The real danger of this thing is actually that bitcoin really becomes secure with PoW only when its proof of work consumes demonstrably more than half of the world's electricity supply.  Maybe bitcoin is the first
https://wiki.lesswrong.com/wiki/Paperclip_maximizer

It turns essentially everything into mining equipment.  First, it enslaved humanity's greed as its work horse.  Now it wants them to make mining equipment.  Maybe between 2008 and 2010, Satoshi realized this, and wanted to stop that madness with a finite block size ?    

One can estimate (I did it elsewhere) that if all world financial investment put 1% of their holdings in bitcoin, we will be forced to consume 1/3 of world electricity production for mining.  Give or take a rough factor of a few.   Maybe between 2008 and 2010, Satoshi did this simple math too.  I'm surprised nobody else did.  It's quite simple.  Here is the outline.

The state of the art mining is the S9 antminer, consuming 0.1J/GH.  (Joule per Giga Hash).
The current hash rate of bitcoin is of the order of 20 billion GH/s right now (that is proved proof of work).  So the very minimal energy consumption of bitcoin right now is 2 GW.  
But right now:
 - bitcoin's miners are most probably a mix of newer and older technologies, consuming more
 - with the very recent price increase, right now mining is very, very profitable (cost of mining much lower than block rewards+fees in the market).  One roughly estimates this a factor of 5.  This is why hash rate is strongly rising, even if price is dropping: hardware installation hasn't caught up yet.

==> at about $10 000 / BTC, we can roughly estimate the equilibrium power consumption at 10 GW (5 times the lowest possible one right now).

Now, what is bitcoin's maximum "success" market cap ?

If bitcoin is a currency, it will eat away the fiat market, which is several tens of trillions of dollars.  ==> factor of several x100 from now.

If bitcoin is a speculative asset (most likely), it will eat away at the speculation finance market.  That's 2 quadrillion world wide (the derivatives market).  ==> factor of 10 000.  If bitcoin represents 1% of all investment in this market, ==> factor of x100 from now.

So everything points to a factor x100 if bitcoin is a total success story: "most of currency" or "1% of speculative assets".  Personally, I think everything points to speculative asset, but that doesn't matter here.

If x100 in market cap, also x100 in mining power consumption.  Not 10 GW, but 1 TW.  World electricity production: 2.8 TW.

QED.

Very funny to see discussions about scaling whine about *disk space* if you realize this.

Yes, that's the idea of trustlessness through decentralization.  The only common thing a large set of untrustworthy people can agree upon, is the original set of rules, even if all of them, individually, would like to deviate from it, but each one in his own way.  But the only way to deviate from it successfully, is to deviate from it in the same way as a majority.  For that, they need to set up a large conspiracy.  And, as you say, they can't trust one another within that conspiracy, to stick to that deviation and not do anything else.  If their conspiracy fails, they are worse off than if they had followed the common original rule set.  

This is also how you can get a large set of people to obey a rule set that they don't like.  This is how a general can get soldiers go dying on the front lines.  If a minority of soldiers rebels and tries to avoid their stupid fate, they will get shot as traitors or deserters.  If a majority of soldiers rebels, the general is done.  But how, as a soldier, to be sure that your majority mutiny will work out ?  If you don't reach majority, you're dead for sure.  If you just go to the front lines, you might survive.  It is the basis game theory of all repressive systems.

This is also the great danger that resides in crypto.  We might all be forced, by crypto, to obey rules we don't like.  Decentralized systems can become very dangerous.  If the King becomes nasty, you can go and kill him.  If a decentralized system starts behaving in a nasty way, there's nothing you can do.  Maybe one day, we may regret that we've shown some how they can lock in people in a decentralized rule set...

See also: https://bitcointalk.org/index.php?topic=2786690.msg29143511#msg29143511

Maybe bitcoin is the "paperclip maximizer" that will destroy us, because it wants to turn everything into proof of work.  Even if all of us, we see it, but individually, we cannot get out of the decentralized rat race, there's nothing we can do: we will be mining, and nothing else !

Bitcoin's power consumption is, at this moment, several GW.  If all mining were done with the latest Antminer S9, which consumes 0.1 J/GH, and we're at 20 billion GH/s right now, bitcoin would consume 2 GW.  However, we see that the hash rate is strongly rising right now, because the price increase of bitcoin, and the new S9 miner's technology, haven't yet reached equilibrium, simply because hardware hasn't yet been installed fast enough to catch up.  Right at this moment, mining is very lucrative.  We can expect most probably a factor of 5 in hash rate.  So we can expect in a few months time, unless the price crashes, that bitcoin will consume 10 GW (or is already consuming this, because of older hardware running).

If bitcoin becomes world-wide accepted as a means of payment, its market cap should rise about x100 because that's the level of market cap of world-wide currencies (tens of trillions).  If bitcoin is not a currency, but a speculative asset (everything points in that direction), the market cap can even go much higher: the total market cap of speculative assets is rather 2 quadrillions.  That's another factor of x100.
So potentially, bitcoin has a factor of 10 000 to go still if it eats the entire speculative asset market.
But let us assume that it eats 1% of that.  That's still a factor of 100.  Now, that should imply also a factor of 100 in proof of work power consumption.  If now, we are at 10 GW, we will then be at about 1 TW.  World electricity production is 2.8 TW.  

So, if all financial investment systems invest 1% in bitcoin, bitcoin will force us to consume 1/3 of world electricity production for mining.

His solution of November 2008 was perfectly alright:  there would be a class of "specialist mining nodes with farms of specialized hardware", and these nodes would be keeping the block chain in full.  They would also act as SPV servers for the light wallets of the millions of users.

Given the investment that "specialist mining nodes" make in their mining hardware, the extra cost of storage and bandwidth, to serve the normal, non-mining users and their SPV wallets, is relatively small.  The "specialist mining nodes" would be small (or big) data centres.  They could serve of course all the normal, non-mining users.

The network and storage load of an SPV wallet is much, much lighter, as Satoshi explains.  So normal users are not concerned with big block chains. It is good enough to autonomously verify that one isn't telling you jokes, and that the transactions you see ARE part of the block chain.  For Satoshi, you only run a full node (with all the data and network burden) if you also "try to make coins" (that is, mine).   A full non-mining node was an absurdity for Satoshi.

As such, even today, his solution can perfectly scale to VISA levels.  The only caveat is that it shows the de-facto centralization in bitcoin.  Satoshi was thinking of hundreds or thousands of "mining nodes".  In reality, we have a handful of mining pools.  



Of course he wouldn't, most of the time.  At that time, there weren't enough transactions to fill a block near to 1 MB, and even if he did, he would only make an orphan, not "fork off".  There's nothing dramatic by having an orphaned block.  Until the middle of last year, miners regularly had blocks orphaned.  It is only when bitcoin mining centralization was complete, that no orphans are made any more (since June 2017 more or less).

The limit was way way higher than the size of needed blocks back then, so there wouldn't be any sensible reason to make such blocks.  So limit or no limit, a normal, well-behaving miner (I count Garzik amongst them) wouldn't, ever, make a block larger than 1 MB.  And if you don't make such blocks, you are going to make blocks that are compatible with the others.  And you accept all blocks by the others.  

Again, the only problem would have been if Garzik's patch were majority adopted, and the problem would then be for those that kept their minority hard limit.  At no point it would have been detrimental TO GARZIK:

- Garzik in minority: accepts still all blocks by others ; will normally only make blocks also accepted by others if reasonable ; if not reasonable, will only make a single orphan monster block.  --> not really detrimental to Garzik

- Garzik in majority: everybody is making big blocks and having fun, except for a minority that keeps the hard limit: they will make a small minority chain that has forked off the rest.  --> also not detrimental to Garzik and his majority.



That's because you should have used a light wallet like electrum.  Electrum is a SPV wallet, like Satoshi proposed for normal users in November 2008.  If your idea was not to mine, there's no reason you download that huge block chain (unless, like me, if you want to look at it with a hex editor...)

I started out in bitcoin with electrum for my own coins.  I only installed, for fun, a full node during a few years, but I never used it for anything else but to look at the blockchain.  As a wallet, it wasn't great.  I didn't want to use a non-HD wallet.  Using a non-HD wallet is very dangerous, because there's a point in time, when you do a transaction, that the very single only place where your private key resides, is in the wallet.dat file.  If at that moment, your disk scratches, you've lost potentially all your funds.  Even if you made a backup 4 minutes earlier.  Very dangerous. 


I could also say: Youtube is bullshit.  I wanted to download the entire datacenter serving Youtube, and it took me a month and a truckload of hard disks.  And I only wanted to look at a single video.  What bullshit, Youtube.  But people don't download all of Youtube to look at a single video.  They connect to a Youtube server, and take what they need.  

As a bitcoin user, you are only interested in a very small part of the block chain: those transactions that you are concerned with (to you, and from you).  That's what a light wallet downloads, together with the cryptographic proof that it is indeed, part of the block chain.  That's what Satoshi already said in that November 2008 mail.

They don't know that decision.  If they knew the exact block (and more, the RULES that decided to make that block) on which they would be giving hash rate, then that sub-miner that would "find the hash" wouldn't send it to the pool, but would publish the block all by himself of course.  So it is essential for the good functioning of a pool, that the miners in the pool don't know on what block they are hashing, or they can screw the pool when they win.  As such, the miners of a pool don't have any "control" over what their pool decides to make as a block.  They could, at best, know on what previous block their pool is hashing, and hence, whether or not their pool has decided to orphan a block, because that's part of what they have to hash.  But the content of the block is hard to guess: they only get the top of the Merkle tree.

Miners are in fact just subcontractors that sell hashes for coins to the pool.    And most of them just configure gear and software, and let it run.  Not much "judgement" in there.

Not really.  It depends on the exact PoS implementation.  I think I'm going to write up my arguments instead of partially typing them in forum posts.

I really would like to emphasize this.  The entire idea of trustlessness is a reverse application of the tragedy of the commons.  

The tragedy of the commons is normally understood that people all agree that it would be better if everyone did X, except that it is individually more advantageous to do Y, even if everybody doing Y is worse in the end.   It is a matter of game theory and Nash equilibrium.  The definition of a Nash equilibrium is this:
"that set of strategies E of a set of N players in a game G, such that, for every player P, if player P changes strategy, and his N-1 peers keep their strategy in E, it is worse for P in the game G than if he applies his strategy in E".

The school example is the Prisoner's Dilemma https://en.wikipedia.org/wiki/Prisoner%27s_dilemma.  Many games have at least one Nash equilibrium.  The whole "tragedy" is that the Nash equilibrium can be far from the optimal global strategy.

"Decentralisation" pushes this thing somewhat further.  One can break out of a Nash equilibrium by collusion with other players.  E may be a Nash equilibrium of game G, if players P and Q collude, they can "game the system" and break out of a normal Nash equilibrium.  In the Prisoner's Dilemma, if the two prisoners collude, they can get away with a better solution than the single Nash equilibrium of the system.

So having a game with a simple Nash equilibrium doesn't protect that equilibrium against collusion.  However, collusion becomes harder and harder if more and more players need to be part of it.  If you can set up a game, in such a way that there's a "Nash++" equilibrium, that is stable against not just one pair of players colluding, but can only be broken if a very large set of players collude, you get an extremely stable system, even if most of the players "would like to leave it in a different way" - that is, would like to game the system in THEIR way.

THAT is decentralisation as a means to obtain trustlessness: having a system in which there is a Nash++ equilibrium, so that even with a large set of "dishonest" players, the equilibrium (the rules of the system) won't be broken.  Because to break it, such a large collusion is needed, that it is too hard to organize and to "trust".

It is like the Prisoner's Dilemma with 500 prisoners.  They will all speak.  Because the probability that not one single one of them will speak, is essentially zilch.  It is the tragedy of the commons from the point of view of the cheaters: even if a collusion would be better for them if they all colluded, it is simply not going to happen: some will cheat the cheaters and then it will be "worse" for those trying to collude.

I can even pinpoint EXACTLY where Satoshi was making this error (at least, in as much as he really wanted to make a currency).  It is here:


http://satoshi.nakamotoinstitute.org/posts/p2pfoundation/3/#selection-41.0-41.460

It is because he thinks he doesn't know how to know the price, that he put in a random emission curve.  The "positive feedback loop" is exactly what's called a deflationary spiral (currency talk) or a speculative bubble (asset talk).

"I don't know how to make a currency, so I'll make you a speculative asset instead".

Because he wrote it in November 2008 maybe ?  http://satoshi.nakamotoinstitute.org/emails/cryptography/2/

That is the normal consequence of bitcoin, even though pretended to be a currency, because it has the monetary policy of a highly speculative collectible, in the same way rare paintings or old sports cars are.  A lot of emission when it has no value (when van Gogh wasn't renowed), and a diminishing emission when it catches on, is the recipe for "get-rich-quickly" speculation.  From the moment that a collectible can rise in value without bounds, it becomes of course a speculative asset.  From the moment the only thing such an "asset" has going for it, is its market price, of course you get speculative bubbles and crashes.

The fundamental property of a good currency is that it is constant in value, and hence neutral in its use.  It is a perfect store of value in the short term as well as in the longer term: what you put in, is more or less exactly what you get out of it.  A currency is NOT an investment, and that neutrality is exactly what turns it into a currency: that you can store value in it when you acquire value, and that you can get the same amount of value out of it when you spend it.  That's the entire fluidifying role of a currency in economic relationships.  Nash called such a hypothetical asset an "ideal currency".  In order for an asset to get close to an ideal currency, its emission should be coupled to its market price.  That's the official role that central banks have.  The criticism to central banks is that they abuse of their power, but that is no excuse NOT to have a regulation mechanism.  Bitcoin doesn't have any.  There's no extra emission of bitcoins when the price rises, as it should, if bitcoin were to have the pretence to become a currency.    The naive gold bugs theory that good money is a collectible, is simply wrong.  Collectibles that serve as money can fall in what's called a deflationary spiral, which is nothing else but "a speculative bubble".  You hoard the asset with the hope for it to rise in value.  That hoarding lowers its velocity, which indeed, makes its value increase, which, in its turn, inspires even more people to hoard it.  If people have been hoarding it like crazy, and only a very tiny fraction of its volume is still used as a currency, the price of the currency is extremely high.  It then becomes tempting for the hoarders to "cash out".  That will significantly increase the small pool of circulating currency: if 99% was hoarded, and 1% cashes out, the circulating supply doubles, from 1% to 2%.  This has a strong effect on the price of the asset, which can lead to people wanting to cash out before their gains are gone.  We get a hyper inflation (a crash).

The words "deflationary spiral" and "hyper inflation" are the monetary speak of "speculative bubble" and "crashing, popping bubble" in stock market language.

A collectible goes through these cycles (once, or several times).  It is useless as money, because of the high volatility.  If there is no price-regulation mechanism, there's no hope.  Bitcoin has no price regulating mechanism.

Another way to regulate the price of an asset, is to make it "redeemable" against something: to back it against something.  As such, the price of the asset will remain in the neighbourhood of the redeemable something.  If you can redeem a bitcoin against a Big Mac,  bitcoin's price will not deviate a lot from the market value of a Big Mac.  It is the fact that it is not backed by anything, that allows bitcoin to have huge value.  Nobody would pay a coin $10 000 if it is redeemable against a Big Mac.  It is because it is NOT redeemable against a Big Mac, that paying $10 000 for a coin doesn't sound crazy.

That will always be the case in an open, anonymous, permissionless system.  The only way to avoid that, is by using "one man (woman, animal, ...) , one vote", and then you need to rely on real-world identities, issued by, well, centrally controlled identity-issuers.  How else do you associate one single human being to one pseudonymous identity on an open network ?  From the moment the network is open and anonymous, sybil sock puppets are possible.  Whatever proxy you use for "person", be it a CPU, an IP address, a smart phone IMEI, a phone number, .... you have the double problem that:
1) there are centrally controlled entities that ISSUE these things and/or
2) the fatter your wallet, the more of them you can afford.

Within a crypto currency system, however, it seems normal that the stake holders can vote according to their stake, in the same way that share holders can vote according to their share.   It would mean that the more vested interest you have in the system, the more you have to say about the system.  In any case, in a value-carrying token system, if a majority of tokens are in the hands of a colluding group, you better get out right away in any case.

With PoW, you get a combination of fatter wallets and better energy and hardware opportunities to be the voting key distribution.

As the URL said, I didn't have time to write a short letter.  If you want it terse, these services are paid-for.  You get the free, long version here.  Contact me for the paid-for short version  


Remember that we are talking about "decentralization".  By putting it so bluntly, you can say "if the boss of Facebook is dishonest, then he can screw Facebook communication.".  That's just as obvious.  What does this have to do with "decentralization" ?  

It has to do with the fact that in a "decentralized system", there needs to be massive collusion of independent and even competing entities to reach that 50%.  So the measure of the number of independent, competing entities is the measure of decentralization.  With Facebook, that number is "1": the boss of Facebook.  With bitcoin, that number is "3".  Ok, slightly better, true.

The whole idea of "decentralisation" was that the number of *independent* entities was so large, that the idea that sufficiently many of them to collude over the same "dishonesty" in the same way, that they can reach majority, is essentially unthinkable.  If you have 500 independent people all over the world with more or less equal voting rights, and opposed, competing interests, then the possibility of 250 of them to collude over the same dishonesty, is relatively small.  If you have 1 million of people with voting rights, all over the world, in different cultures, nations, .... then the chances that they are going to collude is zilch.   The whole idea of trustlessness was the idea that a large set of potentially dishonest participants are nevertheless obliged to follow the common rule set, even if each of them is potentially dishonest, simply because in order to deviate from it, they'd need such a large, common, identical conspiracy that it is not going to happen because impractical and improbable.  Each of them would conspire, but over different things, and they can never reach a majority conspiracy.

If you have 3 guys that see one another regularly at conferences, though, that's indeed SOMEWHAT more decentralized than if you have one guy that needs to agree with himself, but I'm not really impressed.


If these 3 guys screw us, we have no recourse.  In the same way that if Facebook's boss screws us, we have no recourse.  Or if 15 ministers of finance of the Euro zone screw us, we have no recourse.


Nope.  If Zuck screws with Facebook, another one will take his place ; shareholders can fire him ; you might attack him in court ; there are so many things that could screw up Facebook.  Yahoo was also the king of search results and now it is Google.  MySpace was the place to be, now it is Facebook. As a big boss of such a thing, you have to be very careful.   But we agree: the fear of the market is what keeps these monarchs or oligarchs more or less in check.  Not "decentralization".  Hell, the Euro is more decentralized than Bitcoin ! You need a 2/3 majority of finance ministers.  That's more than 3 miner pool owners !

In fact, Intel and AMD didn't profit a lot in this.  If there's one company that even has production problems because of PoW, it is Nvidia, and not for bitcoin, but for ethereum and a few others !  Also, I don't think it is Big Oil that profits, it are the Chinese charcoal mines.  Mining is to be seen quite more literally  

What is startling in this story, is that one almost has the impression that Satoshi's intellect is waning.  His premonitory views in November 2008 were much more to the point than his "incomprehension" in 2010.  Is this because Satoshi was bright, but easily influenced by a dumb ass big mouth that talked him into nonsense ?  Is it because the Satoshi of 2008 was not the Satoshi of 2010 ?  Was Satoshi ill ?  Was he consuming drugs ? Did he get hit on the head ? Or is this the obvious impression that you get from someone trying to trick you into something (a set of lies always look less smart and coherent than a truly brilliant idea) ?

You can't be smart to a point of foreseeing, without one single working system in nature, how thing would evolve in 2008, and be obtuse to the point of failing to see the obvious when it is even spelled out for you by other people in 2010, unless you are someone else, or you were hit on the head, are ill, or are lying.

This is what occurred to me recently ; until now, I simply thought: "well, a big mistake in 2010, but everyone makes mistakes, Satoshi was a bright guy but not a genius".  But now, as I said, the pieces just don't fit together.

What is remarkable, even though nobody ever said it, is that bitcoins' system itself has an indicator of the necessity of the block size: the difficulty !  The difficulty measures the market value, the size of the user community, the technological advances, and the investment mining nodes make in PoW (and the rewards in value they get).    It is quite obvious that things like storage costs and network burden, and their economic cost, are going to be proportional to this.  So it would have been rather obvious to scale block size with difficulty.
The bigger and more valuable the network gets, the more NEED there will be for transaction room, and the more MEANS there are to afford storage and network capacity.  It would also solve the "end problem" of no coin emission, to a point: if fees are too low, and finance PoW, difficulty falls, the blocks shrink to a point where the fee market kicks in, until an equilibrium is reached where the block size allows for a reasonable fee, paying for a reasonable amount of PoW, giving rise to a reasonable difficulty, as a function of demand for transactions, and market price of BTC.

I think that sums it up quite well 

I know you said that, but no cigar.  Bitcoin's mining pools have hash majority with 3.  Let us say, 4.  The 4 biggest mining pools hold
65% hash rate under their knee.  See https://blockchain.info/pools  If these 4 entities decide that your transaction is not going to make it in the block chain, then you can transmit it as many times as you want, it won't make it into the block chain.

They will firstly not include your transaction in the block THEY make.  Moreover, they can decide NOT to mine on top of a block that contains your  transaction, as made by a minority miner.  If ever a minority miner includes your transaction, his block will get orphaned, and given that the 4 entities have 65% hash rate, his orphaned prong will always lose.  They may be friendly to these smaller pools, and inform them that they should exclude your transaction, or lose their blocks.  If these 4 guys maintain a blacklisted set of addresses, then the coins in these addresses are dead.  You'll never know.

Note that mining POOLS are the entities that decide on what block they mine, and what transaction they include.  The miners connected to the pool don't have anything to say in this decision, and don't even KNOW these decisions.

The only thing you can do is shout here, or on coindesk, or on national TV, or on Facebook (oh sweet irony) that your transactions always remain in the mem pool and never get confirmed, and in those cases they get confirmed once, that block gets orphaned.

There's your "decentralization". True, you need 4 guys (strictly speaking, 3 guys) to decide that together.  Zuckerberg can do it on his own.

Now, why isn't this happening ?  Well, maybe it IS happening. There is no way to know that amongst all the rejected transactions in the mempool, there is not one of a poor guy that is excluded. But if it isn't, it is not happening for exactly the same reason it is not happening on Facebook: because of the market.  If it would get known that the mining pools do this, they may lose business.  Miners may connect to another mining pool.  Bitcoin may plummet in the market.  And exactly for the same reason, you get your messages through on Facebook, because Zuckerberg doesn't want you to tell the press that Facebook is censoring you.  

In fact, there's more chance that Zuckerberg is afraid of this bad publicity, than the mining pools.  Remember that they played perfectly according to bitcoin's rules.  No node will complain.  So, your message is just as guaranteed (or even more) to get through on Facebook, than on the bitcoin network.  And not because of decentralization.  Because of fear for bad publicity.

There.  Be careful if you insult a Chinese king of bitcoin.  You may never transact again.  And note that it doesn't cost them a single hash, or a single fee (apart from yours, of course).  And now, re-read what I said above about "decentralization".  Because not only the mining pools have this power.   Core has that power too.  If they decide to blacklist your address in the next version of bitcoin core, then you can do what you want, your transactions will not any more be part of the official bitcoin protocol !  If you think that's a joke, that's more or less what Vitalik did with the DAO hacker.  Ethereum's protocol was modified so as to kill this guys' transactions and pretend they never happened.

All this "decentralization" talk is entirely bogus.  Salesmen snake oil.  But it is not needed (in most cases).  Because of the market and fear of bad publicity.  Like with openly centralized entities. 

In fact, the actual result of difficulty adaption is rather the opposite.  If you have "one bad actor with massive amounts of hashing power that can mine all blocks for himself", then the automatic difficulty adaption is *in the advantage* of this actor, not against this actor.  

Imagine a system where you have 100 "normal" solo miners that mine a block every 10 minutes.  Each miner has, on average, 1 chance in 1/100 to mine a block.  His average time of success is 1000 minutes.  His success series is a Poisson series with a time constant of 1000 minutes.  Each miner has an individual, independent Poisson stream of success with a time constant of 1000 minutes, and the total block rate is hence a Poisson stream (the union of these 100 individual Poisson streams) with an average of 10 minutes.

Now, imagine that our 'bad actor" arrives with about 100 times more hash rate.  Without adaption of difficulty, this doesn't influence the other Poisson streams.  Our new actor will make new blocks on average every 0.1 minutes, but our 100 other actors still have their block every 1000 minutes each.  

Come adaption: difficulty goes up x 101 (let us say, 100).  Or bad actor now has a Poisson stream of success of 10 minutes on average.  Our other actors are now only getting each a success every 100 000 minutes.  That's one block ON AVERAGE every 2 months.

It is not just that our bad actor has diminished revenue of our 100 normal miners: he also increased seriously the FLUCTUATIONS of their revenue.  Mining doesn't only become less profitable, it also becomes riskier.

So the difficulty adaption made life much worse for the "good and numerous" actors.

If you go from 100 to 1000 "good" actors at the start, they go from one block every 10 000 minutes (one block every week) to 1 million minutes (one block every 2 years).   Suppose that the average revenue still allows the good miners to make up for their PoW.  On average every week a block makes them get a relatively steady stream of income ; however, on average every 2 years a block is way, way too risky.

So, difficulty adaption favours the big "bad" actors.  It pushes the smaller miners into oblivion.  You limit the amount of independent solo miners to a small number.  in other words, you centralize, simply already by the effect of income fluctuations and hence, financial risk.  Add to that economies of scale, and you're done.

That's obviously not the case, as I tried to point out.  First of all, Satoshi couldn't know what would be the fraction of the network that would apply the patch.  So by applying the patch, Jeff might just as well get the majority behind him, and it would be to the detriment, not of Jeff, but of those that didn't apply the patch.  The fact that Satoshi assumes that Jeff would automatically be in minority - or at least, wants Jeff to be afraid to be in the minority - is telling.

But secondly, assuming that Jeff himself wouldn't mine, himself, a lot of big blocks (one can hardly think of Jeff wanting to spam the network himself), and assuming that the rest of the network WOULD filter out the big blocks, and only make a small block chain, Jeff wouldn't, at all, "make himself incompatible with the network".  Jeff would accept ALL the small blocks made by the network.  Most probably, Jeff would mine, himself, also only small blocks and get them accepted by the others.  The only difference between Jeff's code and the rest of the network, under the assumption that they don't follow Jeff's patch, is that Jeff's node is more permissive.  So he won't be incompatible with the network.

Satoshi is telling bullshit here. And I can't imagine that he doesn't know.  At no point, his phrase, that it would be "detrimental to Jeff", is true. It would ONLY be true in those rare circumstances that Jeff wants to mine a big spam block, and the majority of the network doesn't follow him.  And it would only be detrimental for Jeff's PoW on that sole block. Jeff's node would accept the fact that his block was orphaned, and that another chain with smaller blocks took over.  From then onward, Jeff wouldn't be "incompatible with the network" any more.

So this is simply utterly wrong.

The only "dangerous" situation is when Jeff is in majority !  When Jeff-like majority miner nodes mine a chain with big blocks, and the minority Satoshi-like nodes don't accept this, and mine their own prong - or adapt.  At no point, anything is "detrimental to Jeff".


I have a hard time imagining that Satoshi didn't understand what he was saying, but nevertheless said it.  I have a hard time thinking that Satoshi:
1) didn't understand the arguments of the danger in the long term of locking in the 1MB limit
2) didn't understand that not applying this limit didn't exclude you from the network at all
3) was capable of suggesting how it could be done safely with an if (blocknumber < ...) {limit} but didn't do so
4) but nevertheless said that it was dangerous to remove his constant as it would exclude you !

except of course if Satoshi was just a sock puppet of Hal Finney and didn't understand zilch himself of this stuff.

It simply doesn't fit together.

I want to pay hookers in paradise with my coins.  No way my family takes them from me.  I only hope I'll remember my secret keys when in heaven... 

I hope you understand how contradictory these two statements are.  Something that changes by so much over such short time scale, cannot ever be a useful currency.  Imagine what that Laszlo must be thinking now !  If he bought a pizza in exchange for a luxury yacht 8 years later ! What a fool !

Eastern Germany was called the Deutsche Democratische Republik, and there was nothing democratic about it.  In the same way, bitcoin is called a crypto currency, and there's nothing currency about it.  In 2010-2011, bitcoin underwent a deflation of a factor over 40000%, followed by an inflation by about 1000% the next year.  In 2012-2013, again a deflation of about 40 000 %, followed by an inflation of 600% the next two years.  In 2017, bitcoin underwent an inflation of 2000%, and this year only started, to get a deflation of about 40% in a few weeks time.

If a third-world fiat currency has such fluctuations, people use bananas as a currency.

Of course, with a PoW consensus mechanism, you need rewards, to compensate for the economic loss in PoW.  The problem is triple:
1) that you then get convergence because of economies of scale (see bitcoin)
2) that you waste a lot of value, in other words, your system is a net value burner.  You need to waste a high amount, because that waste is your only protection.  If the waste is not gigantic, anyone can attack from the outside
3) that you split the system in a set of "wasters/PoW industry" and a set of "users", and both have different objectives.

But that's more a problem of wanting to use PoW based consensus, than an inherent problem of finding consensus.  

As I said, rewards give rise to strategies.  Strategies can then be different than the desire to come to honest consensus.  It is very difficult/impossible to find rewards such that the optimal strategy for those rewards is going to coincide with the desired other outcome, namely consensus.   The systemic difficulty in reaching consensus increases if you reward the one proposing it, instead of making it easier.  Normally, reaching consensus on "sufficiently old mem pool messages" shouldn't be that hard.  Of course, the *current mem pool state* of all active network participants is different, because of network delays.  But a list of "old enough mem pool transactions", broadcast at a certain moment by one of the participants, is easy enough to check and confirm.  The "difficulty" of having several of those lists being broadcast nearly simultaneously, and arriving at different order at different network participants, can also easily be solved by including a symmetry-breaking merit function, part of the protocol, that will assign different preferences of the broadcast lists.  After a reasonable network delay, most nodes can assume they received all of the list candidates, the merit function indicates which one is to be preferred, and all nodes will come to the same conclusion.  That's the consensus.  When the network propagates the message that consensus has been reached, nodes can start thinking of broadcasting a next consensus list, built on the previous one.
If all of this is done without reward, and just on the basis of "altruism" because one wants the system to work, there is no incentive to "game the system", to "make others accept YOUR consensus list" and so on.  There's nothing at stake, apart from contributing to the good functioning of the system.  From the moment you introduce rewards, it becomes strategic to be the one that gets HIS/HER consensus proposal and not your peer's one.  Instead of cooperating in the network, you compete.  

Such a system is totally opaque to any form of long-range roll-back attack, simply because no roll-back is possible.  Reached consensus is reached consensus, done real-time and on-line.   The price to pay is that verification is only on-line.  If you leave the network, you trust your network peers that they continue to build the consensus.  
You can think of ways to recover from the improbable "global network split", by accepting, when the network unites again, a merge consensus, that accepts all that happened on the two split histories.    If you discover nodes that have another prong, with another history from a given point in time, you accept as well their transactions, as those in your prong.  
A priori, these two prongs should be compatible: this is like in a DAG like coin: there can normally not be double spends of the same coin, or it would mean that there was a node, sharing both half-nets and doing double spends, one on both sides.  Splitting the network is already difficult, but splitting the network and being on both at once is even harder.  If ever that happens, one could accept the double spend exceptionally.  That's just some extra coin creation, no problem if it is rare enough.  The important thing is that no roll-back is possible.
Again, this merging is quite straight-forward if consensus DOESN'T come with reward.  Because merging the prongs doesn't mess with rewards that don't exist.


Well, if supply follows demand, price will stabilize.  And supply is value-controlled: if *making* a new coin has a fixed economic cost, you will ONLY make it if its market price is higher than that cost.  So as long as there IS a demand, this will converge to the set price.  You are right of course that if demand simply plummets, the lowest supply (namely NO supply) cannot go lower than zero.  You might introduce a systematic destruction of coins if you think that's a problem: a destroyed transaction fee, which is a given percentage of the transacted value.  It would indicate a "half life of coin": grossly the number of times you can transact a coin before it is gone entirely.  If you put that to, say, 1%, a coin's half life would be about 100 transactions, grossly.  I don't know if such friction is a good idea, though.  At least, it would make speculation impossible.  You put an upper cap on value.

   Funny joke.  150 million dollars gone.  "very upsetting"