Why the fuck did Satoshi implement the 1 MB blocksize limit?

[AGD]

How did you come up with that number? Are you dumb enough to pay $30 while everybody else paying $5? (current fee for a max. 1 hour tx is 100sat/b)

If yes (it seems like a yes), then you are not clever enough for this shit. Stay with credit cards.

I like a man with attitude so wait a few seconds and I will check the charts and be right back to you

Nope it's $15 according to this https://bitinfocharts.com/comparison/bitcoin-transactionfees.html
and was about $30 a week ago when I last looked but I will reduce it to $25 to take in to account
the $55 it was at. Ripple fees are so low that they don't show up on most charts so if $5.00 makes you feel smart then jobs
a good one.

45 sat/b
https://blockchain.info/tx/8c189782b4eeb4074674b1c009b12edfde8ef4264733485c7627cfc8b75106f4

Explain this 1$ transaction.

[1715194637]

1715194637

[1715194637]

1715194637

[1715194637]

1715194637

[Ucy]

http://satoshi.nakamotoinstitute.org/emails/cryptography/2/p

Simple answer:

Satoshi never knew crypto will be this important.

Nope.  Read his e-mail of November 2008:

http://satoshi.nakamotoinstitute.org/emails/cryptography/2/

He understood perfectly how it would go.

The only thing he missed, was the collaboration of mining into pools.  He didn't realize that people could pool together, and sell part of their hashing effort into a fraction of a block reward, so in his opinion, there would have been more individual miner nodes.

That said, mining itself is automatically limited to a finite number of sensible participants.  It wouldn't make sense to be 1 million independent mining nodes (the only nodes Satoshi considered, were mining nodes: he never considered non-mining full nodes as meaningful - and he was right).   The reason that it doesn't make much sense to have 1 million mining nodes, is the fact that there are only 52 600 blocks mined per year.  

Even if there were 1 million mining nodes with all having exactly the same hash rate, that would mean that *on average* a mining node would win ONE SINGLE BLOCK every 20 years.  But given statistical fluctuations, you might as well mine for 50 years and never have a block.  Or you might mine for 3 years and be lucky and find a block.

Given that all capital has a power law distribution, and never a uniform distribution, even if you would only have 5000 mining nodes, where, with a uniform distribution, each node would, on average, mine 10 blocks per year, the power law distribution would give most blocks to the 1000 upper class miners, and almost never a block to each of the 4000 others.  

So it would, in any case, in Satoshi's mind, not have been meaningful, statistically, to have many thousands of nodes.

The strategy of pooling together has of course strongly amplified this effect, and he wasn't aware of that.  In a certain way, mining pools add a small sniff of decentralization to the, in any case, obvious centralization that the economies of scale would induce in the competition for mining.  If mining pools weren't possible, we would quite possibly simply have 5 or 10 mining nodes, period.  Who can, effectively, still compete in *solo mining* as Satoshi saw it ?  The industrial proportions of mining make solo mining only available to a big company.  You'd have Intel vs Motorola vs Nvidia vs AMD or something.  Now, with mining pools, even with a modest mining investment, you can still contribute.  However, the "voting power" of a mining-subcontractor is very small: he can just go to another mining pool.  But it is better than nothing.

What mail was he responding to here?  http://satoshi.nakamotoinstitute.org/emails/cryptography/2/
 
Can I get a link.

[DooMAD]

What mail was he responding to here?  http://satoshi.nakamotoinstitute.org/emails/cryptography/2/
 
Can I get a link.

You only get Satoshi's replies, not the emails that were sent to him/them/whatever.  Keep in mind these are emails, which aren't usually publicly accessible at all.  The email Satoshi replied to on that link is quoted in the body, though.  You have to pay attention to the > and >> marks to see who said what.  You might find more details on the early cryptography mailing lists, but that's down to you and your "googe-fu".

[shield132]

Okay okay, before you rush and say "To prevent spam attacks!!", please wait and read this whole thread.

^The above argument is what I hear all the time. However, there is something not quite right about that reasoning. It doesn't make sense.

Let's start with a little background info...

Satoshi implemented the 1 MB blocksize limit without telling anyone; He just did it randomly. There was no discussion beforehand and after he did it, he did not mention it anywhere. People had to look at the code/use it to see the change. The mannerism in which the 1 MB blocksize limit was added is already strange in itself and as soon as it was done, debates/arguments among the community started happening.

Satoshi never told people that the 1 MB limit was to prevent spam - its just what everyone inferred.

Okay... history lesson over.

Now here is why it doesnt make sense:

The process of a transaction getting confirmed and added to the blockchain goes like this:

1) Tx. is broadcast with a custom fee
2) Tx. is added to the mempool
3) Miner collects tx. from the mempool (usually they will pic tx. based on which has the highest fee and work their way down from there)
4) Miner adds tx. to their block
5) Miner calculates the proof of work
6) Miner publishes block to the blockchain

Okay now we have the process outlined we can analyse the miners incentives/behaviour. I'll be using game theory to explain this and here is where it gets interesting.

The miners main goal is to make profit. This is why he adds tx. to his block in the first place (it allows for more fees and thus, more profits). So we can assume that without the blocksize limit, the miners would add infinite tx. to their block right? WRONG!

Allow me to explain:

Look at step 3.. Collecting the tx. from the mempool and adding it to their block takes a set amount of time and the longer that the miner spends collecting the tx. and adding it to their block, the less time they can spend calculating the proof of work - thereby giving their competition (other miners) the edge. The miners would naturally (based on game theory) find a nash equilibrium between collecting as many tx. as possible and finding enough time to calculate the proof of work in order to give them the maximum profitability. Thus, we can assume, that without a blocksize limit (infinite), the block size would stay relatively the same.

This is why an infinite block size limit is not an issue. I honestly cannot understand why he added the 1 MB limit.
Can someone please, please explain? I have been pondering this for over a month now. Thanks.

This is the first time this year I see an interesting thread/post, thanks for that.
1MB can be considered for preventing spam. Maybe he didn't think if bitcoin could become so popular and also there aren't some things considered like what will happen after mining of last block.
Also if you say that it's because of miners to get much profit, then we can proudly say that satoshi's idea was being a miner and he may be owner of one of the biggest mining companies and that may be the reason of his quiet.
And if his aim was much profit by mining, then why did he decided to give a try to reward halving? Maybe for price rising and he decided to cover that loss by filling it with high tx fees. I have no other idea...

[Anti-Cen]

This is the first time this year I see an interesting thread/post, thanks for that.
1MB can be considered for preventing spam. Maybe he didn't think if bitcoin could become so popular and also there aren't some things considered like what will happen after

Mining using CPU-Wars was and is a total joke and others have proved it and transactions on other coins like ETH don't cost
a tenth of what we are forced to pay today and competition is good but not good by the time you have 20,000 and if you plot the
fees against the mempool then before now when it was at 130,000 we have paid fees of $40 and now we are down to about $15

Clearly something does not add up and something is very wrong if more than 1,000 miners are needed for Bitcoin when it's not
even hosting "Smart Contracts" for other coins and they also say "Lightning Network" bring smart contracts to Bitcoin so where
do I publish my scripts to then if this is indeed the case

[RGBKey]

This is the first time this year I see an interesting thread/post, thanks for that.
1MB can be considered for preventing spam. Maybe he didn't think if bitcoin could become so popular and also there aren't some things considered like what will happen after

...if more than 1,000 miners are needed for Bitcoin...

No amount of miners are needed for Bitcoin. Difficulty scales with the hashrate. The amount of miners is a result of people thinking Bitcoin has value.

[AGD]

This is the first time this year I see an interesting thread/post, thanks for that.
1MB can be considered for preventing spam. Maybe he didn't think if bitcoin could become so popular and also there aren't some things considered like what will happen after

Mining using CPU-Wars was and is a total joke and others have proved it and transactions on other coins like ETH don't cost
a tenth of what we are forced to pay today and competition is good but not good by the time you have 20,000 and if you plot the
fees against the mempool then before now when it was at 130,000 we have paid fees of $40 and now we are down to about $15

Clearly something does not add up and something is very wrong if more than 1,000 miners are needed for Bitcoin when it's not
even hosting "Smart Contracts" for other coins and they also say "Lightning Network" bring smart contracts to Bitcoin so where
do I publish my scripts to then if this is indeed the case

45 sat/b
https://blockchain.info/tx/8c189782b4eeb4074674b1c009b12edfde8ef4264733485c7627cfc8b75106f4

Explain this 1$ transaction.

[dinofelis]

This is the first time this year I see an interesting thread/post, thanks for that.
1MB can be considered for preventing spam. Maybe he didn't think if bitcoin could become so popular and also there aren't some things considered like what will happen after

...if more than 1,000 miners are needed for Bitcoin...

No amount of miners are needed for Bitcoin. Difficulty scales with the hashrate. The amount of miners is a result of people thinking Bitcoin has value.

The "amount of miners" are simply the number of decentralized participants in the consensus decision.  Today, that number is something like 3 or 4.  There are 3 or 4 deciders on the consensus decision: the 3 or 4 mining pools that have majority hash power under their decision.  About 10 deciders, even though they cannot go against the tacit agreement of the 3 or 4 majority deciders, make up the essence of bitcoin's consensus deciding power.

To illustrate what this means: suppose that you are the owner of coins in an address that the 3 or 4 mining deciders are angry about.  If these 3 or 4 deciders really don't want you to transact, they can refuse your transaction (no matter what fee you pay).  If these 3 or 4 deciders also decide that they won't mine on top of another block that contains your transaction, mined by a smaller pool, then they will orphan these blocks.  By the time that the other significant mining pools, not understanding why their earlier block is orphaned, get the message, they will also avoid putting your transaction in their block (even if they don't have anything against you), in order for them not to waste their blocks: they can't win against the majority hash rate that will always orphan their blocks faster than they can construct on it.

The actual miners selling their hash rate will never know.  The block chain that is built is perfectly OK.  Your transaction remains in the pool.  No miner decides to include it, because 3 or 4 entities decided that your coins were excluded from the system.  Their consensus decision is in agreement with bitcoin's protocol.  No "full node in Joe's basement" will alert Joe that something is fishy.

The power of consensus is in the hands of 3 or 4 entities.

Let us see what this means.  Suppose 3 or 4 entities collude and are pissed against a single exchange.  They know that exchange's hot and cold wallets.  They decide that no transaction from one of these wallets is going to make it in the block chain.  If they are serious about it, they will never include one of these transactions, and will orphan every block that does.   People will realize that they cannot withdraw from that exchange.  They will think there's something quite fishy with that exchange.  The transactions will never be confirmed.  They will see the transactions in the mem pool, and never in the chain. 

[cellard]

So... we will never understand what his true intentions were. As far as I know, what we know is having full nodes in the hands of a few corporations only would centralize the system at layer 0 (BCash). If LN centralizes or not, it's yet to be seen. I trust no one.

The dispersed network theory where even if everyone kept open two channels instead of just one because these things
are expensive in miner fees has been blown out the water and it would be like trying to use the Tor network with 300 relay nodes
and that's if you could workout the routing.

I contend that having banking hubs is centralization and was a bank to go down then people would broadcast a message
on the Bitcoin network en-mass and the BTC block-chain at just 7 transactions per second could not cope with demand and unless all
the LN internal channels are able to close down gracefully then chaos would ensue.

In the lighting Network white paper they even talk about "Spending accounts" and "Saving Accounts" to be used for security
and reading whats shown below I can even image insurance brokers being built into the network to reduce risk

If one wishes to broadcast the transaction chain instead of agreeing
to do a Funding Close or replacement with a new Commitment Transaction,
one would communicate with the third party and broadcast the chain to the
blockchain. If the counterparty refuses the notice from the third party to
cooperate, the penalty is rewarded to the non-cooperative party. In most
instances, participants may be indifferent to the transaction fees in the event
of an uncooperative counterparty.

Yes sure, here you go miners have another $30 on me and just where is that button in my
wallet to contest transaction since it is built into the protocol or will the banking nodes all
be running much more sophisticated software which put paid to the saying "Anyone can be a bank"


  

You can think whatever you want about the lightning network, what is clear is that "nodes-as-big-as-needed" (in other words, BCH's approach, as admitted by Roger Ver) is not a real long term solution, and it will end badly, and it will end badly at layer 0.

Whoever is drinking the koolaid that the system would remain as decentralized with a handful of nodes can't be giving lectures on the supposed centralization of the lightning network.

[DooMAD]

The "amount of miners" are simply the number of decentralized participants in the consensus decision.  Today, that number is something like 3 or 4.  There are 3 or 4 deciders on the consensus decision: the 3 or 4 mining pools that have majority hash power under their decision.  About 10 deciders, even though they cannot go against the tacit agreement of the 3 or 4 majority deciders, make up the essence of bitcoin's consensus deciding power.

(...)

The power of consensus is in the hands of 3 or 4 entities.

Let us see what this means.  Suppose 3 or 4 entities collude and are pissed against a single exchange.  They know that exchange's hot and cold wallets.  They decide that no transaction from one of these wallets is going to make it in the block chain.  If they are serious about it, they will never include one of these transactions, and will orphan every block that does.   People will realize that they cannot withdraw from that exchange.  They will think there's something quite fishy with that exchange.  The transactions will never be confirmed.  They will see the transactions in the mem pool, and never in the chain.  

Right, so when Bitcoin advocates say things like:

Bitcoin gives us, for the first time, a way for one person to transfer money to another person anywhere in the world, so that the transfer is guaranteed to be safe and secure, everyone knows that the transfer has taken place, and nobody can challenge or block the transfer.

You're saying that's actually false because these "3 or 4 entities" can block whatever they want and the rest of the network is powerless to prevent it?  In my view, the miners would only get away with that if the rest of the network agrees with that action.  If the majority of non-mining nodes disagree - and there's an incredibly strong likelihood they would disagree, because blocking transactions is categorically not part of Bitcoin's ethos or underlying principles - you can expect that a change of algorithm would be promptly forthcoming and there would be many expensive bricks that used to do something profitable left to rot on a shelf somewhere.  Not to mention that the users of that exchange would be innocent bystanders and wouldn't deserve to have their withdrawals blocked like that.  There would be uproar if miners pulled a stunt like that.  So no, "3 or 4 entities" do not have "consensus deciding power".  Miners might decide what transactions go in their blocks, but users decide which blockchain they freely choose to follow.  If the users aren't accepting those blocks, no one cares who they did or didn't include.

Whoever says miners have all the power is wrong; whoever says users have all the power is wrong; whoever says developers have all the power is wrong.  It should be more than self-evident by now that's not how this shit works.  It all exists in equilibrium.  No one group can act unilaterally if the others don't agree, unless one of them is prepared to fork off or be forked off.

[Destrodream]

I love how people think they are smarter then Satoshi

[dinofelis]

You can think whatever you want about the lightning network, what is clear is that "nodes-as-big-as-needed" (in other words, BCH's approach, as admitted by Roger Ver) is not a real long term solution, and it will end badly, and it will end badly at layer 0.

Whoever is drinking the koolaid that the system would remain as decentralized with a handful of nodes can't be giving lectures on the supposed centralization of the lightning network.

Point is, bitcoin isn't decentralized.  It is centralized in the code, it is centralized in the mining, if we understand by centralized: a small oligarchy has most of the decision power.  As I pointed out several times in this thread, one should have several copies of the unique block chain out there.  However, there is no need to have thousands of them.  If there are sufficient copies around so that there's always one that is accessible, that's good enough for the cryptographic verification of *authenticity*.   The data burden is large in the hypothesis that every user needs a full copy, but the data burden is in fact very small as compared to other public data repositories.   The availability of thousands of identical copies only plays a very minor role in the so-called decentralization of bitcoin, because the actual decision power is in the hands of a small oligarchy of coders on a single repository (Github), and the few deciders who run the most important mining pools (less than you have fingers).  The true "verification" of the whole system needs to be done by those who want to convince themselves, but here, we have a contradiction.  What value has the fact that a piece of software that you didn't even READ tells you that some unique data set of which you can verify the cryptographic authenticity (that is, this is not a fake copy of the "original", it is the sole original), has not been constructed "correctly" according to rules in a piece of software you didn't check yourself ?  Software you got from a much more centralized entity than the system you're trying to verify ?

Needing thousands of identical verifications by a piece of centralized software, of a big piece of data of which, in any case, there's no other version around, is not what I'd call "decentralisation".  It is waste of resources.  Now, as long as one can afford waste of resources, why not ?  It doesn't hurt.  But if this waste of resources is taken as an argument to tell us that the system cannot scale, that's ridiculous.

In fact, bitcoin is not truly decentralized, but that's not needed.  From the moment that bitcoin is working correctly, that's good enough.  Decentralization is a method, not a goal.  Given that bitcoin is now a speculative asset that is in any case highly sensitive to belief, it has in any case lost the robustness for which true, weaponized decentralization would make a difference.  From the moment that bitcoin would be so much under attack (by governments for instance) that it would NEED true decentralization, in any case its market value would plummet so hard that it would break down as a store of value.  Most of its market value now comes from trading it on centralized exchanges.  The very rumour of a government cracking down on centralized exchanges somewhere, makes its value plummet.  

So, in reality, if a few mining pools keep one another in check, and if there are a few tens of independent repositories around the world, and if a few people read the code on the centralized monopoly of its code, and if some geeks check the validity of the block chain, that's good enough for bitcoin to work correctly.   And that's all its users actually need: that it works correctly.  It can be nice to have the illusion that you have an unbreakable ledger, with a totally decentralized piece of software, so that even if the First Order cracks down on bitcoin, there will be the Resistance that has still a few copies of the block chain left, with a piece of authentic code.  But by that time, a bitcoin trades for half a hamburger of yesterday.    All this are nice illusions.  They are not true.  But even if they are not true, bitcoin works.  There are no double spendings.  You can do transactions.  That's good enough.  

Another point is that bitcoin has been compared, already by Satoshi, to VISA.  However, bitcoin is not a currency, and never will be.  Its monetary theory behind it makes it fail as a currency.  Bitcoin is a highly speculative asset.  Of course, it has some liquidity which permits you to transact ; but its volatility (which is part of its design) makes that it is not suited as a currency.   So, no, bitcoin will not need VISA-like transaction volumes.  Because almost nobody will use it as a day-to-day currency, like we use VISA.  Simply because it is so volatile.  Bitcoin is a financial asset with high volatility.   Like risky financial products (say, Apple options).    There's no need to go to very high volumes.

All this makes that we are putting technical limits to achieve fake goals.  We don't need every Joe to have his copy of the block chain.  A few tens of copies that are publicly available is good enough - in any case, the true power structure of bitcoin is already so much centralized, that this essentially useless form of "decentralization" doesn't add much.  And we don't need to aim for true currency applications, because in any case the asset is ill-designed to become a true currency: its domain of application is in the speculative finance sector.

[dinofelis]

Right, so when Bitcoin advocates say things like:

Bitcoin gives us, for the first time, a way for one person to transfer money to another person anywhere in the world, so that the transfer is guaranteed to be safe and secure, everyone knows that the transfer has taken place, and nobody can challenge or block the transfer.

You're saying that's actually false because these "3 or 4 entities" can block whatever they want and the rest of the network is powerless to prevent it?

Yes, by design.  They can.  They won't mostly.

 In my view, the miners would only get away with that if the rest of the network agrees with that action.  If the majority of non-mining nodes disagree - and there's an incredibly strong likelihood they would disagree, because blocking transactions is categorically not part of Bitcoin's ethos or underlying principles - you can expect that a change of algorithm would be promptly forthcoming

Ha. Another algorithm.  If core writes it quickly.  Core, the other oligarchy.  And if exchanges list that new version as "bitcoin" and unlists the original.  The third oligarchy.  If core doesn't move its ass, and if exchanges continue to list the original coin with the original algorithm, as the thing they call "bitcoin", you can always dream of a change.  What are you, "decentralized user with your node that is happy", going to do ?    After all, miners are free to build the blocks they want.  They are free to build upon the blocks they want.  That's the consensus mechanism in bitcoin.  They even played by bitcoin's rules.  Your expensive non-mining node wouldn't see anything fishy. It simply turns out that the 4 most important mining pools decide not to include transactions from user X or exchange X, and decide not to mine on blocks that do so.  That's the consensus by proof of work.  Changing algorithm is not evident, because suddenly, there wouldn't be an army of miners, ready to provide enough PoW with the new algorithm.  From one day to another, there wouldn't be enough different hardware available.  What algorithm ?  Etherhash ?  Scrypt ?  Cryptonight ? ...  How to sell this in the bitcoin narrative ?  What if different forks appear ?

The *real* reason why this won't happen, is simply the market.  This is why strong decentralisation is not very important.   Most of these oligarchs are in bitcoin and need the market to appreciate it.  So they most probably wouldn't do something that breaks the market entirely.  Although, with the recent possibility to short bitcoin naked with futures in cash, things may be less evident.

and there would be many expensive bricks that used to do something profitable left to rot on a shelf somewhere.  Not to mention that the users of that exchange would be innocent bystanders and wouldn't deserve to have their withdrawals blocked like that.  There would be uproar if miners pulled a stunt like that.  So no, "3 or 4 entities" do not have "consensus deciding power".  Miners might decide what transactions go in their blocks, but users decide which blockchain they freely choose to follow.  If the users aren't accepting those blocks, no one cares who they did or didn't include.

In an industrial proof of work system, that blockchain needs to be made first, before you can decide to follow it.  For that, you need someone to code it (if it isn't Core, it most probably will not be called "bitcoin") ; if it remains a PoW coin, you still need an industrial amount of mining hardware to mine it ; and you need exchanges to list it.  

My point is that if these power structures, which are oligarchies simply because there are only a few deciders in each of them, keep one another in check, that's good enough, and that's what will keep bitcoin running.  But it has not much to do with decentralization.  And if these entities collude over certain aspects, they will dictate their will.  But in the end, they are kept in check by the market, as long as they are looking for gains.

Whoever says miners have all the power is wrong; whoever says users have all the power is wrong; whoever says developers have all the power is wrong.  It should be more than self-evident by now that's not how this shit works.  It all exists in equilibrium.  No one group can act unilaterally if the others don't agree, unless one of them is prepared to fork off or be forked off.

Absolutely.  But this is why this talk about decentralisation is bogus.  You have a few oligarchies that keep one another in check.  Like everywhere else.  You have the miners (the industry).  They are essentially 3 or 4 to decide.   You have Core.  They are a handful to decide.  You have the important exchanges.  They are a few tens to decide.  And then you have all the users that gamble on its tokens.  They are the market that finance the whole circus.  And that's the power structure of bitcoin.  Like most other financial ecosystems.  Not much "decentralized".  Samsung is also depending on its customers.  Apple also.  Bitcoin's industry also.  That's what keeps this up and running.  Not "decentralization".  Because it isn't.

So, like Apple could prevent you from having an iphone, but won't, the mining pools could prevent you from transacting, but won't.  Most of the time.  Unless they have a good reason to.  There's not much of a difference.

If Facebook pisses off most of its users because it behaves "unethically" towards some of them, or if the mining oligarchy pisses off most bitcoin users because they behave "unethically" towards some of them, that will have grave consequences in the market for both of them, and that's why Facebook is nice to its users, and the miner oligarchy is nice to bitcoin users.  In general.  That's why it works.  Not because of the myth of decentralization.  Even though it makes for a good story to tell at the camp fire.

[DooMAD]

You're saying that's actually false because these "3 or 4 entities" can block whatever they want and the rest of the network is powerless to prevent it?

Yes, by design.  They can.  They won't mostly.

I think many would argue that 3 or 4 entities making all the decisions was definitely not part of the design.  You seem to be advocating that it would be better if miners were in full control of consensus, but clearly most people aren't going to buy into that vision.  Any suggestion that implies a central controlling oligarchy is preferable will be met with resistance.  Which is precisely why the overwhelming consensus shows that maintaining a large number of non-mining full nodes is still the plan going forward.  

I'd concede that having thousands of non-mining nodes clearly wasn't part the original design either.  But there was no way to foresee the advent of ASICs and the way mining in general became as centralised as it did, at the rapid pace as it did.  So the compromise is the thousands of volunteers validating what has been put into the blocks by the centralised miners.  You might claim they aren't needed to keep the miners honest, but that's not something that can be proven empirically unless everyone is prepared to take a risk and start trusting our new supposed overlords, which, again, runs wholly contrary to the ethos and underlying principles of Bitcoin.  Ain't happening.

Your expensive non-mining node wouldn't see anything fishy. It simply turns out that the 4 most important mining pools decide not to include transactions from user X or exchange X, and decide not to mine on blocks that do so.  That's the consensus by proof of work.

Yeah, I'm not saying it's like you get a popup alert or anything, but people do have eyes.  When lots of people notice their withdrawals haven't confirmed and it becomes one of those big issues where everyone piles in with an opinion, as customarily happens on the internet, someone's going to notice and figure it out.  The more people who can instantly check and see that the transactions weren't included, the faster the community can react.  Hence why more nodes are better than fewer.

Changing algorithm is not evident, because suddenly, there wouldn't be an army of miners, ready to provide enough PoW with the new algorithm.  From one day to another, there wouldn't be enough different hardware available.  What algorithm ?  Etherhash ?  Scrypt ?  Cryptonight ? ...  How to sell this in the bitcoin narrative ?  What if different forks appear ?

The *real* reason why this won't happen, is simply the market.  This is why strong decentralisation is not very important.   Most of these oligarchs are in bitcoin and need the market to appreciate it.  So they most probably wouldn't do something that breaks the market entirely.

I'm not suggesting it would be clean, but it's an option.  A somewhat nuclear option, granted.  But even if there was a period where the network had to suffer with a lack of hashrate, that's what difficulty adjustments are for.  And the more nodes who have a full copy of the blockchain, the stronger that nuclear option looks.  Hence why more nodes are better than fewer.

In an industrial proof of work system, that blockchain needs to be made first, before you can decide to follow it.  For that, you need someone to code it (if it isn't Core, it most probably will not be called "bitcoin") ; if it remains a PoW coin, you still need an industrial amount of mining hardware to mine it ; and you need exchanges to list it.  

My point is that if these power structures, which are oligarchies simply because there are only a few deciders in each of them, keep one another in check, that's good enough, and that's what will keep bitcoin running.  But it has not much to do with decentralization.  And if these entities collude over certain aspects, they will dictate their will.  But in the end, they are kept in check by the market, as long as they are looking for gains.

Yes, but the general idea is you don't want to make it easier for entities to collude and that's what people argue will happen with fewer nodes.  The greatest level of transparency, neutrality, trustlessness and permissionlessness is achieved through the blockchain being shared and updated in real time between as many users on the network as possible.  Hence why more nodes are better than fewer.

Whoever says miners have all the power is wrong; whoever says users have all the power is wrong; whoever says developers have all the power is wrong.  It should be more than self-evident by now that's not how this shit works.  It all exists in equilibrium.  No one group can act unilaterally if the others don't agree, unless one of them is prepared to fork off or be forked off.

Absolutely.  But this is why this talk about decentralisation is bogus.  You have a few oligarchies that keep one another in check.  Like everywhere else.  You have the miners (the industry).  They are essentially 3 or 4 to decide.   You have Core.  They are a handful to decide.  You have the important exchanges.  They are a few tens to decide.  And then you have all the users that gamble on its tokens.  They are the market that finance the whole circus.  And that's the power structure of bitcoin.  

And the reason most users are prepared to gamble on its tokens is because it is decentralised.  If you take that away and 3 or 4 entities can make unilateral decisions without consequence, the vast majority of participants lose faith and the entire experiment fails.  I'm not prepared to gamble on Ripple's or Stellar's tokens because I don't think they're decentralised enough, although I do have to say even they sound more decentralised than your vision for Bitcoin.  So the "market" you keep talking about wouldn't be as strong as it is if we didn't believe the mere presence of this ever watchful eye of the thousands of nodes was keeping miners in check.  Hence (you guessed it) why more nodes are better than fewer.

[dinofelis]

You're saying that's actually false because these "3 or 4 entities" can block whatever they want and the rest of the network is powerless to prevent it?

Yes, by design.  They can.  They won't mostly.

I think many would argue that 3 or 4 entities making all the decisions was definitely not part of the design.  You seem to be advocating that it would be better if miners were in full control of consensus, but clearly most people aren't going to buy into that vision.

I agree that is was not part of the *indicated purpose* of the design, but it *was part* of the design, as we see it experimentally confirmed, and game-theoretically explained.  PoW always will evolve into an oligarchy, like any market does, but in this particular case, with no "product differentiation" and open, anonymous competition, it is evident that this would happen.  I'm not expressing any advocacy.  I'm just telling you what was the logical consequence of the actual design, confirmed in reality by the experiment. I think that Satoshi had more in mind "a few hundred" rather than "four".  That's the only difference between his vision and the outcome of the experiment. But obviously, he clearly stated that it wouldn't be decentralized in the way you think it is: that is, over a large portion of its participants.  Was Satoshi lying, was he simply wrong, or didn't it matter in his mind, we don't know of course.  But that's the design that he proposed, and that's what came out of it, with the caveat of less deciders than anticipated, because of the unanticipated formation of pools.

Any suggestion that implies a central controlling oligarchy is preferable will be met with resistance.  Which is precisely why the overwhelming consensus shows that maintaining a large number of non-mining full nodes is still the plan going forward.  

As I indicated many times, this is a joke, because non-mining nodes have no power.  Decentralization is about power, not about technical distribution over a geographical area.  And Satoshi explained from the start why non-mining nodes *shouldn't* have any power: because they are easily sybilled.  And that part of his design works well.  If it were possible for non-mining nodes to weight upon any decision, for no cost (as compared to mining costs), tomorrow, you swamp the network with hundreds of thousands of non-mining nodes and impose whatever you impose.  So it is fairly obvious that non-mining nodes have no power, bitcoin was designed for them to have no power, and bitcoin works correctly in that respect.  This is why this whole point is moot and such an obvious joke.  If non-mining nodes were to have power, Satoshi wouldn't have had to include proof of work.  He explains in his paper why you cannot base anything on "node count" because it is so simple to sybil it.  All the difficulty in the design of bitcoin was how to make it such that non-mining nodes DIDN'T have power.  

I'd concede that having thousands of non-mining nodes clearly wasn't part the original design either.  But there was no way to foresee the advent of ASICs and the way mining in general became as centralised as it did, at the rapid pace as it did.

How do you explain Satoshi's November 2008 phrase "farms of specialized hardware" then ?  How do you explain his use of the phrase "left to specialists" then ?  Of course it was evident to him.  He simply thought they would be somewhat more numerous than 3 or 4.  As I showed, you cannot reasonably expect many 1000 as the statistical lottery of winning a block becomes too fluctuating then.    If you introduce a power law distribution, like in any market, you'd arrive at best at a few 10 or 100.

 So the compromise is the thousands of volunteers validating what has been put into the blocks by the centralised miners.

They are not "validating".  They are only "checking".  That's the whole point. A full node only gives information to his owner whether the centralized miners did what he thought they were supposed to do.  If ever they didn't, his full node, nor an army of full nodes, is going to stop these miners from doing what they are doing.  And, to repeat myself, that's exactly how bitcoin was designed, to protect from a sybil attack on full nodes.  And Satoshi explains it in his paper.  

You only get information out of a full node, you don't do anything about the building of the block chain.  

 You might claim they aren't needed to keep the miners honest, but that's not something that can be proven empirically unless everyone is prepared to take a risk and start trusting our new supposed overlords, which, again, runs wholly contrary to the ethos and underlying principles of Bitcoin.  Ain't happening.
 

It happened already.  The whole story about full nodes is simply to obfuscate that hard-to-swallow fact.  This is what I'm explaining here: bitcoin IS centralized.  Bitcoin HAS a new oligarchy of overlords.  There is the oligarchy of miner pools ; there is the monopoly of code with the oligarchy of pushers on the github repository.  And there is the handful of exchanges.  These ARE the overlords of bitcoin.  And they keep one another in check, and they care about the market, like Tim Cooke cares about Apple customers.  Because that's where their power and money comes from, in the end.

Your expensive non-mining node wouldn't see anything fishy. It simply turns out that the 4 most important mining pools decide not to include transactions from user X or exchange X, and decide not to mine on blocks that do so.  That's the consensus by proof of work.

Yeah, I'm not saying it's like you get a popup alert or anything, but people do have eyes.  When lots of people notice their withdrawals haven't confirmed and it becomes one of those big issues where everyone piles in with an opinion, as customarily happens on the internet, someone's going to notice and figure it out.  The more people who can instantly check and see that the transactions weren't included, the faster the community can react.  Hence why more nodes are better than fewer.

What you are talking about, are the news feeds.    Coindesk, this forum, other centralized places of news.  That's the "eyes of the market".  That's where people get informed (or misinformed).  No "full node runners" would pay attention to it, because, as you say, there won't be pop-ups.   If you would like to *demonstrate* it here, you'd link to things on blockchain.info.  You wouldn't use your own full node for that.  The few geeks who are able to look at that using their full node wouldn't be believed as much as a link to blockchain.info.  

This is what really keeps those oligarchs, our overlords of bitcoin, in check: the buzz on internet, the market sentiment, their business.  Not the full nodes.  Most full node runners wouldn't even know where to look or what to do to see this.

Guess what ?  Exactly such an intermediate system is what Dash put in place: the master nodes.  The master nodes are a PoS layer between the user and the miners.  They will make a pre-confirmed mem-pool which the miners have to include.  But it is proof of stake.  There, it are the master node owners which are the overlords, but they are more of them.

I'm not suggesting it would be clean, but it's an option.  A somewhat nuclear option, granted.  But even if there was a period where the network had to suffer with a lack of hashrate, that's what difficulty adjustments are for.  And the more nodes who have a full copy of the blockchain, the stronger that nuclear option looks.  Hence why more nodes are better than fewer.

This is one of those make-believe stories again.  There is absolutely no strength in holding thousands of identical copies, because whatever that nuclear option is, it will be decided by a monopolist: Core, the overlord of bitcoin, because he has the brand name.  If not, there will be many forks.  And if that happens, how much time would it take ?  I cannot believe that they can do this in a few weeks time.  Hell, they cannot even roll out their pet LN in two years' time.  And the suspicion would be that they planned this in advance, with tons of specific hardware ready to do the new algo.  This story is of the type "I hide a big nasty dog in my basement, and if you rob me, I will release him".  "Our system is decentralized and robust against attacks, and if you dare, our Good Overlord in the basement will change the system the next week so that your attack fails."

Yes, but the general idea is you don't want to make it easier for entities to collude and that's what people argue will happen with fewer nodes.  The greatest level of transparency, neutrality, trustlessness and permissionlessness is achieved through the blockchain being shared and updated in real time between as many users on the network as possible.  Hence why more nodes are better than fewer.

As I have now outlined many different times, it doesn't make much of a difference, especially because bitcoin was designed not to leave power to non-mining nodes that are easily sybilled.  More nodes are very slightly better than fewer, if one doesn't introduce system limitations in order to have more of them.  Because the very very small added decentralization by having thousands of copies is not offset by the huge performance hit you introduce by taking the requirements for this into account, like limiting the amount of transactions that can be done.

And the reason most users are prepared to gamble on its tokens is because it is decentralised.  If you take that away and 3 or 4 entities can make unilateral decisions without consequence, the vast majority of participants lose faith and the entire experiment fails.

No, the reason why most users are prepared to gamble on its token is because they BELIEVE it is decentralized.  In fact, not even.  They believe that others believe that it is decentralized. Bitcoin is a Keynesian beauty contest.  It doesn't matter what you find best, it matters what you think others believe will find best.

This is  why this make-believe story of many nodes is necessary, to maintain that erroneous belief.  This is exactly like one needs to do human sacrifice for the Great God Dodo, simply to maintain the belief in the power of the Great God Dodo.  If the Great God Dodo wouldn't be able to require human sacrifice, the Great God Dodo wouldn't be powerful, wouldn't he ?

In the same way, if there's no necessity to have "many full nodes", it would be fairly obvious that bitcoin is centralized.  So one needs to propagate the myth that bitcoin is decentralized, and that Joe, with his node in his basement, is "contributing to the decentralization of bitcoin".  So that you can show that there are 11 000 Joes that all copy the same data set made by 10 pools out there, and tell people "look how decentralized our system is: there are 11 000 identical copies out there, and all of them have checked that everything is OK.  Power to the people !  Bitcoiners of the world, unite !".   While the fundamental design of bitcoin was exactly to avoid any power of these cheap nodes.

From Satoshi's white paper, third page:
"The proof-of-work also solves the problem of determining representation in majority decision
making. If the majority were based on one-IP-address-one-vote, it could be subverted by anyone
able to allocate many IPs."

This is where Satoshi says that full nodes shouldn't have power.

 I'm not prepared to gamble on Ripple's or Stellar's tokens because I don't think they're decentralised enough, although I do have to say even they sound more decentralised than your vision for Bitcoin.

I have no "vision for bitcoin", I'm telling you the factual, actual state of bitcoin's power structure, and how that actual state of bitcoin was already foreseen by Satoshi in November 2008, and why this is a simple logical consequence of his design.  Apart from having only 3-4 pools instead of a few hundreds of solo miners.  Yes, this goes against the narrative of bitcoin.  Satoshi wanted to make a system that works, for all practical purposes.  The decentralization narrative was good propaganda to make it going, but it doesn't correspond to reality.  When one is going to cripple the system in order to feed the narrative, one is making a big mistake.

You are expressing your "belief".  And that's what all these "decentralization stories" are about: make believe.  The Great God Dodo doesn't exist.

[AlexGR]

Satoshi never told people that the 1 MB limit was to prevent spam - its just what everyone inferred.[/i]

He did express his desire to both intentionally combat micro-tx spamming and to keep the blocks as low as possible in order to get adoption going.

Bitcoin isn't currently practical for very small micropayments.  Not for things like pay per search or per page view without an aggregating mechanism, not things needing to pay less than 0.01.  The dust spam limit is a first try at intentionally trying to prevent overly small micropayments like that.

...

It would be nice to keep the blk*.dat files small as long as we can.

The eventual solution will be to not care how big it gets.

But for now, while it's still small, it's nice to keep it small so new users can get going faster.

Satoshi cared about the wellbeing of bitcoin and he tried to avoid damaging circumstances, whether it was spamming, large blocks, or even ...wikileaks funding.

Basically, bring it on.  Let's encourage Wikileaks to use Bitcoins and I'm willing to face any risk or fallout from that act.

No, don't "bring it on".

The project needs to grow gradually so the software can be strengthened along the way.

I make this appeal to WikiLeaks not to try to use Bitcoin.  Bitcoin is a small beta community in its infancy.  You would not stand to get more than pocket change, and the heat you would bring would likely destroy us at this stage.

As you can see he knew that bitcoin wasn't 100% resilient against all kinds of attack vectors from the get-go, and acted accordingly.


Now regarding your other point:

Okay now we have the process outlined we can analyse the miners incentives/behaviour. I'll be using game theory to explain this and here is where it gets interesting.

The miners main goal is to make profit. This is why he adds tx. to his block in the first place (it allows for more fees and thus, more profits). So we can assume that without the blocksize limit, the miners would add infinite tx. to their block right? WRONG!

Your game theory calculations only deal with actors that are inside the system. They are also projections of our time period to 8 years ago, when money generated from fees were non-serious. I mean even the bitcoins generated were non-serious money.

Actors from outside the system that want bitcoin to die, can spam it to death - AND include the blocks on their own / at their own expense, because the cost of bitcoin back in 2010 was peanuts.

Why would Satoshi allow that? He'd be an idiot if he saw a vulnerability and left it there waiting to be exploited.

[dinofelis]

Actors from outside the system that want bitcoin to die, can spam it to death - AND include the blocks on their own / at their own expense, because the cost of bitcoin back in 2010 was peanuts.

The point is that by PoW consensus, every previous block is voted over by PoW by the peer miners.  So if the PoW consensus is working, you might think that even if a crazy attacker makes a 10 GB block that is correct according to the bitcoin protocol, the next miner should most probably decide NOT to build upon it because he estimates this not to be the good consensus of the bitcoin state.

I think that Satoshi, who also made the *mining software*, built in this 1 MB as a MINING directive, not as a PROTOCOL specification - because bitcoin miners were also taking Satoshi's software to MINE.  In fact, in those days, a node was a mining node (one CPU/one vote).  So the 1MB limit was probably seen, by Satoshi, as the "no-go" for a honest miner to not build upon a crazy block.  Not as a protocol specification.

He couldn't of course, protect bitcoin against any serious outside attack, because bitcoin was proof of work and the proof of work in those days was peanuts.  If it would have been a PoS coin, it would have been protected (especially given the big stash that Satoshi had himself) as the attacks would be limited to insiders.  Nobody attacked bitcoin at that time, because it was considered a joke by those that would now most probably curse themselves from not having done it !

[AlexGR]

Actors from outside the system that want bitcoin to die, can spam it to death - AND include the blocks on their own / at their own expense, because the cost of bitcoin back in 2010 was peanuts.

The point is that by PoW consensus, every previous block is voted over by PoW by the peer miners.  So if the PoW consensus is working, you might think that even if a crazy attacker makes a 10 GB block that is correct according to the bitcoin protocol, the next miner should most probably decide NOT to build upon it because he estimates this not to be the good consensus of the bitcoin state.

But another miner might see that another block was issued as #xxxxx instead of the 10gb block and then reject that one for not building upon the 10gb one. So it's a gamble that others will do the same...

[dinofelis]

Actors from outside the system that want bitcoin to die, can spam it to death - AND include the blocks on their own / at their own expense, because the cost of bitcoin back in 2010 was peanuts.

The point is that by PoW consensus, every previous block is voted over by PoW by the peer miners.  So if the PoW consensus is working, you might think that even if a crazy attacker makes a 10 GB block that is correct according to the bitcoin protocol, the next miner should most probably decide NOT to build upon it because he estimates this not to be the good consensus of the bitcoin state.

But another miner might see that another block was issued as #xxxxx instead of the 10gb block and then reject that one for not building upon the 10gb one. So it's a gamble that others will do the same...

This is why this is a consensus decision.  If a majority of hash rate thinks that the 10GB block is a good thing, then it will be included (as you say).  If the majority of hash rate thinks that it is a spam block that shouldn't be part of the consensus, they will mine on top of the smaller block.

A miner can put a lot of considerations in his decision ; in as much as other miners agree with it, his decisions are confirmed, in as much as other miners disagree with it, his decisions are rejected.

Note that the risk for the second miner is also on his side if he mines on top of the 10 GB block.  After all, the third miner may be of the same opinion as the first one, and think that such a large 10 GB block is not reasonable, and orphan the 10 GB block plus the second miner's block.  So the second miner has to try to guess what is the most likely consensus decision that will be taken by the third miner.  And so on.

Most probably, it was with this in mind that Satoshi introduced his 1 MB block size.  When he introduced it, 1 MB was way larger than any reasonable block at that time, so it was a good "miner directive" to say that blocks larger than 1 MB should be considered as spam (given we measured blocks in KB at that time).  It was a "good miners' directive" not a "protocol standard".  The idea was most probably, in Satoshi's mind, to give a clear rule to the miners as what was to be considered "prohibitive spamming".  It could systematically be put a factor of 100 or so above the "usual block size".   It was a trivial formalization of the good practice "don't accept obvious spam in the consensus decision".

But then it became part of the "protocol of bitcoin" by some crazy narrative.

It is important to realize that in mining, there's no "accumulation of success".  At the point that the second miner receives the small block from his peer, even if he was already mining (without success) on top of the 10 GB block, he doesn't lose any opportunity from switching to his peer's smaller block.  It is not because he has been mining for 5 minutes on the 10 GB block that his chances are bigger when he continues to do so.

But on top of that, chances are that he has faster the full small block of his peer rather than the 10 GB block, if his network connection is of limited capacity.  So from the moment that 10 GB is a network issue, *in any case* it will arrive late at the miner's place.

So there are enough arguments to say that there's no reason to expect that the block chain could have been spammed by gigantic blocks: miners would sensibly reject them in their consensus decision, and those blocks pay already a penalty because they lose time in network transmission.

[AlexGR]

Most probably, it was with this in mind that Satoshi introduced his 1 MB block size.  When he introduced it, 1 MB was way larger than any reasonable block at that time, so it was a good "miner directive" to say that blocks larger than 1 MB should be considered as spam (given we measured blocks in KB at that time).  It was a "good miners' directive" not a "protocol standard".  The idea was most probably, in Satoshi's mind, to give a clear rule to the miners as what was to be considered "prohibitive spamming".  It could systematically be put a factor of 100 or so above the "usual block size".   It was a trivial formalization of the good practice "don't accept obvious spam in the consensus decision".

That's the soft limit... which could be set as a parameter, like 250k, 750k, etc... the 1mb was a consensus rule - and obviously he told garzik that this would make him incompatible with the network if he changed it => https://bitcointalk.org/index.php?topic=1347.msg15139#msg15139