4 ms·

I see 3 scenarios right now: 1. Bitcoin + lighting network solves everything 2. Bitcoin's dominance ends and a more scalable payment-focused currency start to

by nadam 9y ago

I see 3 scenarios right now: 1. Bitcoin + lighting network solves everything 2. Bitcoin's dominance ends and a more scalable payment-focused currency start to dominate (like Dash) 3. Bitcoin remains the main store of value, and people will convert some of their bitcoins to an altocin to do daily paymnt where there will be debates on the level of decentralization, but is designed to be extremely high (Visa level) tx/sec, and very low latency. My bet is currently EOS. I still think any of these can become true, but I am more and more predicting (3) to be the case. (That is why my main investments are BTC and EOS, but I also have some Dash in case (2) becomes true.)

fragsworth 9y ago
> 1. Bitcoin + lighting network solves everything As far as I can tell, the Lightning Network's design isn't robust enough to solve everything. It allows for an arbitrary number of back-and-forth transactions between any two parties. It will be good, but it's not clear to me that this extends to solving all of the scalability problems.
- Frogolocalypse 9y ago
  I would suggest actually learning how it works before commenting on it. Lightning allows users to take a single bitcoin transaction, and use it over a finite space of time, in order to shift bitcoin back and forth, with no or vanishingly small transaction fees, as iou's between what are called 'lightning channels'. The actual transaction size will be the same as a normal transaction, with a flag that says it can only be committed to the blockchain at a specific time (or block depth). In effect, these lightning transactions are cryptographically signed 'iou's. The 'thousands' of back-and-forth iou changes are discarded when the channel is closed (massive privacy dividend), and only the aggregated change is committed to the blockchain. This is the great thing about lightning. It is a single transaction with as many inputs and outputs (one/two) as defined in the original transaction. Only the transaction values change. Consider a channel link of this : John->Bob->Sally->Anne. Only the people next to each other in the chain have a 'channel open' with each other (John/Bob, Bob/Sally, Sally/Anne). Assuming each of the channels between all of the participants have adequate funds in order to achieve this, if John wanted to send 1btc to Anne. A transaction ledger would read : John->Bob channel : John -1btc, Bob +1btc Bob->Sally channel : Bob -1btc, Sally +1btc Sally->Anne channel : Sally -1btc, Anne +1btc. Anne now has 1btc more, and John has 1btc less, but the net effect in each channel is simply a modification to the distribution of bitcoin in that channel. Imagine this happening, back and forth, and extended to thousands, tens of thousands, and millions of people. As long as there's a chain between them, you use the chain. You might even have your transaction split over multiple chains. For links that don't have a chain, you create a channel on-the-fly. Each channel, therefore, only ends up with the same single input, and the same two outputs, just differing values. It's significantly more complex than that (transaction routing, and channel closure mechanics), but that's the idea. I wish I could remember the lightning dev that explained it. It really is very clever. I haven't actually questioned about the effect of having channels with multiple inputs (John/Sally) and outputs (Bob/Anne) at inception. Technically, I can't see any reason why this wouldn't be possible. Just a much weirder set of ramifications. I would suggest that the method for creating ln channels is most likely : Carol wants to pay Bob 0.5btc. She creates a ln channel with 1btc, of which Bob is allocated 0.5btc, and carol is allocated 0.5btc. Bob gets paid 0.5btc that is accessible once confirmed. Carol has 0.5btc that is now accessible once confirmed. That way the ln can be bootstrapped by existing txns. I'd also suggest that it would be easier to understand if Bob were Bob Inc. The incentive will be for the buyer to reduce txn fees. If Carol uses a store (coffee?) she will want to reduce paying those fees. So she creates a channel with 0.1btc and 0.001 is allocated to Bob Inc. For the coffee she buys today, coffee + txn fee. Ever more? Zippo txn fee. When your channel is almost zero, you buy bitcoin, it is delivered back to you by the rebalancing of that channel. And because it's Bob Inc the path through the channels from a source of bitcoin is reduced, maybe even two hops ( exchangeA/exchangeB -> exchangeB/Bob Inc. -> Bob Inc./Carol).
  - SnowProblem 9y ago
    This is too complicated. I would gladly pay small txn fees to avoid having to manage payment channels.
    - Frogolocalypse 9y ago
      It's no more complex than running any other bitcoin wallet. It is, after all, bitcoin. The only difference really is that lightning is instantaneous and immutable, where blockchain transactions require you to wait until a transaction has been confirmed in the relevant blockchain in order to be secure. And no-one will force you to use it. If you choose to continue using bitcoin transactions natively, no-one is going to stop you.
    - ericb 9y ago
      The surface area of lightning scares me. Don't get me wrong, I hope it works, but it just sounds really unlikely to.
  - nextstep 9y ago
    If John wants to pay Sally, but the cheap/practical way is to use the existing path of channels through Bob and Carol and whatever, then it’s not peer-to-peer anymore! I don’t get why this fundamental architecture change is so appealing to some people.
    - pmorici 9y ago
      Because Bob and Carol will make money by routing the transactions. They will effectively be banks.
    - Frogolocalypse 9y ago
      Because you don't understand it. Do you think about the connections between your computer and the server you are reading this page from?
- em3rgent0rdr 9y ago
  Care to elaborate?
- gwillen 9y ago
  What you're describing is a payment channel. The Lightning Network is that _plus_ protocols for negotiating the routing and chaining of payments across multiple channels.
_red 9y ago
>3. Bitcoin remains the main store of value I think this is probably right. And I would add 2 things (a) The size and the hash power of BTC give it a HUGE advantage. If you want "immutable proof that you own this specific decentralized block" then BTC is the only game in town. And that comes at a cost (hence high fees because of lots of electricity, computers, ASIC fabs, etc). (b) Secondly (and I realize this analogy has been beat to death...but), Gold didn't want Silver to exist. The gold industry naturally wanted to monopolize every transaction, but like Bitcoin, it turned out that their very strengths were also simultaneously their weaknesses. Its an immutable law of economics that BTC isn't going to be able to be all things to all people, so I think its natural that among their various options, they will choose to be the "Store of Value". Ultimately thats the equivalent of being a reserve currency. By capturing the store of value corner, it means that alternatives like ETH, LTC, DASH, IOTA, et al will be valued against and perform a support function to Bitcoin. In a sense, by being the Store of Value, the other cryptos perform support roles that further entrench your position.
- pishpash 9y ago
  If you owned a national currency, you would have the legal taxing authority and guns behind your papers. If you owned gold, you would have a shiny incorruptible metal in your hand to hand over for industrial if not artistic uses, in a pinch. Heck, even if you owned a potato, you could eat it -- and in dire circumstances that has a huge amount of value and maybe you can convince somebody to trade something else valuable. If you "own" bitcoins, you have some numbers graffitied over people's hard drives. Yes they can't change it because electricity, and the numbers only get so big, and you have some different numbers that can be turned into the graffitied numbers by one particular algorithm that people like at the moment. Without transactions, these properties are completely arbitrary. I don't see how by themselves they can give anything approaching the necessities for a store of value. The truth is, without competitive transactional uses bitcoin has no fundamental value. Without fundamental value it can't be much of a store of value, since literally anything else without fundamental value can be equally good. You can store something like the goodwill of the network, as now, which isn't nothing, but it can be notoriously fickle.
jackmodern 9y ago
#3 is how I have allocated my bets as well.

by nadam 9y ago

fragsworth 9y ago

> 1. Bitcoin + lighting network solves everything As far as I can tell, the Lightning Network's design isn't robust enough to solve everything. It allows for an arbitrary number of back-and-forth transactions between any two parties. It will be good, but it's not clear to me that this extends to solving all of the scalability problems.

Frogolocalypse 9y ago
I would suggest actually learning how it works before commenting on it. Lightning allows users to take a single bitcoin transaction, and use it over a finite space of time, in order to shift bitcoin back and forth, with no or vanishingly small transaction fees, as iou's between what are called 'lightning channels'. The actual transaction size will be the same as a normal transaction, with a flag that says it can only be committed to the blockchain at a specific time (or block depth). In effect, these lightning transactions are cryptographically signed 'iou's. The 'thousands' of back-and-forth iou changes are discarded when the channel is closed (massive privacy dividend), and only the aggregated change is committed to the blockchain. This is the great thing about lightning. It is a single transaction with as many inputs and outputs (one/two) as defined in the original transaction. Only the transaction values change. Consider a channel link of this : John->Bob->Sally->Anne. Only the people next to each other in the chain have a 'channel open' with each other (John/Bob, Bob/Sally, Sally/Anne). Assuming each of the channels between all of the participants have adequate funds in order to achieve this, if John wanted to send 1btc to Anne. A transaction ledger would read : John->Bob channel : John -1btc, Bob +1btc Bob->Sally channel : Bob -1btc, Sally +1btc Sally->Anne channel : Sally -1btc, Anne +1btc. Anne now has 1btc more, and John has 1btc less, but the net effect in each channel is simply a modification to the distribution of bitcoin in that channel. Imagine this happening, back and forth, and extended to thousands, tens of thousands, and millions of people. As long as there's a chain between them, you use the chain. You might even have your transaction split over multiple chains. For links that don't have a chain, you create a channel on-the-fly. Each channel, therefore, only ends up with the same single input, and the same two outputs, just differing values. It's significantly more complex than that (transaction routing, and channel closure mechanics), but that's the idea. I wish I could remember the lightning dev that explained it. It really is very clever. I haven't actually questioned about the effect of having channels with multiple inputs (John/Sally) and outputs (Bob/Anne) at inception. Technically, I can't see any reason why this wouldn't be possible. Just a much weirder set of ramifications. I would suggest that the method for creating ln channels is most likely : Carol wants to pay Bob 0.5btc. She creates a ln channel with 1btc, of which Bob is allocated 0.5btc, and carol is allocated 0.5btc. Bob gets paid 0.5btc that is accessible once confirmed. Carol has 0.5btc that is now accessible once confirmed. That way the ln can be bootstrapped by existing txns. I'd also suggest that it would be easier to understand if Bob were Bob Inc. The incentive will be for the buyer to reduce txn fees. If Carol uses a store (coffee?) she will want to reduce paying those fees. So she creates a channel with 0.1btc and 0.001 is allocated to Bob Inc. For the coffee she buys today, coffee + txn fee. Ever more? Zippo txn fee. When your channel is almost zero, you buy bitcoin, it is delivered back to you by the rebalancing of that channel. And because it's Bob Inc the path through the channels from a source of bitcoin is reduced, maybe even two hops ( exchangeA/exchangeB -> exchangeB/Bob Inc. -> Bob Inc./Carol).
- SnowProblem 9y ago
  This is too complicated. I would gladly pay small txn fees to avoid having to manage payment channels.
  - Frogolocalypse 9y ago
    It's no more complex than running any other bitcoin wallet. It is, after all, bitcoin. The only difference really is that lightning is instantaneous and immutable, where blockchain transactions require you to wait until a transaction has been confirmed in the relevant blockchain in order to be secure. And no-one will force you to use it. If you choose to continue using bitcoin transactions natively, no-one is going to stop you.
  - ericb 9y ago
    The surface area of lightning scares me. Don't get me wrong, I hope it works, but it just sounds really unlikely to.
- nextstep 9y ago
  If John wants to pay Sally, but the cheap/practical way is to use the existing path of channels through Bob and Carol and whatever, then it’s not peer-to-peer anymore! I don’t get why this fundamental architecture change is so appealing to some people.
  - pmorici 9y ago
    Because Bob and Carol will make money by routing the transactions. They will effectively be banks.
  - Frogolocalypse 9y ago
    Because you don't understand it. Do you think about the connections between your computer and the server you are reading this page from?
em3rgent0rdr 9y ago
Care to elaborate?
gwillen 9y ago
What you're describing is a payment channel. The Lightning Network is that _plus_ protocols for negotiating the routing and chaining of payments across multiple channels.

_red 9y ago

>3. Bitcoin remains the main store of value I think this is probably right. And I would add 2 things (a) The size and the hash power of BTC give it a HUGE advantage. If you want "immutable proof that you own this specific decentralized block" then BTC is the only game in town. And that comes at a cost (hence high fees because of lots of electricity, computers, ASIC fabs, etc). (b) Secondly (and I realize this analogy has been beat to death...but), Gold didn't want Silver to exist. The gold industry naturally wanted to monopolize every transaction, but like Bitcoin, it turned out that their very strengths were also simultaneously their weaknesses. Its an immutable law of economics that BTC isn't going to be able to be all things to all people, so I think its natural that among their various options, they will choose to be the "Store of Value". Ultimately thats the equivalent of being a reserve currency. By capturing the store of value corner, it means that alternatives like ETH, LTC, DASH, IOTA, et al will be valued against and perform a support function to Bitcoin. In a sense, by being the Store of Value, the other cryptos perform support roles that further entrench your position.

pishpash 9y ago
If you owned a national currency, you would have the legal taxing authority and guns behind your papers. If you owned gold, you would have a shiny incorruptible metal in your hand to hand over for industrial if not artistic uses, in a pinch. Heck, even if you owned a potato, you could eat it -- and in dire circumstances that has a huge amount of value and maybe you can convince somebody to trade something else valuable. If you "own" bitcoins, you have some numbers graffitied over people's hard drives. Yes they can't change it because electricity, and the numbers only get so big, and you have some different numbers that can be turned into the graffitied numbers by one particular algorithm that people like at the moment. Without transactions, these properties are completely arbitrary. I don't see how by themselves they can give anything approaching the necessities for a store of value. The truth is, without competitive transactional uses bitcoin has no fundamental value. Without fundamental value it can't be much of a store of value, since literally anything else without fundamental value can be equally good. You can store something like the goodwill of the network, as now, which isn't nothing, but it can be notoriously fickle.

jackmodern 9y ago

#3 is how I have allocated my bets as well.