Why Bitcoin Core Developers won't compromise

[Heruur]

The Lightning Network can take a huge amount of transactions largely offchain.  The idea of having all transactions fully onchain is not a matter of principle, it's a matter of control from miners so that they can receive transaction fees more often.  SegWit allows a slight increase in onchain capacity which is enough for the short term while this offchain scaling can also be implemented.

You're wrong because off-chain transactions are against the nature of Bitcoin and Satoshi vision. All transactions must be ON CHAIN due to many reasons, technology is not in cause here. The decentralization model should also be for Core Dev we have to rethink the process of BIP proposal and reward Dev for it.


For exemple Bitgo Instant is great tool and might be considerated as Off-chain transactions network. You don't need a solution On the network but Off the network it's called business man

[1714783624]

1714783624

[1714783624]

1714783624

[1714783624]

1714783624

[1714783624]

1714783624

[1714783624]

1714783624

[Viper1]

I don't blame either Bitfury or Bitmain -- they are at least signaling what they perceive to be the best solution.

How do you know that's what they're doing? Are you on the inside hearing all the internal discussions/plans so you can state that for a certainty? I'm certainly highly skeptical as to that being their reason from the information I see.

How are investors  to blame for trying to make logical arguments about what is happening?

If you're refering to yourself as the investor making logical arguments, I've been reading a lot of your stuff the last couple weeks and I'm sorry, but you're clearly not "just" some lowly investor lol.

Bitcoin core devs Would agree to a similar agreement?  Very doubtful.

I have no bloody idea what they're thinking. The HK thing was a clusterfuck and as far as I'm concerned the real reason we're where we are despite all the supposed altruistic reasons some people like to promote.

Yes, that's their roadmap, but who decides when "if its needed"?  Greg?  He'll never say its needed.  If they can't agree we already need it badly right now, I don't see why there would be any logical explanation for them to act reasonably in the future.

Until segwit and LN are working AND people stop playing games with the network, we won't know whether it's needed or not at that point. Please don't try and tell me it's badly needed now given the "Shenanigans" going on. As for Greg, I think he's a very talented developer and very bright when it comes to that (although his view on bitcoin prior to coming on board was funny), but that's all he should do.

Could be.   I hope it doesn't end bad as I'm a hodler.   Maybe there will be a network split.  Who knows.

Could happen. When "Man" tries to gain control over something, they tend to end up destroying it. They often think they know when to stop but often momentum takes over and it's too late. History shows that time and again. So with the way things are going it's a possibility here.

[jonald_fyookball]

I don't blame either Bitfury or Bitmain -- they are at least signaling what they perceive to be the best solution.

How do you know that's what they're doing? Are you on the inside hearing all the internal discussions/plans so you can state that for a certainty? I'm certainly highly skeptical as to that being their reason from the information I see.

They are signaling for segwit or BU because they thats what they want.  Why else would they?  I'm always up for a good conspiarcy theory.

How are investors  to blame for trying to make logical arguments about what is happening?

If you're refering to yourself as the investor making logical arguments, I've been reading a lot of your stuff the last couple weeks and I'm sorry, but you're clearly not "just" some lowly investor lol.

and a guy that likes to post a lot?  Not sure what you're saying.
 

Yes, that's their roadmap, but who decides when "if its needed"?  Greg?  He'll never say its needed.  If they can't agree we already need it badly right now, I don't see why there would be any logical explanation for them to act reasonably in the future.

Until segwit and LN are working AND people stop playing games with the network, we won't know whether it's needed or not at that point. Please don't try and tell me it's badly needed now given the "Shenanigans" going on. As for Greg, I think he's a very talented developer and very bright when it comes to that (although his view on bitcoin prior to coming on board was funny), but that's all he should do.

What games and shenanigans?   Look at the all time chart for avg. blocksize.  Look at the bitcoin market dominance chart, and the median fee chart.  If you don't have a sense of urgency, I don't know what to say.   If you agree "that's all Greg should do" (develop), then why do you seem to be ok with him setting economic policies by calling for full blocks and a fee market? 
 

[Viper1]

If you don't have a sense of urgency, I don't know what to say.

Almost 6 months ago, when the number of transactions etc was a lot lower than it is now, Segwit was released which, if it was activated so it could be used, would alleviate some of the issue. It would allow LN to be implemented (both litecoin and vertcoin have performed real transactions on their mainnets using LN already). For 6 months it's been in the hands of the miners. They're the ones that have been in the position to determine whether bitcoin explodes or not since it's their voting that makes the decision. Until the outcome is decided, there's nothing to do and assuming they continue to block it, that's almost 6 month away. What I don't get frankly is that as far as I see, the vast majority of users/businesses want segwit activated. So why aren't the miners doing it?

I'm sorry, but if you think you can convince me that core would be to blame at that point you're sadly mistaken. There's plenty of blame to go around but as things stand right now, failure of bitcoin will be due to the miners as they could just suck it up and activate segwit. There are plenty more battles to be fought over all this stuff.

If you agree "that's all Greg should do" (develop), then why do you seem to be ok with him setting economic policies by calling for full blocks and a fee market?

I wasn't aware that he unilaterally decided what code ended up being released. As far as I'm aware it's the actual bitcoin developers that decide and he isn't one any more.

As far as the "fee market", that's controlled by the miners and always will be. The intent was always that as mining rewards decreased miners would get more out of fees. So higher fees to replace block rewards is just the natural progression of bitcoin over time.

[jonald_fyookball]

If you don't have a sense of urgency, I don't know what to say.

Almost 6 months ago, when the number of transactions etc was a lot lower than it is now, Segwit was released which, if it was activated so it could be used, would alleviate some of the issue. It would allow LN to be implemented (both litecoin and vertcoin have performed real transactions on their mainnets using LN already). For 6 months it's been in the hands of the miners. They're the ones that have been in the position to determine whether bitcoin explodes or not since it's their voting that makes the decision. Until the outcome is decided, there's nothing to do and assuming they continue to block it, that's almost 6 month away. What I don't get frankly is that as far as I see, the vast majority of users/businesses want segwit activated. So why aren't the miners doing it?

I'm sorry, but if you think you can convince me that core would be to blame at that point you're sadly mistaken. There's plenty of blame to go around but as things stand right now, failure of bitcoin will be due to the miners as they could just suck it up and activate segwit. There are plenty more battles to be fought over all this stuff.

If you agree "that's all Greg should do" (develop), then why do you seem to be ok with him setting economic policies by calling for full blocks and a fee market?

I wasn't aware that he unilaterally decided what code ended up being released. As far as I'm aware it's the actual bitcoin developers that decide and he isn't one any more.

He is the author of the core roadmap and one of the key "deciders", yes.  He recently gave up github commit access so the Blockstream conflict of interest wouldn't be quite as glaring.

As far as the "fee market", that's controlled by the miners and always will be. The intent was always that as mining rewards decreased miners would get more out of fees. So higher fees to replace block rewards is just the natural progression of bitcoin over time.

A market is made of supply and demand.  By deciding the supply (1 meg per block), a fee market is created.  This is admitted.

[franky1]

Fact: If you raise the blocksize up to a point where people can't run their own nodes, you cannot call it a peer to peer network anymore.

get the "gigabytes by midnight" script out of your head. the rises of blocksize can grow at a natural progressive rate that nodes can cope with.
core already admit 8mb is safe..
with all the code efficiencies since 2009 (libsecp256k1=5x efficient for instance),
the fact that we are not average homeline of 512mbit/s(38mbyte/10min) ADSL, but alot more as an average now
the fact that hard drives ar cheaper
the fact that the baseline raspberry Pi is now raspberrypi3

all show that 8mb is safe and admitted as such, but even so just going to 4mb is also ok. with a few tweaks ONTOP to further becoming extra safe such as limiting txsigops to 4k per tx or less forever...
all would show that there is nothing technically hindering the ability to run a full node at home

No amount of tricks can overcome the importance of a full validating node, so forget about SPV. The moment people can't have full validating nodes the whole concept of "peer to peer cash" it's game over.

and i now hope you see why the whole filters(gmaxbuzz) bridging(lukeJrbuzz) to create a cesspit of a TIER network by going soft is something i have hate of.

[ArticMine]

...
A market is made of supply and demand.  By deciding the supply (1 meg per block), a fee market is created.  This is admitted.

Yes this is correct. I commented on this in another related thread. https://bitcointalk.org/index.php?topic=1847331.msg19009361#msg19009361

[aarturka]

I'm glad they don't compromise, that makes bitcoin decentralized and very reliable. If they succumb to pressure of some Chinese monopolists Bitcoin will be in great danger.

[dinofelis]

How do you maintain the promise of so called "peer to peer cash" to scale globally without centralizing the network due huge blocks that people cannot afford to run at home, therefore not anymore peer to peer cash but peer to corporation to peer transaction? (aka what we have already in the current baking system)

I'm amazed that this point is repeated over and over, and I've been arguing, demonstrating, proving.... that there's no *power* to be had in running a full node that is not a mining node.  There is *information* to be had when running a full node: indeed, if you want to *verify yourself* what the system is doing *but without any means to act on it*, you can run a full node.  But that's informational, and is not a matter of power.

I'm amazed that people always put forward the "decentralization aspect" of full nodes, while Satoshi himself explained from the very beginning:
1) that the consensus system is Proof of Work, especially to avoid "proof of node" simply because that would be open to Sybil attacks, and as such, nullifying the decision power of non-mining full nodes ON PURPOSE.  PoW was introduced exactly for that !
2) that if the block chain becomes very succesfull, only people mining new coins need to run a full node, and that other users can use their light wallets to connect to them.

So, concerning the decentralization of bitcoins consensus mechanism, there's absolutely no use for non-mining full nodes.  As an individual power user, you may want to check for yourself whether bitcoin is still working how they told you it was working, and invest in a full node - but the only thing you will get out of that is *information* ; you cannot INFLUENCE bitcoin that way.

I've argued this very logical point, nobody has ever countered it, and it is fairly obvious from the writings of its creator that non-mining full nodes have no consensus power at all.

In other words, your permissionlessness, and your ability to transact peer-to-peer are totally INDEPENDENT of whether there are a lot of non-mining full nodes or not, because ALL THAT is decided by the consensus of miners.  The protocol they agreed upon to build the block chain, is the de facto protocol of bitcoin, and they decide if they include your transaction or not.  You don't need full nodes to transmit them your transaction: if you connect DIRECTLY to their nodes, they will get it.  And that was how bitcoin was designed !  Consensus is decided by those who deliver proof of work and explicitly NOT by the number of full nodes.

It is rather strange that one argues that the peer-to-peer ability to pay is compromised because Joe cannot run his full node in his basement any more (while this node never intervened in any consensus decision). but that the peer-to-peer ability would NOT be compromised by needing a lot of hubs in the LN network to agree to your transaction: hubs to which you are TIED with a payment channel which you cannot settle easily (as per definition that the on chain system is "compromised" and doesn't have, per design, the capacity for you to easily settle).  Being forced off-chain looks to me like a much higher danger to the peer-to-peer permissionlessness of transactions, than having to rely on the consensus of Proof-of-work providers only.

The current, actual reality is that all of bitcoins' consensus, including the protocol, the permission to transact, the fees, and everything, are the consensus that happens between about 20 entities, the pools, that together, have more than 99% of the decision power (PoW) under their control.  The consensus that emerges between these 20 entities is what we call "bitcoin", and bitcoin was designed to be like that.  

[jonald_fyookball]

How do you maintain the promise of so called "peer to peer cash" to scale globally without centralizing the network due huge blocks that people cannot afford to run at home, therefore not anymore peer to peer cash but peer to corporation to peer transaction? (aka what we have already in the current baking system)

I'm amazed that this point is repeated over and over, and I've been arguing, demonstrating, proving.... that there's no *power* to be had in running a full node that is not a mining node.  There is *information* to be had when running a full node: indeed, if you want to *verify yourself* what the system is doing *but without any means to act on it*, you can run a full node.  But that's informational, and is not a matter of power.

I'm amazed that people always put forward the "decentralization aspect" of full nodes, while Satoshi himself explained from the very beginning:
1) that the consensus system is Proof of Work, especially to avoid "proof of node" simply because that would be open to Sybil attacks, and as such, nullifying the decision power of non-mining full nodes ON PURPOSE.  PoW was introduced exactly for that !
2) that if the block chain becomes very succesfull, only people mining new coins need to run a full node, and that other users can use their light wallets to connect to them.

So, concerning the decentralization of bitcoins consensus mechanism, there's absolutely no use for non-mining full nodes.  As an individual power user, you may want to check for yourself whether bitcoin is still working how they told you it was working, and invest in a full node - but the only thing you will get out of that is *information* ; you cannot INFLUENCE bitcoin that way.

I've argued this very logical point, nobody has ever countered it, and it is fairly obvious from the writings of its creator that non-mining full nodes have no consensus power at all.

In other words, your permissionlessness, and your ability to transact peer-to-peer are totally INDEPENDENT of whether there are a lot of non-mining full nodes or not, because ALL THAT is decided by the consensus of miners.  The protocol they agreed upon to build the block chain, is the de facto protocol of bitcoin, and they decide if they include your transaction or not.  You don't need full nodes to transmit them your transaction: if you connect DIRECTLY to their nodes, they will get it.  And that was how bitcoin was designed !  Consensus is decided by those who deliver proof of work and explicitly NOT by the number of full nodes.

It is rather strange that one argues that the peer-to-peer ability to pay is compromised because Joe cannot run his full node in his basement any more (while this node never intervened in any consensus decision). but that the peer-to-peer ability would NOT be compromised by needing a lot of hubs in the LN network to agree to your transaction: hubs to which you are TIED with a payment channel which you cannot settle easily (as per definition that the on chain system is "compromised" and doesn't have, per design, the capacity for you to easily settle).  Being forced off-chain looks to me like a much higher danger to the peer-to-peer permissionlessness of transactions, than having to rely on the consensus of Proof-of-work providers only.

The current, actual reality is that all of bitcoins' consensus, including the protocol, the permission to transact, the fees, and everything, are the consensus that happens between about 20 entities, the pools, that together, have more than 99% of the decision power (PoW) under their control.  The consensus that emerges between these 20 entities is what we call "bitcoin", and bitcoin was designed to be like that.  

great post QFT

[dinofelis]

They cannot honestly and openly admit "yes we want to change Bitcoin from Satoshi's peer to peer cash into a settlement network" because it would be so radical that it would have a high probability of getting backlash from the community. Therefore, they have opted to be sneaky about it.

How do you maintain the promise of so called "peer to peer cash" to scale globally without centralizing the network due huge blocks that people cannot afford to run at home, therefore not anymore peer to peer cash but peer to corporation to peer transaction? (aka what we have already in the current baking system)

Satoshi's vision was that eventually, when the network gets large, only miners need to run full nodes.   Ordinary users can use SPV clients.

As the network gets larger, running a full node will become more expensive, but please realize that:

A)  part of that cost will be offset by the natural decline in processing and bandwidth costs.

and...

B)  By that time, Bitcoin network will be so big that it will still be very decentralized even if the cost barriers to mining increase.

You can have a different opinion than Satoshi, but I find it very suspicious that those who do aren't forthright about it.  

For about "satoshis" vision and focus on the facts.

Fact: If you raise the blocksize up to a point where people can't run their own nodes, you cannot call it a peer to peer network anymore.

Fact: if the entire consensus mechanism as it was designed, is decided by 20 entities, it is not a peer-to-peer CONSENSUS SYSTEM any more.  And that it the current reality.

Bitcoin is not about peer-to-peer communication (like, say, the Tor network).  Bitcoin is not about communication on top of the internet.  Bitcoin is about building a secured ledger of transactions according to consensus deciders, deciding what is the actual list of past transactions.  
Well, that list is about ENTIRELY decided by 20 entities.

That's bitcoin's reality.  So there's not much peer-to-peer any more.  Yes, the peer-to-peer part is between these 20 entities.  The rest is just a communication network that doesn't serve much of a purpose: whether I send my transaction through 8 hops from P2P to one of these 20 entities, or whether I send it directly with TCP/IP to them, doesn't alter much apart from some anonymity aspects, and then I can use true P2P networks such as Tor to obtain the same.

In the end, my transaction has to be accepted by one of these 20 entities, written in the unique block chain these 20 entities produce, and my counter party has to get it from them (directly, or through a proxy server-full-node) to see that my payment got accepted, confirmed, and written into the unique eternal ledger, produced by these 20 entities.

I don't see what's peer-to-peer in that system.

You have a consortium of 20 ledger-producers on one hand, and an army of customers (the users) on the other hand.

[jonald_fyookball]

but those 20 pools can only continue to operate if they more or less play by the rules.   otherwise they will dissolve.  Is there a better KNOWN system than Bitcoin (other than some unknown one someone is working on behind the scenes),

[Viper1]

Fact: if the entire consensus mechanism as it was designed, is decided by 20 entities, it is not a peer-to-peer CONSENSUS SYSTEM any more.  And that it the current reality.

Which, sadly, implies it's a failure. "The system is secure as long as honest nodes collectively control more CPU power than any cooperating group of attacker nodes.". Course one could "argue" about what the definition of "attacker" is but it is what it is at this point.

[dinofelis]

but those 20 pools can only continue to operate if they more or less play by the rules.   otherwise they will dissolve.

This is like in any industry: you have the main industries, like the car manufacturers (the pools), you have the subcontractors (the owners of mining hardware) who do in fact most of the manufacturing work, and you have the customers paying for the joke (the users, buying and selling bitcoins).

The specificity of this industry is that the product is a unique global product: the block chain, and not individual products you can buy from pool A or from pool B.  They are building one single big bridge, say, and not many independent cars.  This is what keeps the immutable rules in place: no-one can start deviating from the rules, because his pieces wouldn't be accepted by the peers, and he would lose all of his production.

Bitcoin remains immutable as long as these 20 entities (well, 5 would be sufficient in bitcoin's case) don't sit in a single room and decide something together.  They are bound to remain with the only rule set on which they have de facto agreement: the actual rules, until they can sit together and invent something more lucrative for them.  More lucrative also means: not scare away all customers.

[dinofelis]

Fact: if the entire consensus mechanism as it was designed, is decided by 20 entities, it is not a peer-to-peer CONSENSUS SYSTEM any more.  And that it the current reality.

Which, sadly, implies it's a failure. "The system is secure as long as honest nodes collectively control more CPU power than any cooperating group of attacker nodes.". Course one could "argue" about what the definition of "attacker" is but it is what it is at this point.

I think it is quite clear what an attacker is in bitcoin's case: it is an entity wanting to modify the published consensus history.
In other words, a miner building on top of the chain is never an attacker (a miner orphaning the last block isn't, either).  Someone trying to overdo the last 50 blocks, however, is an attacker.
I think one could define an attacker as someone orphaning ON PURPOSE any block older than 6 confirmations.

It is here that proof of work is a terribly BAD cryptographic security.  You can't find worse.  In as much as consensus FINDING can be done with just any sufficiently fair and random way, past consensus securing should be done with better cryptographic means than proof of work.

In fact, as long as these 20 entities are not attackers, and they have never been, the system is not a failure in the sense that it allows people to transact.  But it is not a peer-to-peer system, and it does have points of failure (for instance, law enforcement intervention).  But it seems to work.  Not as it was conceived, but it works. 

As such, my idea is that given THIS deviation from the original P2P vision, the discussions concerning "full nodes in your basement" are after the battle.  The system is not a P2P system since quite a while.  So if these entities are the ones deciding on the whole bitcoin consensus, they may just as well also operate the full nodes that everyone connects to with light wallets: it won't change the power structure of bitcoin's industry.

[Viper1]

Fact: if the entire consensus mechanism as it was designed, is decided by 20 entities, it is not a peer-to-peer CONSENSUS SYSTEM any more.  And that it the current reality.

Which, sadly, implies it's a failure. "The system is secure as long as honest nodes collectively control more CPU power than any cooperating group of attacker nodes.". Course one could "argue" about what the definition of "attacker" is but it is what it is at this point.

I think it is quite clear what an attacker is in bitcoin's case: it is an entity wanting to modify the published consensus history.
In other words, a miner building on top of the chain is never an attacker (a miner orphaning the last block isn't, either).  Someone trying to overdo the last 50 blocks, however, is an attacker.
I think one could define an attacker as someone orphaning ON PURPOSE any block older than 6 confirmations.

It is here that proof of work is a terribly BAD cryptographic security.  You can't find worse.  In as much as consensus FINDING can be done with just any sufficiently fair and random way, past consensus securing should be done with better cryptographic means than proof of work.

An attacker that controls more than 50% of the network's computing power can, for the time that he is in control, exclude and modify the ordering of transactions. This allows him to:
Reverse transactions that he sends while he's in control. This has the potential to double-spend transactions that previously had already been seen in the block chain.
Prevent some or all transactions from gaining any confirmations
Prevent some or all other miners from mining any valid blocks

I do agree, given what we now know for a certainty, that the bitcoin method of using PoW for "all things" is a major flaw.

In fact, as long as these 20 entities are not attackers, and they have never been

As of today. But that "20" will continue to shrink until someone can control 51%. It's inevitable assuming "all things" stay the same.

[jonald_fyookball]

It is here that proof of work is a terribly BAD cryptographic security.  You can't find worse.  In as much as consensus FINDING can be done with just any sufficiently fair and random way, past consensus securing should be done with better cryptographic means than proof of work.

Options are limited when it comes to distributed consensus obviously.  What's better?

What do you think of Byteball/DAG ? They have a 'finality' feature.

[dinofelis]

As of today. But that "20" will continue to shrink until someone can control 51%. It's inevitable assuming "all things" stay the same.

Even if someone was having 51% of the hashing power, that wouldn't mean that bitcoin was not functioning - only, we would now have a "CEO of bitcoin".  That said, the relationship is somewhat more complicated, because that 51% power is mostly outsourced to hardware owners.  There hardware owners cannot decide directly, (they only sell their hash rate), but they are part of the "block chain producing industry" and wouldn't want to see the system destroyed in which they are hardware-invested.   So even if a single pool had more than 51% of the hash rate, I don't think it would mean that bitcoin was compromised.  The day that that pool owner "goes beserk", the hardware owners might decide to jump to another pool if the jumping of the pool owner is not seen in their lucrative advantage.

There's no reason for a big pool owner to go beserk ; however, all miners (hardware owners and pool owners) will want to find the most lucrative way of running bitcoin for them, that's for sure.   My idea is that the current protocol is actually very lucrative to them. 

[Sr.Urbanist]

Transaction times are taking longer and fees are increasing.  Bitcoin started small.  I don't know what everybody's deal is with nodes.  It's so flippin' easy to run a full node, I didn't even know I was running three different ones (Bitcoin Core, Bitcoin Unlimited, Litecoin and DASH).  I did notice the significant investment in purchasing mining equipment - including dedicated circuits.

If the question is about SegWit/Lighting, then read the 2004 Ripple white paper by Ryan Fugger.  

If we are talking about the purpose of Bitcoin, we should remember the context of its inception and the message encrypted in the genesis block:

"The Times 03/Jan/2009 Chancellor on brink of second bailout for banks"

This was probably intended to comment on the instability caused by fractional-reserve banking and to provide people with another option:   peer-to-peer cash.  

If they really believed (Blockstream / Core Developers) in Bitcoin, they would conduct all of their business in Bitcoin.   I've got too much invested for a 20 year old, developer who lives in grandma's basement, to think s/he's an economist. If it takes two coins and a hard fork.  I was against it a few months ago and, now, I say the sooner the better.

EDIT:

Even if someone was having 51% of the hashing power, that wouldn't mean that bitcoin was not functioning ... the hardware owners might decide to jump to another pool if the jumping of the pool owner is not seen in their lucrative advantage.

Well put and insightful.  I've switched pools multiple times and even having miners on multiple pools.  Though I've recently switch them all to a BU pool because I am sick of these long confirmation times and high tx fees.

[dinofelis]

It is here that proof of work is a terribly BAD cryptographic security.  You can't find worse.  In as much as consensus FINDING can be done with just any sufficiently fair and random way, past consensus securing should be done with better cryptographic means than proof of work.

Options are limited when it comes to distributed consensus obviously.  What's better?

In bitcoin, and in several other crypto, one has confused several different functions, to bring them together in one single thing, and I think it is quite obvious now, with hindsight, that this was a bad idea.

The different functions that PoW has been assigned to, are these:

1) burning seigniorage for new coins.  The ability to print "money out of thin air and get its value" (seigniorage) has always seen as a bad trait of any monetary system ; in fact, it is the sole justification of "sound money doctrine", namely that NOBODY should be able to print money because of seigniorage.  The Austrian school considered gold, that came from "long ago" as an acceptable sound money, because nobody could "print" gold - and mining gold was about as costly as what it was worth.  If you make a new currency, of course you have to "print" it, and bitcoin's way of killing the seigniorage was exactly, mining with PoW: you got bitcoin's value, but you had wasted about the value you obtained in hardware and electricity.

==> this is a sensible use of PoW

2) allowing new users to compete for coins without being in the "old boys network".   PoW allows just anyone to create coins, without permission or acceptance or whatever.

==> this is a sensible use of PoW

3) consensus formation.  This is already somewhat more doubtful.  NORMALLY, all transactions that are transmitted over the network, should simply be part of the consensus, apart from double spends, in which case, one has to decide "arbitrarily" which of the two double spends, if any, should be part of the consensus.  One should also include, in the consensus, who had the right to win coins through PoW.

==> this has nothing to do with PoW a priori.  In case there's a doubt, a random, but agreed-upon decision should simply be taken.

In fact, consensus formation is almost trivial apart from double spends close in time so that they arrive in different orders at different parts of the network, which is the "hard part" to solve, but which could in fact be solved easily: if ever there is a double spend in a relatively close interval of time, the spending address is blacklisted, and the coins are lost for ever, which would be a serious disincentive to double spend.  Without double spends, the consensus formation is trivial: it is the full list of all transactions, and the order doesn't matter.

4) cryptographic securing of the consensus formation.  In as much as the consensus is "the biggest set of transactions", there's in fact not much securing to do.  It is the possibility of *excluding* transactions that needs cryptographic securing, so that nobody *excludes* old transactions.  In as much as it would be "the biggest list", no cryptographic securing is in fact needed, but if the list has special requirements, it needs to be secured once the list is accepted generally.

In fact, it is the severe consensus mechanism, to put everything in a block chain, that needs cryptographic securing, and here, PoW is really bad.  Just any digital signature scheme would be better because cryptographically more secure. The whole of bitcoin's drama comes from the fact that the PoW, which was meaningful for coin creation, was also used as a very severe consensus decision mechanism, and as a cryptographic securing of this very severs consensus decision, leading to the mining industry and the uselessness of network nodes.