4 ms·

Interesting comments! Do you mind sharing your thoughts on why proof of stake won’t happen?

svennek 5y ago
IMHO because it depends on uncloneability of digital assets.. Which I believe is not possible. The whole thing that makes cryptocurrencies work now is the un-cloneability of the amount of work done (which is why the 51% problem exists in the first place). The only good way of doing it would be through a cartel with strict enforcement of hashrates etc.. And welcome to the financial banking system...
- dom96 5y ago
  I'm also curious about this but don't really understand this explanation. Can you elaborate more on what "uncloneability of digital assets" means exactly and why it's a pre-requisite of proof-of-stake? Happy to take some links if there are any that explains this well.
  - svennek 5y ago
    All kind of trade depends on value and value depends (mostly) on scarcity. In the real world cloning money (historically precious metals) is really really hard (and btw. much easier with paper money that the metals that have a basket of physical attributes that cannot be emulated at once). Digital bits on the other hand are indefenately cloneable (by design, that is one of the base properties of "digital"). Just for you to see my text here, along the way somewhere around 20 perfect copies were made by various routers (i.e. computers)..
- vbs_redlof 5y ago
  You're confusing concepts and blockchain fundamentals. "Uncloneability" is the double-spending problem solved by proof-of-work/hashcash scheme. The 51% attack is not the problem, it's just a requirement for a PoW based blockchain to work. Just like how you have 1/3 honest node requirements in traditional BFT distributed database designs.
  - svennek 5y ago
    So cloneable cryptocoins is a thing? Where do I sign up wink. IMHO cloneablitiy IS the problem, PoW is just a (very, very unfortunate) solution to that. PoS (stake as in possession, not pledging) is IMHO inherently impossible, so PoW is all we got, now and for ever...
pcthrowaway 5y ago
Proof of stake is already happening. It's on Ethereum (though they haven't switched over to the beacon chain from proof-of-work yet), and it's on many others. If you look at the top 100 cryptocurrencies by market capitalization, maybe bitcoin and 1 are 2 others are proof of stake. There are a bunch of 'tokens' cryptocurrencies which exist on top of another blockchain, but the remainder of the layer 1 blockchains and all of the layer 2s that use a consensus mechanism are using proof of stake now. Arguing against crypto because of proof of work is like arguing that computers won't see mass adoption because vacuum tubes make them impractical to fit inside of a typical home.
- svennek 5y ago
  Last I looked the so-called proof-of-stake was more a system where somebody put up a stake to be witnesses (in lack of a better word). When I say proof of stake, I mean it in the way of "proof of possession"
  - pcthrowaway 5y ago
    Proof of stake is a decentralized consensus mechanism that is (usually) trustless, and prevents attacks like double spending. Most notably, Solana, Cardano, Polkadot, Avalanche, Terra, Algorand, Polygon, Elrond, and Fantom are all using permissionless networks where nodes which have some amount of the native token of those blockchains staked produce blocks and validate transactions
    - svennek 5y ago
      Your answer reads like a marketing blurb and only tells about the goal (on which we agree) and not the mechanism (where we disagree). I tried reading a few of the coins you mentioned and all of them were witness ("validator") based... They still do not prove that they have something in possession, they stake their own coin on being honest... BIG difference. Do they allow you to prove that you (and you alone) have possession/control of the coin itself, and if so, how? (That is the part that I consider impossible). The whole problem is that proof-of-stake has been redefined by the new PoS networks and does have the original meaning of "stake" (i.e. sole possessor) anymore. I can prove to you that I have access to a gold coin by taking it out of my purse and showing it to you. You can do three simple tests (weight, volume and ringing) that ensures that it cannot be any other material than pure gold (due to any material that is both cheaper and heavier than gold is tungsten and that does not ring). You can also be sure that it is not a clone of another coin so that it "just disappears" or gets double spent. There is also no witness/validator that can break your trust and betray you. Every POS network I have ever looked into (and I admit that it has been a while) relies on witnesses. And that removes my way of knowing whether the witnesses lie or not. Is there any of the socalled PoS networks that actually allow you to prove that you have exclusive possession/control of the digital coin in question without any witnesses ('validators')? If so, their founders can already book their ticket to the Nobel Gala next year, as that would AFAIK be a feat in the same scale as disproving gravity...

IMHO because it depends on uncloneability of digital assets.. Which I believe is not possible. The whole thing that makes cryptocurrencies work now is the un-cloneability of the amount of work done (which is why the 51% problem exists in the first place). The only good way of doing it would be through a cartel with strict enforcement of hashrates etc.. And welcome to the financial banking system...

dom96 5y ago
I'm also curious about this but don't really understand this explanation. Can you elaborate more on what "uncloneability of digital assets" means exactly and why it's a pre-requisite of proof-of-stake? Happy to take some links if there are any that explains this well.
- svennek 5y ago
  All kind of trade depends on value and value depends (mostly) on scarcity. In the real world cloning money (historically precious metals) is really really hard (and btw. much easier with paper money that the metals that have a basket of physical attributes that cannot be emulated at once). Digital bits on the other hand are indefenately cloneable (by design, that is one of the base properties of "digital"). Just for you to see my text here, along the way somewhere around 20 perfect copies were made by various routers (i.e. computers)..
vbs_redlof 5y ago
You're confusing concepts and blockchain fundamentals. "Uncloneability" is the double-spending problem solved by proof-of-work/hashcash scheme. The 51% attack is not the problem, it's just a requirement for a PoW based blockchain to work. Just like how you have 1/3 honest node requirements in traditional BFT distributed database designs.
- svennek 5y ago
  So cloneable cryptocoins is a thing? Where do I sign up wink. IMHO cloneablitiy IS the problem, PoW is just a (very, very unfortunate) solution to that. PoS (stake as in possession, not pledging) is IMHO inherently impossible, so PoW is all we got, now and for ever...

pcthrowaway 5y ago

Proof of stake is already happening. It's on Ethereum (though they haven't switched over to the beacon chain from proof-of-work yet), and it's on many others. If you look at the top 100 cryptocurrencies by market capitalization, maybe bitcoin and 1 are 2 others are proof of stake. There are a bunch of 'tokens' cryptocurrencies which exist on top of another blockchain, but the remainder of the layer 1 blockchains and all of the layer 2s that use a consensus mechanism are using proof of stake now. Arguing against crypto because of proof of work is like arguing that computers won't see mass adoption because vacuum tubes make them impractical to fit inside of a typical home.

svennek 5y ago
Last I looked the so-called proof-of-stake was more a system where somebody put up a stake to be witnesses (in lack of a better word). When I say proof of stake, I mean it in the way of "proof of possession"
- pcthrowaway 5y ago
  Proof of stake is a decentralized consensus mechanism that is (usually) trustless, and prevents attacks like double spending. Most notably, Solana, Cardano, Polkadot, Avalanche, Terra, Algorand, Polygon, Elrond, and Fantom are all using permissionless networks where nodes which have some amount of the native token of those blockchains staked produce blocks and validate transactions
  - svennek 5y ago
    Your answer reads like a marketing blurb and only tells about the goal (on which we agree) and not the mechanism (where we disagree). I tried reading a few of the coins you mentioned and all of them were witness ("validator") based... They still do not prove that they have something in possession, they stake their own coin on being honest... BIG difference. Do they allow you to prove that you (and you alone) have possession/control of the coin itself, and if so, how? (That is the part that I consider impossible). The whole problem is that proof-of-stake has been redefined by the new PoS networks and does have the original meaning of "stake" (i.e. sole possessor) anymore. I can prove to you that I have access to a gold coin by taking it out of my purse and showing it to you. You can do three simple tests (weight, volume and ringing) that ensures that it cannot be any other material than pure gold (due to any material that is both cheaper and heavier than gold is tungsten and that does not ring). You can also be sure that it is not a clone of another coin so that it "just disappears" or gets double spent. There is also no witness/validator that can break your trust and betray you. Every POS network I have ever looked into (and I admit that it has been a while) relies on witnesses. And that removes my way of knowing whether the witnesses lie or not. Is there any of the socalled PoS networks that actually allow you to prove that you have exclusive possession/control of the digital coin in question without any witnesses ('validators')? If so, their founders can already book their ticket to the Nobel Gala next year, as that would AFAIK be a feat in the same scale as disproving gravity...