Prior related invention in 2012 by my co-founder: https://bitcointalk.org/index.php?topic=91843.0
Limitations of this side-chains:
- Can't use proof-of-stake
- It requires many changes in the Bitcoin reference client
- It's not very clear if you can increase the frequency of blocks in the side-chain
Can you please explain why PoS is incompatible with sidechains?
IANA blockchain expert, but this is my understanding:
You need to be able to show a "proof" to the bitcoin main network that the coins are being transferred back to the bitcoin blockchain. This proof needs to be hard to generate, which pretty much makes POW mandatory.
Do you know if that's 1) "must use POW for the transaction that burns the sidecoins", 2) "must have proof of enough total work in the sidechain atop the block in question", or 3) "must only have POW blocks in the sidechain"? Superficially, 2 seems most likely.
If you take this a bit broader, you could not just link other crypto currencies to Bitcoin, but can use it to timestamp any types of data. For example, in mid-2013 I created a "public ledger" service based on Bitcoin: https://vog.github.io/bitcoinproof/ (HN discussion: https://news.ycombinator.com/item?id=5796935).
And even that wasn't really new. It dates back to forum entries and similar projects from 2011 and possibly earlier.
So while I like the basic idea of Sidechains, I don't like their presentation in the article as something totally new, without giving any credit to the history they build upon.