All coins built upon PoW suffer the same scaling issues as bitcoin. Since Bitcoin is the most popular, it hit the limitations first. Second layer technologies such as Lightning Network and Rootstock do solve various problematic use cases. Bitcoin is the most widely held cryptocurrency and is the only cryptocurrency with second layer technologies being rolled out into production as we speak.
Lightning Network exceeds the throughput of VISA. The scaling problem is being addressed -- just not in the naive "let's throw away decentralization and increase the blocksize" way.
I've heard mentions about the lightning network quite a lot in previous years, but never got what the implementation was. I get it it's "offchain transactions", but how does it relate to bitcoin, exactly? Is it "sidechains", like lisk (each third party managing its own blockchain), or is it simply "outsourcing" money and transactions and waiting for a general report from the trusted party to engrave it in the blockchain?
It seems to work by setting up channels of funds between entities. To create a channel and fund it requires a bitcoin transaction. Payments from the fund to the recipient can then proceed off-chain to a maximum of the fund amount, then another on-chain transaction settles it.
This sounds somewhat cumbersome, but AFAICT the gains apparently come from routing between channels, so if A wants to pay B, and channels exist from A->C and C->B then payments can route that way, without the need for more on-chain transactions.
To me this sounds really quite complex, and also like it's going to involve intermediary 'hubs' which will process off-chain transactions and... well it sounds like banks and centralisation again, something the BTC crowd apparently hate.
I see, thanks for the explanation. As is, it kind of seems to defeat the idea of the blockchain, dropping it to do a meaningful amount of transactions. But it's well over my head, I'll wait and see how it plays out.
> Bitcoin effectively froze when Blockstream decided against increasing block size.
You mean when the Bitcoin Core implementation of the Bitcoin protocol decided decentralization was an important feature of the currency and transaction compression should be the prioritized means of scaling.
That's the "glass is very half-full" way of putting it :)
They could've easily gone both ways: block size increase to 2mb, segwit and lightning network. Hell, Litecoin is well on its way to becoming the de-facto transfer currency between exchanges because it iterates so much quicker (and those 2.5 minute blocktimes sure help).
That’s a cute theory, but housing costs are much more related to the cost of money. When the federal reserve makes money cheap, asset prices increase. Since mortgage rates are at historic lows, people simply borrow as much as possible until their monthly payment is at some threshold. If you want housing prices to return to historic norms, you need to ask the Fed to first return the cost of money to normal.
I'd have to agree. :) Many of these new technologies like electron and node are more RAM hogs than anything else. Sure they boost productivity but you need a really high end laptop with at least 8GB of RAM to run anything smoothly.
C is neither here nor there. Really anything compiling down to the platform code would improve things. Running a browser environment for these applications is just not as efficient. But also the UI tooling outside of web is also not very good. While you're right about productivity, we also just happen to have a large workforce that knows the webstack very well, but not so much QT or UWP or AppKit.
That doesn't seem to me like a fundamental law. Certainly with "internal blockchains" or other cases where at least a majority of the operators are trusted, mining and tokens are completely unnecessary.
A blockchain is a decentralized trustless ledger. If the operators are trusted, you can just use a database. It’s much more efficient. It’s pretty clear from the comment that you don’t understand Bitcoin or why, when people say “we like blockchain but not bitcoin”, that they’re talking nonsense.
> If the operators are trusted, you can just use a database. It’s much more efficient.
This is true! But databases don't have that magic "blockchain" tech.
David Gerard's book Attack of the 50 Foot Blockchain touches on the ridiculousness of this idea, actually. And, yes, it's ridiculousness, but it's what some people are trying to sell.
So should I take it that you're one of the people who thinks blockchains have little applications outside of currency?
No, I think blockchains provide many useful functions. However, providing a token with value is a necessary feature. I’m very excited about things like Rootstock which add smart contract functionality to the bitcoin blockchain.
As far as I can tell, the purpose of tokens is to provide an incentive for running the ledger client and processing transactions: mining for uptake and transaction fees to maintain scale long-term. You need as many people to run the client as possible.
If you can find some other reason for processing transactions and can guarantee that the population is big enough then the coin becomes unnecessary.
Distributed ledgers (e.g. blockchain) are based on security by game theory. It's financially rewarding to agree with each other, and financially very costly to disagree or attack it. You can't have game theory in the protocol without intrinsic rewards, as far as I know.
I assume you mean extrinsic reward. Yeah, it's unrealistic to expect people to burn CPU cycles processing the ledger client just for the fun of seeing the dials on their power meter spin faster.
An extrinsic reward doesn't have to be financial though, it's anything that you want to get out of an activity that isn't part of the activity itself.
If the blockchain ledger was a part of a greater whole and the client software of which it forms a part was somehow intrinsically fun, useful or needed, the token itself wouldn't be necessary.
