Hacker News Reader

> Mining difficulty is scaled to the hashpower of the network. In the long run the EV of ten minute time between blocks is maintained.

The key part here is “in the long run”. IIRC the difficulty is only adapted after a certain number of mined block, so in case of a mining power collapse you still need to mine blocks at the old difficulty for a while, which would be really slow in the aforementioned scenario before a really long time has elapsed (more than a thousand years).

Anyway, in such scenario, you would probably not have a worldwide reliable network anymore, so the whole concept of bitcoin would just be be obsolete.

> the block size could never be touched without ruining the vaguely-defined “decentralization”

Fun fact: All forks of Bitcoin which attempted to remove it actually ruined their decentralization in the process and are now either dead or valued <1/100 of Bitcoin. Seems like the market values a stable and secure Layer 1 over altcoins promising free lunches.

The market capitalization of a cryptocurrency has no relation to its utility as a currency, except insofar as a volatile currency - particularly in the downward direction - hampers its usage as a currency without converting from USD at the time of tx

"Popularity contests don't think much of forks" seems like a much stronger thesis to me.

The second most popular (i.e. valuable) cryptocurrency (ETH) is a fork of Ethereum, while the original (now called ETC) is practically dead.

I would disagree with calling "ethereum" the fork. But even accepting that premise, it's still an example where exactly one version gets to live, and the one made by a group of average people is the one that died. There is effectively no forking allowed, and implementing a fork of bitcoin is just a slipshod alternative to BIPs.

If you redefine 'forks' to refer to the minority chains you are of course correct that "popularity contests don't think much of forks", but this would be a) tautological and b) not the correct use of a well-defined technical term.

ETC is not a fork and it wasn't 'made' by anyone. It is just the original, unaltered transaction history of Ethereum. ETH was 'made' by introducing a hard-fork and altering parts of Ethereum's transaction history, diverging from prior consensus rules.

> tautological

It's not tautological to say that you get 100% or 0%. You can't effectively split the population. It almost happened with ethereum, and has never really come close with bitcoin.

It's also not tautological to say "The thing called 'bitcoin' at the time has always been the winner. The thing called 'ethereum' at the time has always been the winner."

> not the correct use of a well-defined technical term

If you want to be really technical, every single update is a fork, and every single time two blocks are mined at the same time it's a fork.

I would rather go with what people were actually calling 'bitcoin' or 'ethereum' the day after any forking of the chain. I don't think it's reasonable to say that we're on a fork of a fork of a fork of a fork of a fork of a fork of bitcoin or ethereum.

> ETC is not a fork and it wasn't 'made' by anyone. It is just the original, unaltered transaction history of Ethereum. ETH was 'made' by introducing a hard-fork and altering parts of Ethereum's transaction history, diverging from prior consensus rules.

All the blocks after the divergence point were made by someone. The original transaction history is part of both.

And I would say that using consensus to change the consensus rules is part of the consensus rules.

> It's not tautological to say that you get 100% or 0%.

I was referring to your argument that "popularity contests don't think much of unpopular (i.e. minority) forks".

> The thing called 'ethereum' at the time has always been the winner.

The thing called 'Ethereum' at the time is now called ETC, and it lost.

> every single update is a fork, and every single time two blocks are mined at the same time it's a fork.

Only updates which change consensus are considered forks. Two blocks mined at the same time is not a fork (because consensus rules have not been changed), but a temporary chain split. Forks can sometimes cause permanent splits though (ETH, Bitcoin alts) when both sides retain hashrate to mine/support it.

> I was referring to your argument that "popularity contests don't think much of unpopular (i.e. minority) forks".

So am I, but you're interpreting my words wrong.

I wasn't saying that "in the realm of blockchain forking, the less popular one will be less popular", which would be a tautology. I was saying that "in the realm of blockchain forking, the less popular one will get demolished".

> The thing called 'Ethereum' at the time is now called ETC, and it lost.

Right before the fork, they were both called Ethereum.

Right after the fork, ETH was called Ethereum. And it won.

"at the time" and "the day after" are supposed to be roughly synonyms in my comment there.

And you didn't address the issue of how validation logic has changed many other times on coins, with the original continuing to mine for a while before sputtering out and dying. Is "Bitcoin" dead? Nobody is running the original logic anymore.

> I was saying that "in the realm of blockchain forking, the less popular one will get demolished".

Sorry I still find myself reading your words the same way. If you can only determine if a fork was "demolished" by rating its popularity (i.e. market value or hashrate), you are back to the tautology:

"The most popular fork wins" seems to be your argument. But "winning" here means having higher market cap/more hashrate. And being more popular also means having higher market cap/more hashrate.

Let's say a fresh fork happens causing a chain split. Both chains have 50% market cap and hashrate. How do you measure "popularity" at this point so that your prediction of "the most popular fork wins" is actually helpful? ETH forking away from ETC was a very controversial topic at the time.

> And you didn't address the issue of how validation logic has changed many other times on coins, with the original continuing to mine for a while before sputtering out and dying.

This is exactly what we are discussing with ETC. ETH changed consensus and the original (ETC) continues to mine for a while before sputtering out (it is not dead... yet).

> Is "Bitcoin" dead? Nobody is running the original logic anymore.

Not sure what this is supposed to support? There are still Bitcoin Qt 0.8.x nodes on the network which are ~10 years old.

> Let's say a fresh fork happens causing a chain split. Both chains have 50% market cap and hashrate. How do you measure "popularity" at this point so that your prediction of "the most popular fork wins" is actually helpful? ETH forking away from ETC was a very controversial topic at the time.

You have to go forward into the future to evaluate the claim. The claim is that if you look at things... let's say 5 years later, one of those is going to be dead. If you find that one of the forks still has, for example, 20% as much hash power as the other, you have disproved the claim. So it's not a tautology.

Another way to formulate that part of the claim would be to say something like "the half-life of a fork is only X amount of time, so it's a very bad idea to try to pull one off".

> There are still Bitcoin Qt 0.8.x nodes on the network which are ~10 years old.

There's no way they're fully compatible. At the very least, consider that block that generated 184 billion bitcoins and had to immediately get the code changed. And look at all of these: https://blog.bitmex.com/bitcoins-consensus-forks/

> The claim is that [...] one of those [forks] is going to be dead.

This is much weaker claim than your original one (~"the market values popularity over technical features") and a much less controversial one.

> something like "the half-life of a fork is only X amount of time, so it's a very bad idea to try to pull one off".

> If you find that one of the forks still has, for example, 20% as much hash power as the other, you have disproved the claim.

Again, ETH was forked off the original Ethereum chain and won (it has ~98% hashrate) so it is not necessarily a bad idea to attempt one.

> There's no way [~10 year old Bitcoin 0.8.x nodes] are fully compatible.

You can download v0.8.1 here (https://github.com/bitcoin/bitcoin/releases/tag/v0.8.1) and it should be able to fully sync the Bitcoin blockchain (i.e. including blocks mined today). If you use older clients for mining you might generate blocks that are ignored by other miners but your old client will still keep in sync and follow the same correct chain. (Something that is not true for ETH or Bitcoin forks).

According to your link you can apparently sync even older clients (e.g. v0.3.24) if you take care of non-consensus related bug fixes that would otherwise crash your client.

> This is much weaker claim than your original one (~"the market values popularity over technical features") and a much less controversial one.

You keep putting different versions of words in my mouth, but I would endorse that formulation of the "original". Popularity will crush technical features in a pretty short timeframe. The technical features have an opportunity to pull people in, but it is very limited and then that version will be pushed toward 100% or 0%. It is not sustainable to split the market, unlike most markets. You can compare different markets this way, seeing if it's true or false for that market, so it's definitely not a truism.

> Again, ETH was forked off the original Ethereum chain and won (it has ~98% hashrate) so it is not necessarily a bad idea to attempt one.

It means it's a bad idea to try to split things, because even in that case splitting didn't work.

So you'd better be damn sure that a powerful majority will want your version. You can't rely on merely many people wanting it.

Also, you don't think that having it be named Ethereum is a bit of a special case?

If you fork bitcoin or ethereum, you won't be taking the name with you.

Forks are arbitrary. There's no need to literally gather 51% consensus, it just means users of Fork A are in an entirely different universe than Fork B and that's fine.

For example when BTC split into BTC and BCH, for me the ticker for Bitcoin went from BTC to BCH because BCH was the continuation of "real bitcoin". And the absurdly-small-blocks crowd obviously felt the opposite. And that's completely fine.

One of the most pernicious beliefs of the BTC crowd is that a hard fork is some evil, unforgivable thing. It shows a total misunderstanding of how free software works.

> One of the most pernicious beliefs of the BTC crowd is that a hard fork is some evil, unforgivable thing. It shows a total misunderstanding of how free software works.

It's not about software, it's about network effects when the only thing giving a coin value is those network effects. Nobody cares if you fork the software. Diverse software is encouraged!

When the hash rates are at 1 per month, I don't think we will have enough hashing power, probabilistically, to match 10 minute marks. Verification of the hash to validate and consent to the block will also be at 1 per month, for consideration.