Removing OP_return limits seems like a huge mistake

[Wind_FURY]

This BIP-110 supporter actually said what we have been saying about dick pics/fart sounds and the fee market. It's laughable how they debate against the Core Developers without understanding themselves.



He's admitting that we're right.

That's fine, but jpegs are going to get priced out anyway. They're not a threat to the network
https://x.com/mattkratter/status/1732812053122064795

This however might not the case forever. We cannot just hope ordinal craziness will simply go away.

  ¯\_(ツ)_/¯

Spam prevention is the one of the main functions of the fee market.

Sure, but this is quite a simplified view on it and is contrary to what Bitcoin is supposed to be. We can't expect it to be able to act as a digital peer to peer cash and simultaneously only rely on fees to price out undesirable behaviors.

I have already posted/debated about my viewpoint on that many times in BitcoinTalk. I am not going to keep repeating myself like a broken record. BUT if your viewpoint is that Bitcoin is, "Peer To Peer Cash" because "It's in the white-paper", then OK. You do your own sort of thinking.

Plus if I'm going to nit-pick because I believe that the "because white-paper" debate is a nit-pick, then I'll tell you, technically Bitcoin isn't a peer to peer network. It's a broadcast network.

[gmaxwell]

Sure, but this is quite a simplified view on it and is contrary to what Bitcoin is supposed to be. We can't expect it to be able to act as a digital peer to peer cash and simultaneously only rely on fees to price out undesirable behaviors.

Some people have some internal definition of what "digital cash" is and can't tell the difference between their own meaning of a term and reality.  Imagine someone howling that bitcoin has to be made of paper because the white paper uses the word cash and cash is paper-- except that complaint is too obvious, and those people didn't make it that far because they first got hung up on the white paper itself not being paper.

Bitcoin is what it is, the analogies used in the white paper are to help people generally understand what it's talking about.  They're not an imposition of a definition and are absolutely not a imposition of a very specific definition that exists only in your head.  Bitcoin has always had fees as part of the design, and fees have always been the proposed mechanism to fund security absent inflation.  Cash by any conventional usage doesn't have fees, one of its most important differences compared to other forms of payment.

Oh no!  Satoshi says bitcoin is cash, but cash doesn't have fees, so Bitcoin can't be cash, so if Satoshi is right then satoshi is wrong which makes him right which makes him wrong ... Can god make a rock so heavy that even god can't lift it??!?!    What happens when an unstoppable force meets an unmovable object?



Dog Pig Dog Pig Dog Pig loaf of bread S̷̖̒͊̔̎͐̈́̆͌͌̏͊͝͝Y̴̢̜̪̪͍̫͇͈̦̭̣̘̣̤͍̖̆̉̑̓͒̈́͌͐̿̈́̚͜͠S̷̢̛͈̰̙̳̟͓̟̲̰̟̥̎̒̈́́̀͐̌̊̍͗̿̾͆͝ͅT̷̫͛̇̏̀̏͘É̸̡̡̺̦̫̘̯͍̆͗̀͒̊͆́̉͌͜M̵̡̢̝̲̯͍̖͕̰̳̮̖͍̬͐͑̐̂͑͗̈́̅͐͂͑͘͝ ̷̧̢͓͎̤̘̻̩͕̞͉̰̼̻̇̇͊̈́͝E̷̳̜̰̠͎͔͂̽̑͐́͐͑̽̍̾̕͜Ŕ̶̡͕̝͔̠̺͙̻̲̖̈́͆̒̊̈̽̈́R̸̤̋͒̑̽̓̈́͑̚Ǫ̷̮̝͔͇̻̬̥͇͍͓͂̀Ŗ̷̨̧̹̹̠̪̩̫̰͓͙̹̀́ͅ.



Your head will only stop exploding when you realize overreading meaning into other peoples writing that was not ever intended and would never have made sense.  Bitcoin is peer to peer electronic cash.  Does that mean it needs to behave exactly the same as some other cash system (which is not peer to peer, or not electronic, or just not Bitcoin) or even behave particularly similarly? Not at all.  Terminology games are for idiots.  Bitcoin just needs to be Bitcoin, it needs to be useful-- and it is.  On day one anyone could have said "hey wait a minute, you're saying this has fees as part of the design, how can it be cash?"  But no one was saying that then and so it makes even less sense to raise it now. The feedback of online discussion must have created idiocy today more profound than existed in 2008.  Satoshi had to refer to existing concepts to teach people about Bitcoin was but in doing so he wasn't making any deep, subtle, or overly prescriptive remarks-- his audience was people who had no idea of what Bitcoin was or that it was even possible.  They just had to get the general idea.  Satoshi did not write that Bitcoin is a "functionally identical online replica of the argentenian peso", he simply said it's a peer to peer electronic cash.  He needed to say that or otherwise some readers might think he was writing about tacos or something.

Bitcoin's nature creates some limitations-- e.g. such as floods of high fee transactions can delay your transactions if you're not willing to outbid them.  I think it's still useful in spite of those limitations.  Somehow before Bitcoin we managed to survive since time immemorial with person to person money that couldn't even be sent online at all.  Compared to that I think that irreversibility when sending online having potential delays isn't a particularly serious limit.  In particular because the mechanisms that people have used to be assured that payment would be completed can still also be used when Bitcoin itself isn't settling quickly.

[stwenhao]

On day one anyone could have said "hey wait a minute, you're saying this has fees as part of the design, how can it be cash?"

On day one, many transactions were free, and it was standard to pay zero satoshi, to send some coins. And you can probably remember times, when 0.01 BTC fees appeared, and how some people reacted to that, how some tried to avoid it, build an alternative client without it, and so on.

I don't think the threshold should ever be 0.  We should always allow at least some free transactions.

I wonder, if it will ever be implemented in that way, as the current default implementation provides no room for free transactions (even if they can fit into a block). Because now it seems there will always be at least one satoshi per transaction (when fees will decrease from 1 sat/vB or 0.1 sat/vB into 0.001 sat/vB or similar). And even if technically there will be some transactions with zero fees, then still: they will be just a part of a bigger package, where some transaction with zero fee will be pushed to the block, by some CPFP transaction with non-zero fees, and they will form a package.

[Wind_FURY]

On day one anyone could have said "hey wait a minute, you're saying this has fees as part of the design, how can it be cash?"

On day one, many transactions were free, and it was standard to pay zero satoshi, to send some coins.

And you can probably remember times, when 0.01 BTC fees appeared, and how some people reacted to that, how some tried to avoid it, build an alternative client without it, and so on.

I don't think the threshold should ever be 0.  We should always allow at least some free transactions.

I wonder, if it will ever be implemented in that way, as the current default implementation provides no room for free transactions (even if they can fit into a block). Because now it seems there will always be at least one satoshi per transaction (when fees will decrease from 1 sat/vB or 0.1 sat/vB into 0.001 sat/vB or similar). And even if technically there will be some transactions with zero fees, then still: they will be just a part of a bigger package, where some transaction with zero fee will be pushed to the block, by some CPFP transaction with non-zero fees, and they will form a package.

OK, then in that "Zero-Fee Decentralized Bitcoin Network", how do you effectively create a spam-protection mechanism? Filter and hope?

In my personal opinion, if Satoshi came back he'll probably change some of his/her viewpoints about Bitcoin, especially after learning how the network is working in practice.

This is the first time the world has ever had something like Bitcoin. Satoshi probably didn't actually have all of the answers. The smartest developers in the Bitcoin community are probably still learning as we all are also still learning.

[stwenhao]

how do you effectively create a spam-protection mechanism? Filter and hope?

Proof of Work is a quite good filter. Today, you can pay only with fees. But it is possible to allow lower fees, if you provide more Proof of Work instead. And I guess there is some equilibrium somewhere, where users could accept paying N satoshis fee, with M difficulty. And then, you can decide, if you want to provide more coins, or more hashrate, to get your transaction confirmed.

Also, as long as OP_CHECKSIG, or its equivalent, is used almost everywhere, enforcing it is quite easy: the smaller your DER signature is, the more Proof of Work is needed to make it. Which also makes the whole transaction smaller, and can save you some satoshis on fees.

So, it is even possible now, to some extent. The main problem is that doesn't contribute to the whole network's chainwork. But well, if someone is against Merged Mining, then it can be used as it is: if you require a signature smaller than N bytes, then everyone, who will want to double-spend it, will need to re-mine it, by solving the same Proof of Work challenge from scratch.

Also, we can look at some numbers, to estimate, how many hashes are needed, if there is a lot of competition:

Code:

hash=00000000000000000000eaa0b9031bf81c368df29bb62a27794f7d7a38308f36
difficulty=1701f303
target=00000000000000000001f3030000000000000000000000000000000000000000
satoshis=314736509=0x12c27f7d
target*satoshis=000000000000249156c825770000000000000000000000000000000000000000

And then, some users will never meet 0x1701f303 difficulty. But more of them could meet 0x1a249156 difficulty instead, which could be a price for lowering your fee by one satoshi, if there will be many people, trying to do the same thing.

Which means, that if fees will be high, then it may be worth grinding your signatures, and making some satoshis out of it, by sending your transaction with the same fee rate, which could take less bytes.

[Satofan44]

Sure, but this is quite a simplified view on it and is contrary to what Bitcoin is supposed to be. We can't expect it to be able to act as a digital peer to peer cash and simultaneously only rely on fees to price out undesirable behaviors.

Some people have some internal definition of what "digital cash" is and can't tell the difference between their own meaning of a term and reality.  Imagine someone howling that bitcoin has to be made of paper because the white paper uses the word cash and cash is paper-- except that complaint is too obvious, and those people didn't make it that far because they first got hung up on the white paper itself not being paper.

I don't have a strict definition of what that is, I was using terminology that is familiar to most readers. You can call it many other things that it it usually called or bring up many others functions that it is supposed to fulfill, and a similar point can be made since it impacts a lot of these things.

Bitcoin's nature creates some limitations-- e.g. such as floods of high fee transactions can delay your transactions if you're not willing to outbid them.  I think it's still useful in spite of those limitations.  Somehow before Bitcoin we managed to survive since time immemorial with person to person money that couldn't even be sent online at all.  Compared to that I think that irreversibility when sending online having potential delays isn't a particularly serious limit.  In particular because the mechanisms that people have used to be assured that payment would be completed can still also be used when Bitcoin itself isn't settling quickly.

The conclusion is, nothing at all can be done or should be done to improve this aspect (I am not arguing for any particular idea)? That's not a very forward way of thinking on topic of usability, is it? It is a bit strange to see this response, perhaps we don't understand each other's intent clearly -- when LN itself is an improvement on the issue that I am talking about (and many improvements to LN itself are as well) -- it just does not work if you are not already in LN and the chain is congested. So why would we stop there or not want more? In essence we have made huge progress for certain scenarios, but for others we are stuck in the same place where we were at the time of spam attack a decade ago or so.

[gmaxwell]

If you have a solution-- by all means suggest it.  I think it's fundamental:  the resource needs to be limit in supply to control operating costs and to drive income for security.  Free market action turns that into a market.  If someone is outbidding what you're willing to pay that is a problem for you but it's also a success for the system.

Essentially this is just repeating the age old arguments against free markets-- they don't necessarily optimize for human flourishing to the extent a ideal spherical dictator in simple harmonic motion might be conjectured to do.   But we find that e.g. communist regimes tend to create appalling living conditions when they're not busy commit massing murder on unimaginable scale ... while dirty amoral capitalism has reliably produced the best living conditions ever known to humanity in spite of the fact that it regularly screws up some stuff.

Part of the issue is that there already are other solutions-- people just want to ignore them perhaps because stuff that works is boring.  You mention lightning but it's also perfectly possible for altruistic miners to skip over higher fee paying transactions and prioritize stuff they think is good for the welfare of the system.   And that works fine without coercive consensus changes or mob action that deprives people of individual choice.

And of course this also gets ignored because all this gnashing of teeth is over a problem which is currently inactive.   Knots supporters are a bored immune system that has turned on the host because Bitcoin's success has for the moment robbed them of any real enemies they're capable and willing to fight.

[Wind_FURY]

how do you effectively create a spam-protection mechanism? Filter and hope?

Proof of Work is a quite good filter. Today, you can pay only with fees. But it is possible to allow lower fees, if you provide more Proof of Work instead. And I guess there is some equilibrium somewhere, where users could accept paying N satoshis fee, with M difficulty. And then, you can decide, if you want to provide more coins, or more hashrate, to get your transaction confirmed.

OK, but the cost of spam-protection must come from somewhere. Why don't we choose the simplest path and let users pay fees in Bitcoin instead? I believe the miners would appreciate it greatly, and for them to keep giving security for the network.

Although for your idea, you probably should start your own topic for it. It could be something good to learn from the discussions that could come from it.

  👍