Taproot script can't because it's needed for validation (I may be wrong here though). The effect on block space itself would however be negligible.

The whole point thus would be to make OP_RETURN the most convenient means to store data.

even 1000 a day of just 30kb small memes is alot

Yep, it would still take significant block space. But it would be a number which is currently managable by the blockchain, i.e. at weekends and often even at some hours in working days the fee would be allowed again to go down to 1 sat/vbyte.

Average block size before the Ordinals boom was about 1.2 - 1.5 MB, and 200 kB more would not have mattered much in current "crypto winter" conditions (currently the average size, with full blocks due to Ordinals, is about 2,3 MB). That would of course change in a full-scale bull market like in 2021, so I hope the ordinals boom will be over/relocate to other chains once the next hype begins (or sidechains become operational).

I also didn't say I consider this number "acceptable". It's only the number I expect it to go down to in a couple of weeks. Ideally of course it would be even much less.

[Hispo]

-snip-

-snip-

Even if the popularity of NFT's continues to fade away, this new boom in your main chain could lead to people having to pay higher fees. I would like to point out that Bitcoin adoption is strong in countries like Venezuela, Argentina, Nigeria, etc. Places where it is not only about decentralization but also about saving money.

thanks to Bitcoin people can move dozens or even hundreds of dollars for cents, while other services like Paypal would stripe away many dollars from us, living in countries where each bill counts, that is friction is not good. I have been in the same place where Artemis3 is, using 1 sat/vbyte (specially at night), so I am aware that situation is frustrating.

The only positive thing I can see about this in the mid term is the possibility of developers coming with a good solution to improve scalability. Letting people to mint whatever they want and at the same time, do not spamming the mempool beyond recognition.