Hacker News Reader

Comment by emef | Hacker News Reader

Transaction cost and speed are enormous problems for bitcoin, not sure why that would only be mildly appealing to solve. Not vouching for Libra, but my understanding of bitcoin was that it caps out at something ridiculously low like 11 transactions/second, which each cost multiple USD. Clearly that is the bottleneck that needs to be addressed by whatever would lead crypto moving forward (I remember reading visa processes more than 10k transactions/s).

View on HN

Bitcoin is the oldest and thus technically least advanced cryptocurrency. There are plenty of coins that are much better than Bitcoin in terms of privacy and scalability; Libra should be compared against these.

Wrong. Bitcoin will use Lightning Network for most everyday transactions and it's speed is only limited by network liquidity and bandwidth

Genuinely curious: Does the lightning network provide all of the same guarantees that the standard Bitcoin network provides? What are the downsides? Why isn't the lightning network used all the time instead of the Bitcoin protocol?

Lightning is new, so that's part of the adoption story. It suffers from a centralization risk similar to banks (viz a viz using a well-connected intermediary via a wide channel to transmit to arbitrary addresses off chain with less pain that establishing a dedicated channel directly).

The downside is that there is no market for watchtowers yet (so you need be online while having channels open).

Another disadvantage is that if you control your key, you must be online. Like, if you run an HTTP server, you must have a machine. With plain Bitcoin you can spend to a pubkey and it doesn't have to be online

Having to be online to be secure is a pretty big hole, don’t you think? Sure you could outsource it to some trusted party, but that feels awfully bank-like and I thought bitcoin was all about being your own bank. Am I wrong?

Watchtowers are not trusted parties and you do not have to be always online to be secure. There is a configurable period where you can broadcast a revocation transaction. Watchtowers can do it for you but it's designed so they cannot frame you. The penalty for broadcasting an out of date transaction is loss of all funds in the channel.

To summarize.

1. Attacks are extremely unlikely in the first place.

2. Watchtowers can prevent them if they do happen.

3. Watchtowers are trustless. They in no way resemble banks.

4. Without a watchtower you are not required to be always online to be secure.

If I am not online and don't use a watch tower, how is the breach remedy transaction published in time?

How does the transaction appear on the blockchain if you're not online?

If you are not online, you can't receive or send at all (this is why I said "Another disadvantage")

The channel state appears on the blockchain once either party closes the channel.

It requires some additional investment on the user's part.

Think of it like opening a secured credit card with cash. Faster and more convenient to use, since you can now use the credit card networks instead of the cash, but requires tying up the funds you want to use to pay with.

Also, the merchant/recipient has to be able to accept the payments, and the software is new / in beta still.

Too bad you still need to use the blockchain to get into and off of the LN. which means if BTC got widely adopted you’d still run into major scaling limits. Of course that is assuming LN will ever exist. Which it won’t. It “exists” only a way for shysters to dodge the “bitcoin can’t scale” question when pitching the blockchain to the next batch of rubes.

Also, there is a certain unavoidable irony in suggesting the way to scale bitcoin is to avoid using it.

Unfortunately very few alternatives have been created which ticks the same boxes as bitcoin, but some with speed improvements. It's too bad that so many developers have stranglehold their innovations by premining, receiving development tax, keeping a percentage of all coins by themselves. Or lacking a good process for protocol innovation like BIP. Because there are some really shiny coins out there.

I thought that transaction cost and speed was a problem, but I am not so sure anymore. There are second layer solutions, like Lightning Network. And it is also possible to use current solutions like Visa, through pre-filled cards or debit cards. Of course not optimal.

It's not that long ago when we got to choose network protocols in games for example, IPX/SPX in favor of TCP/IP because of speed. Maybe it is a similar time, we'll see.

There is an extremely easy way to get over that issue. It has already been solved for all fiat currencies. When big banks transfer dollars between one another, their balances at the Fed get updated to reflect this. But this doesn't happen for 99.9% of transactions.

Reply on news.ycombinator.com