Profile

Are Ethereum's block times better than Bitcoin's?

Are Ethereum's block times better than Bitcoin's?

Ethereum likes to tout its 12 second block times as one of its primary features. But is it really a feature? And what is the opportunity cost? This question was ...

Transcription

What's up, party people? Chris DeRose here, community director of the Counterparty Foundation. Today I'm shooting this video on Miami beach. It's on the north end of the beach. It's Miami beach still but it's getting into not quite Hollywood I don't know where it is just over there. Nonetheless, it's a beautiful day and it's a good place to talk about Etheorems block times. More specifically, how are the Ethereums of block times improvement over Bitcoins? Or aren't there? This question was asked by mmortal03.

Mmortal03 this ones for you. So, I told you a lot about the Ethereum and I follow the project off and on. But certainly it's a crucial time for the project right now. I think it's time we start talking about some of their design decisions. So right off the bat I know that there's a much ballyhooed sort of representation of their block times. This notion that, I don't know what their at now, 12 second block times, or minute block times, or something like, are wonderful and are a major improvement.

Now, I'm very very sceptical of those claims for a lot of reasons. First and foremost, you got to understand, especially if you're not a programmer that the block times themselves could be any number that you would have wanted in Bitcoin, it's just more important for the network to be on the same page. To have this innovation that is faster block times, typically consists of like making a couple of changes to the source code. Like coin did it I think and it took them a very trivial amount of time, maybe there's five spots they have to change it. So it's not like some R&D mission or some sort of laboratory work that resulted in this great innovation that was a faster block interval. The reason why we have ten minute block times at Bitcoin is because, quite simply, they are very secure and they have fewer forks.

So, if you have a block interval that is as low as 12 seconds, or some such thing you will have more forking. Forking is when the network splits because two people got the same result at approximately the same time. At that point consensus is not on the same unitary conclusion, but there are in fact two consensuses fighting at that point that blocks get larger and larger until they eventually fold onto one that is faster arriving than the next. So when you have a shorter confirmation interval, those times then would exert a great propensity for the blocks to split. The problem when you do that of course is that the network is not in a state of consensus. They may be very similarly represented blocks, but they are not identical blocks, and for people to invest in one branch or another is slightly risky.

So when you have a shorter block time, right off the bat there's more risk in that block time. And a lot of those factors have to do with the number of people mining, the speed of internet connections, the speed of blocks, and the speed of network connections and some such thing, bandwidth and what not. So, in the case of Bitcoin, it's conceivable that maybe one day, and I don't think it's very likely, but it's conceivable that they would lower the block times where we don't do it because it's just not needed. For people who seem to think that instant confirmations are what we need, we have solutions like that in Bitcoin. We have in the form of insurance, by the way I forgot what bibbet was, bib 70 I believe, and we will probably have in other forms, consisting of people monitoring the mempool. There's all kinds of stuff.

Lightning network even actually tackles this to a significant degree. Side chains may also tackle this to a significant degree, seeing how that project goes. There are many solutions, but keeping the heartbeat up ten minutes adds a lot of security because that ten minute interval isn't going to fork very often and there's a crap load of action power that goes into every one of those blocks, which yes is somewhat analogous to having a more spectrum sort of representation of block validity in the terms of hashing strength. But, nonetheless is much harder to unroll Bitcoin blocks than it is ether blocks and all these things. It's important to recognize when you're evaluating Ethereums claims that these things are not without their opportunity costs, when you make these decisions, there's a reason that it existed in the first place and it's very easy to criticize Bitcoin for saying that block intervals are not big enough or some such thing, when you don't know the rest of the story and the rest of the story is usually pretty good. All of the problems that a Ethereum is going to have in the next couple years, how ever many years it may be, are problems that we've talked about at length at Bitcoin, that have been solved at length at Bitcoin, and that have been worked on at length at Bitcoin.

And the Ethereum community is just now starting to explore them. So, when they make these changes, like block intervals that are faster, they don't really know the ramifications yet, they don't quite know where it's going. I think it's important that people understand that it's probably not an innovation. What's probably going to happen is you're going to have more forking, you'll have less reliability in the Ethereum Chain, and on and on. This doesn't even get into the notions of what is the Ethereum Chain, because after all even that seems to be in a state of flopsis. The developers have announced that this is just one phase in a multiphase of deployment because of the deficiencies they are going to have immediately with the system.

So, I don't know we'll see. But hey, maybe you're into Ethereum and you want to comment on this issue, by all means use the comments below, I'd love to hear your opinion and check out my other videos and I'll talk about Ethereum on occasion, certainly, Counterparty is my main interest, but we use Serpent, as well as the Ethereum, so at some level there's a lot of overlap and I certainly would love to have you around. So subscribe to the channel, tweet me on Twitter @trosetech, and stick around. Later party people.