Gold collapsing. Bitcoin UP.

[smooth]

If anything, we should get the blocksize to the largest rationally supportable size

This is exactly where reasonable people differ.

That is a key word right there "reasonable".  I certainly haven't had any great epiphany as to how to choose the correct size or how to make it a reliably updated thing.   Predictability and anything that removes human consensus making from an ongoing process is what I would favor.   

The only way to remove human consensus from the ongoing process is to leave it exactly the way it is. No changes to the consensus rules ever (MP argument).

That is btw pretty much Wladimir's view. He's not going to back any change that doesn't have human consensus. So you either have human consensus or, removing it from the process, no changes at all.

Is this your position too?

No, I've already said I'm in favor of an immediate modest increase to 2-3 MB (and generally only keeping up with hardware technology unless there are other changes to the software that improve its ability to scale decentralized, which so far there have not been)

Which then leaves Monero as a lifeboat where I recall some recent post about it having massive scaling capacity, no 1MB rubbish for them!

The Monero scaling is somewhat similar to BIP100. I consider it very much an open question how that will work out (perhaps okay, perhaps quite poorly), and I wouldn't want to impose something so unproven on Bitcoin.

[I_bitcoin]

You are never going to get human consensus on an unlimited block size to the point where it isn't controversial. Just isn't going to happen.

The good news is that Bitcoin is not software.   It is a protocol  .   Of course you have to get people to agree to use your fork but that is another thing entirely.  Errr, consensus?

[nby]

You are never going to get human consensus on an unlimited block size to the point where it isn't controversial. Just isn't going to happen.

The good news is that Bitcoin is not software.   It is a protocol  .   Of course you have to get people to agree to use your fork but that is another thing entirely.  Errr, consensus?

The problem is that currently Bitcoin (the protocol) is conflated with Bitcoin Core (the software). Until this state of things changes we are de facto running a centralized system, with veto power in the hands of a few selected ones. And this is what bothers me the most.

[smooth]

You are never going to get human consensus on an unlimited block size to the point where it isn't controversial. Just isn't going to happen.

The good news is that Bitcoin is not software.   It is a protocol  .   Of course you have to get people to agree to use your fork but that is another thing entirely.  Errr, consensus?

The problem is that currently Bitcoin (the protocol) is conflated with Bitcoin Core (the software). Until this state of things changes we are de facto running a centralized system, with veto power in the hands of a few selected ones. And this is what bothers me the most.

It is not a protocol that is amenable to successful modification by a marketplace of different implementations doing different things (with respect to the consensus rules). The result of that is chaos and likely failure.

Perhaps it is possible to build such a protocol, and some ideas along those lines have been discussed here and elsewhere, but there is a lot more to that than just multiple implementations.

Then again the argument could be made that we already have such a protocol/marketplace and it is called altcoins. But if you want to stick with Bitcoin itself, you need more to successfully make changes without human consensus.

[solex]

No, I've already said I'm in favor of an immediate modest increase to 2-3 MB (and generally only keeping up with hardware technology unless there are other changes to the software that improve its ability to scale decentralized, which so far there have not been)

Good. I am fine with that as well as that is a perfectly sensible position.
It is also a compromise, which is why it is so dismaying to see Core Dev all but ignore Jeff's bare-minimum BIP 102. This is a problem, they are not offering any alternative themselves.

As mentioned before too: IBLT is a known change which can help a lot with decentralised scaling.

[iCEBREAKER]

I have heard this statement "tx fees are underpriced"  and "fee market is broken" several times and I still don't get it.    Mainly because the way it gets portrayed is that we must raise fees now vs. let things develop naturally otherwise miners can't stay in business.   When the block reward halves what do you think happens?   My expectation is that miners will still need to pay the bills in fiat so the fiat per BTC ratio will increase.    This increase, by it's very nature, increases the fiat "profit" per transaction fee.   So, if fiat/BTC increases due to reward changes the fees go up from the perspective of a fiat holder.   Heck, I suspect the block reward will be a major factor in driving fiat/BTC ratio until the block reward gets close to the average transaction fee.  .

If anything, we should get the blocksize to the largest rationally supportable size and allow nature to take it's course.   Fees, from a fiat perspective, will go up even if the miners do nothing to prioritize transactions by fee or increase the bitcoin denominated standard fee.    Miners will still make a profit as scarcity drives the fiat/btc ratio and the fee market will continue to develop as a means of preserving scarce resources and preventing spam.   It seems to have done an adequate job so far and this may be the best possible outcome with respect to keeping bitcoin globally relevant.   One of the biggest drivers of adoption we have at our disposal is the fiat/btc ratio. 

There must be something I have been missing in the arguments I see about this.   Where does the above thinking not work?

I'm less about raising fees "now" than I am about not suppressing, via centralized governance, fees now and forever.

Block subsides have taken Bitcon this far, we don't need to enhance the subsidies.  There is a line around the block for Satoshi's Better Mousetrap, so we don't need to undercharge nor promote with giveaways.

Good point about fiat/btc ratio driving adoption!  We always operate under the key assumption Bitcoin is and will continue growing.

You also already seem to get the fact that a ~5 cent BTC tx uses electricity equivalent to about 1.5 days worth of typical American home (ie $15-$20; cite: some post on reddit I can no longer find).  That's a good starting point.

What you are missing in order to get from there to my position is an understanding of (the implications of) the fully intentional very high difficulty of forking Bitcoin.  Even uncontroversial soft forks like BIP66 entail some degree of FUD/chaos/drama.  Therefore, it is not best practice to posit anything remotely like an easy, flexible, fine-grained path for us to "get the blocksize to the largest rationally supportable size and allow nature to take it's course" via consensus and subsequent commits.  Let's work with, leverage, and indeed celebrate, what we have (ossification at 1MB) and not make the actual the enemy of the perfect ("the largest rationally supportable size").  "Rationally" presupposes a purpose, and subjective individual interpretations of Bitcoin's Ultimate Purpose^TM differ wildly.   

Until block subsidies taper off, we only need (as in must) raise tx fees sufficient to exclude spam via prioritization.  But the spam exclusion/wallet improvement/RBF initiatives also present an excellent early opportunity for the rough contours of a fee market to emerge.  IMO, sudden massive blocksize increases waste that opportunity.

The last step to see my PoV is an appreciation of what sidechain/LN's Layer 2 will do to scale and thereby enhance Layer 1 tx.

As Frappuccino purchases move off main chain and are consolidated by Layer 2 processes, each of Layer 1's 7 tps becomes increasingly valuable, demonstrating it is not the size of the block that matters, but how you use it.

[NewLiberty]

If anything, we should get the blocksize to the largest rationally supportable size

This is exactly where reasonable people differ.

That is a key word right there "reasonable".  I certainly haven't had any great epiphany as to how to choose the correct size or how to make it a reliably updated thing.   Predictability and anything that removes human consensus making from an ongoing process is what I would favor.   

The only way to remove human consensus from the ongoing process is to leave it exactly the way it is. No changes to the consensus rules ever (MP argument).

That is btw pretty much Wladimir's view. He's not going to back any change that doesn't have human consensus. So you either have human consensus or, removing it from the process, no changes at all.

Is this your position too?

No, I've already said I'm in favor of an immediate modest increase to 2-3 MB (and generally only keeping up with hardware technology unless there are other changes to the software that improve its ability to scale decentralized, which so far there have not been)

Which then leaves Monero as a lifeboat where I recall some recent post about it having massive scaling capacity, no 1MB rubbish for them!

The Monero scaling is somewhat similar to BIP100. I consider it very much an open question how that will work out (perhaps okay, perhaps quite poorly), and I wouldn't want to impose something so unproven on Bitcoin.

This is pretty much where I am also.

Either:

• A 1 time increase, (without all this "I know the future and what will be needed and what we can do when" stuff).  This could be 2-3mb without significant additional risk, and would give a bit of time for other developments to be ready for a more long term change.
• BIP 100, after some more testing.  The monero method has been working for a good long while, but hasn't really had as much stress testing as would be nice to see.

Any hard fork presents an edge case where unusual risks are presented.  This is especially true when these hard forks are so rare.  There is no practiced resilience to the sorts of things that creative minds can cook up to exploit events. 

[thezerg]

You also already seem to get the fact that a ~5 cent BTC tx uses electricity equivalent to about 1.5 days worth of typical American home (ie $15-$20; cite: some post on reddit I can no longer find).  That's a good starting point.

This is a perfect example of your flawed reasoning.  This comment is analogous to claiming that a sailboat is incredibly inefficient because its energy use is equal to the total wind power available in the ocean divided by the number of sailboats.

Instead the volume of wind power (energy spent mining) is present for a reason that is completely orthogonal to sailboats (transactions).  It is proportional to the amount of energy being input into the system.  In the Bitcoin world, the "energy input" is the conversion of value into bitcoin representation (i.e. long term purchases).  When the block subsidy dies down, this energy input will be reflected in coin appreciation.  When the block subsidy ultimately approaches a negligible quantity, boats will (finally) have to turn on their motors and pay their way across the bitcoin network.

Additionally, you make the circular fallacy of defining the number of total boats allowed on the ocean, and therefore finding the ocean winds to be inefficient.

Ultimately there will be a fee market.  Miners will turn off hashing power until transactions appear that will make mining the block profitable.  We don't need to centrally force this to happen.  You are falling into the fallacy of thinking that others are forcing a lack of a fee market but in fact you are centrally forcing one to exist.  It is technically feasible to have larger blocks, and having them makes each transaction more efficient, so anything less than larger blocks is central planning and reduces efficiency.

And by forcing a fee market you are running the sails and the motors simultaneously, uselessly, and dangerously.

Why is it dangerous?

Money is defined as better by users fundamentally by its transfer efficiency.  This was hard to measure with fiat or PM currencies but it is simple with Bitcoin.  It is the ratio of the fee to the quantity transmitted.  Once the block subsidy diminishes this ratio will be nowhere near zero like it is today.  It is ironic that you ignore this because it is what fundamentally gave Bitcoin value in the early days while nobel laureate economists were insisting that Bitcoin would fail because it had "no intrinsic value".

Once the motor's (transaction's) true efficiency is not hidden by the wind (block subsidy), the most efficient (and accessible) cryptocurrency implementation will "win" just like gold "won" over silver.  Just like Apple is learning, getting a head start does not matter much when a huge percentage of the world simply cannot afford your product, so MUST choose a competitor.

[thezerg]

PS: this hard fork "danger" overrated, like Y2K.  We'll have tested moving across the limit over and over.  And if for some reason TSHTF, the very centralized mining pools will simply get together and fall back to the old code (unwinding blocks > 1MB), just like they did during the accidental hard fork a few months ago.

This is one great property of Bitcoin, since every node validates (processes) every block, you can't have a bug lying latent in a block that's (say) 500 deep in the chain before someone suddenly triggers it.  Half of the nodes will reject the new block RIGHT NOW and the other half will accept it.

[NewLiberty]

PS: this hard fork "danger" overrated, like Y2K.  We'll have tested moving across the limit over and over.  And if for some reason TSHTF, the very centralized mining pools will simply get together and fall back to the old code (unwinding blocks > 1MB), just like they did during the accidental hard fork a few months ago.

This is one great property of Bitcoin, since every node validates (processes) every block, you can't have a bug lying latent in a block that's (say) 500 deep in the chain before someone suddenly triggers it.  Half of the nodes will reject the new block RIGHT NOW and the other half will accept it.

Y2K was pretty easy, the problem was finite, scope driven, and easily prioritized.  Lots of things were just turned off and replaced (though the replacements weren't necessarily better, just maintainable).

There are a number of issues under development that will also reduce the hard fork risks, and block size increase risks.

[iCEBREAKER]

Is this your position too? Because the way Bitcoin is today with a protocol level transaction cap so small it is too crippled to ever become a significant part of the world economy. A cap too small even for LN to provide much scaling because LN needs the capability to close a lot of payment channels simultaneously. SC is not a scaling solution unless Pieter is wrong.
Which then leaves Monero as a lifeboat where I recall some recent post about it having massive scaling capacity, no 1MB rubbish for them!

Icebreaker is transparent like an ice-cube when it comes to campaigning for a crippled Bitcoin, clearly so he can execute a massive Monero put during a time of Bitcoin crisis and Monero pump, hoping to catch the reverse trade when the 1MB is finally fixed and quadruple his coinage accordingly.

Yes, that's right.  I joined Bitcointalk in mid-2011 as part of my sinister master plan of "campaigning for a crippled Bitcoin, so [I myself] can execute a massive Monero put during a time of Bitcoin crisis and Monero pump, hoping to catch the reverse trade when the 1MB is finally fixed and quadruple [my] coinage accordingly."

Who told you?  Was it (that fucking troll) smoothie?  

But seriously...you have no evidentiary basis on which to validly assert what is or is not "a protocol level transaction cap so small it is too crippled to ever become a significant part of the world economy."  That is an empirical question; we do not at this time know what sidechains/Lightning will do to relieve Layer 1 tx/fee pressure.

Talking out of your ass may impress Frap.doc, but the Team Core and I know better.

Didn't you learn your lesson about bullshitting when you spouted off some nonsense about 5 year deadlines for DoS regulator deprecation predicated upon an embarrassing dearth of expertise in Bitcoin 101?

[iCEBREAKER]

You also already seem to get the fact that a ~5 cent BTC tx uses electricity equivalent to about 1.5 days worth of typical American home (ie $15-$20; cite: some post on reddit I can no longer find).  That's a good starting point.

This is a perfect example of your flawed reasoning.  This comment is analogous to claiming that a sailboat is incredibly inefficient because its energy use is equal to the total wind power available in the ocean divided by the number of sailboats.

Instead the volume of wind power (energy spent mining) is present for a reason that is completely orthogonal to sailboats (transactions).  It is proportional to the amount of energy being input into the system.  In the Bitcoin world, the "energy input" is the conversion of value into bitcoin representation (i.e. long term purchases).  When the block subsidy dies down, this energy input will be reflected in coin appreciation.  When the block subsidy ultimately approaches a negligible quantity, boats will (finally) have to turn on their motors and pay their way across the bitcoin network.

Additionally, you make the circular fallacy of defining the number of total boats allowed on the ocean, and therefore finding the ocean winds to be inefficient.

Ultimately there will be a fee market.  Miners will turn off hashing power until transactions appear that will make mining the block profitable.  We don't need to centrally force this to happen.  You are falling into the fallacy of thinking that others are forcing a lack of a fee market but in fact you are centrally forcing one to exist.  It is technically feasible to have larger blocks, and having them makes each transaction more efficient, so anything less than larger blocks is central planning and reduces efficiency.

And by forcing a fee market you are running the sails and the motors simultaneously, uselessly, and dangerously.

Why is it dangerous?

Money is defined as better by users fundamentally by its transfer efficiency.  This was hard to measure with fiat or PM currencies but it is simple with Bitcoin.  It is the ratio of the fee to the quantity transmitted.  Once the block subsidy diminishes this ratio will be nowhere near zero like it is today.  It is ironic that you ignore this because it is what fundamentally gave Bitcoin value in the early days while nobel laureate economists were insisting that Bitcoin would fail because it had "no intrinsic value".

Once the motor's (transaction's) true efficiency is not hidden by the wind (block subsidy), the most efficient (and accessible) cryptocurrency implementation will "win" just like gold "won" over silver.  Just like Apple is learning, getting a head start does not matter much when a huge percentage of the world simply cannot afford your product, so MUST choose a competitor.

Money currency is defined as better by users fundamentally by its transfer efficiency.

fify.  Better: Money is defined as better by users fundamentally by its storage of value efficiency.

Please excuse the digression.  

I like the wind of subsidy in early tx's sails analogy.  But the recent appearance of "cosmic background spam" makes it outdated.

The problem is that sailboats are expensive to construct and operate, while Bitcoin tx are not.

If we are to think about Bitcoin in the long term, and given Bitcoin's reasonably critical economic mass of ~$5 bil, current temporary block subsidies can be discounted as we consider fundamental questions about how and when to change one of Satoshi's Holy Numbers.

[nby]

[...]
I like the wind of subsidy in early tx's sails analogy.  But the recent appearance of "cosmic background spam" makes it outdated.

The problem is that sailboats are expensive to construct and operate, while Bitcoin tx are not.
[...]

Yes, they're not expensive, especially the spam ones. And guess why is that so ? Because there's no room for those transactions in a 1MB block, and  whoever is sending them can just resend them over and over again without actually spending a dime on them.

[dnaleor]

As Frappuccino purchases move off main chain and are consolidated by Layer 2 processes, each of Layer 1's 7 tps becomes increasingly valuable, demonstrating it is not the size of the block that matters, but how you use it.

I LOL'ed

[sickpig]

If we are to think about Bitcoin in the long term, and given Bitcoin's reasonably critical economic mass of ~$5 bil, current temporary block subsidies can be discounted as we consider fundamental questions about how and when to change one of Satoshi's Holy Numbers.

To which holy number are you referring to?

One last question: what's your position on block size limit, never change it or change (in the way you like the most) in the future?

[thezerg]

Unfortunately for money to have store of value efficiency it must ultimately have value which with bitcoin rests again on its currency efficiency.

Miners could only store non dust UTXOs in easily accessible storage and ignore incoming txns that spend them unless the fee is worth the cost to look the UTXOs up. There are so many possibilities.   Your problem is that you are a central planner even tho you dont know it -- you are forcing a particular solution (expensive limited txns) onto the network as a whole.

[sickpig]

Increasing block size does add features (and/or bugs), in the form of higher tps and whatever concomitant other new auto/adaptive regulation mechanisms come along with the eventual solution.  100k max tx size is probably only the first such required adjustment.

In the very thread you linked gmaxwell said solution proposed by
gavin adopt an indirect and simplistic approach, to which gavin replied:

But I would REALLY hate myself if in ten years a future version of me was struggling to
get consensus to move away from some stupid 100,000 byte transaction size limit
I imposed to mitigate a potential DoS attack.

So I agree, a limit on sigops is the right way to go. And if that is being changed,
might as well accurately count exactly how many sigops a transaction actually
requires to be validated...

the bolded part really makes me laugh, though.

that said it seems that there won't be any cap to tx size for the moment.

[sickpig]

If anything, we should get the blocksize to the largest rationally supportable size

This is exactly where reasonable people differ.

That is a key word right there "reasonable".  I certainly haven't had any great epiphany as to how to choose the correct size or how to make it a reliably updated thing.   Predictability and anything that removes human consensus making from an ongoing process is what I would favor.  

The only way to remove human consensus from the ongoing process is to leave it exactly the way it is. No changes to the consensus rules ever (MP argument).

That is btw pretty much Wladimir's view. He's not going to back any change that doesn't have human consensus. So you either have human consensus or, removing it from the process, no changes at all.

i wonder what all those guys will say next year when the Blockstream guys want to change the code for SC's and LN if Gavin objects.

Sidechains can be done with a soft fork, which means all they need to do is sign up miners (and of course find customers who want to use the side chain). It doesn't require a global consensus.

But if SC, and also LN,  needs a block size increase in any case (source, source2), doesn't this mean that introducing SC implies an hard-fork?

[sgbett]

The only way to make software secure, reliable, and fast is to make it small. Fight Features. - Andy Tanenbaum 2004

Oh yeah? So adding complicated stuff like sidechains and lightning network is making bitcoin code smaller?

To the extent those build on top of Bitcoin they aren't adding features to it or making the the core code bigger. If they require changes to the core code or adding new features to it, that's a different matter that needs to be considered carefully.

But if the claim is that sidechains are the 'solution' to the 'problem'. Then you are saying they are part of bitcoin [the ecosystem] whether they are part of core or not.

Increasing block size does not add features *and* it 'solves' the 'problem'. Without introducing any other layers of complexity, or attempting to artificially manipulate the fee economy that is growing organically just as it was always intended.

What frightens me is that the whole thing seems to have turned into a pissing contest.

The fee market isn't a problem right now, the block size is. So why are people trying to pre-emptively fix the fee market in an ass backwards way to address the block size problem.

Nobody is claiming to have predetermined that sidechains (or LN) are necessarily the ENTIRETY of the solution to the scaling problem, only that they are potentially part of it.  Bitcoin is an ongoing bleeding-edge experiment, and we are working with intuition, educated guesses, hypotheses, and prototypes, not off-the-shelf kit in neat little boxes.

Increasing block size does add features (and/or bugs), in the form of higher tps and whatever concomitant other new auto/adaptive regulation mechanisms come along with the eventual solution.  100k max tx size is probably only the first such required adjustment.

And thus, the 100k max tx adjustment neatly destroys your claim that larger blocks do not introduce additional complexity.  More is different; the dose makes the poison...

The Red Queen interpretation, whereby we must change Bitcoin ASAP for the sake of keeping it the same, is absurd.  Moving the tx supply curve with the goal of controlling the range where it intersects that of demand is prima facie centralized market manipulation.  As long as tx fees are absurdly underpriced, in terms of their cost and what users are willing to pay, the fee market is completely broken.  Bitoin's 'free sample/loss leader' viral marketing campaign phase ended with the emergence of omnipresent 'cosmic background spam.'

The "pissing contest" which "frightens" your delicate sensibilities is exactly the "fight" to which Tannenbaum exhorts us, because the "adversarial process is valuable in assuring [features] do not compromise security or reliability."

Misrepresenting my position then arguing against that isn't going to cut it.

I said the problem is that the block size is too small, and that the solution is to make the block size bigger. That doing this requires no additional functionality. I contrasted this to the sidechain solution which does require additional functionality. As you rightly say nobody is claiming that is entirely the solution, but that is irrelevant, the point is that this or other solution(s) that require extra functionality are the very thing that your Tanenbaum quote warns against.

All that stuff you said about how we need to change TX size, thats some other thing. Either you are intentionally conflating the two, which is disingenuous or you really can't tell the difference, which I doubt is the case.

It looks to me like your emotional attachment to your position is causing your reasoning to become irrational. I don't think anyone's argument is absurd. I can see how enforcing higher fees benefits some parties, I think that misses the big picture which is that it *requires* additional functionality.

I'm not frightened of pissing contests, I think they are childish. You don't need to "fight" anything. As a smart human being we all need to listen, think and reason. Not inject hyperbole, and inflammatory language into posts to try and bully your opposition. Your argument should stand on its own merit and not the vehemence with which it is delivered.

[inca]

Unfortunately for money to have store of value efficiency it must ultimately have value which with bitcoin rests again on its currency efficiency.

Miners could only store non dust UTXOs in easily accessible storage and ignore incoming txns that spend them unless the fee is worth the cost to look the UTXOs up. There are so many possibilities.   Your problem is that you are a central planner even tho you dont know it -- you are forcing a particular solution (expensive limited txns) onto the network as a whole.

He refuses to understand where the speculative value of bitcoin lies, the decentralised cheap unlimited transactional nature of bitcoin coupled with first mover advantage and importantly the expectation of progressive adoptive waves of new users.

If bitcoin is crippled and day to day transactions are forced off chain then bitcoin ceases to maintain either it's attractiveness or advantages over competing usurper chains. It isn't hard to see a migration of new users to alternate chains or digital monetary media and a collapse in the exchange price of bitcoin if that scenario plays out.

Does the Core team really believe bitcoin will maintain it's speculative value if it continues to be a play thing for 10,000 people globally?