Sidechain whitepaper

[jonald_fyookball]

If someone has deeply read the sidechain whitepaper and would like
to summarize some of the main points or explain in basic concepts
how it works, that'd be super.

[brg444]

May I suggest this article:

http://gendal.me/2014/10/26/a-simple-explanation-of-bitcoin-sidechains/

[maokoto]

May I suggest this article:

http://gendal.me/2014/10/26/a-simple-explanation-of-bitcoin-sidechains/

Nice article, although it does not look as "simple". I got mentally tired just by looking at the pics LOL.

[Peter R]

If someone has deeply read the sidechain whitepaper and would like
to summarize some of the main points or explain in basic concepts
how it works, that'd be super.

The paper proposes two-way pegged sidechains as an extension mechanism for Bitcoin.  The idea is that coins would be able to move from the Blockchain, to a sidechain, and then back again in a trustless way.  This would allow sidechains to implement properties that are not feasible to implement on Bitcoin itself, while preserving the total number of Bitcoins.  

The feasibility (excluding snarks) of trustless sidechains hinges on the existence of a new cryptographic primitive called "SIDECHAIN_PROOF_VERIFY."  This is the "magic" that would allow the sidechain to prove to the Bitcoin network that coins on the main chain should be unlocked.  

The authors briefly describe SIDECHAIN_PROOF_VERIFY in 270 words on pages 8 and 9 of the white paper.  They mention that the challenge is making these proofs "compact."  For example, if the transaction from the sidechain to the Blockchain (the reverse TX) included the entire sidechain blockchain, then obviously the Bitcoin nodes could figure out whether the transfer was valid or not.  However, this would lead to unwieldy reverse transactions that could become arbitrarily large.  

To address this, the authors include Appendix B, which suggests that it might be possible to make these "sidechain proofs" compact [I haven't personally verified the math so I cannot endorse that claim].

As far as I know, there is no working implementation for SIDECHAIN_PROOF_VERIFY with compact proofs.  

[jonald_fyookball]

The sidechain is a separate blockchain right?

What group of miners would secure it?

[Peter R]

The sidechain is a separate blockchain right?

Yes.  And there can be more than one sidechain.

What group of miners would secure it?

This is undefined.  Sidechains could be merged mined, or even secured by proof-of-stake. 

[Blawpaw]

May I suggest this article:

http://gendal.me/2014/10/26/a-simple-explanation-of-bitcoin-sidechains/

Sidechains are one of the best proposals that can effectively solve all the Bitcoin protocol limitations. This article is a very good read since it explains all the good points Sidechains present.

[jonald_fyookball]

The sidechain is a separate blockchain right?

Yes.  And there can be more than one sidechain.

What group of miners would secure it?

This is undefined.  Sidechains could be merged mined, or even secured by proof-of-stake. 

is there any way they can be used to scale Bitcoin while using decentralized nodes backed by the security of the main chain?

[brg444]

The sidechain is a separate blockchain right?

Yes.  And there can be more than one sidechain.

What group of miners would secure it?

This is undefined.  Sidechains could be merged mined, or even secured by proof-of-stake. 

is there any way they can be used to scale Bitcoin while using decentralized nodes backed by the security of the main chain?

Define scale?

Scale transactions? Not adequate.

Scale properties and use cases? Definitely

[LiteCoinGuy]

https://www.youtube.com/watch?v=jE_elgnIw3M

[spartacusrex]

is there any way they can be used to scale Bitcoin while using decentralized nodes backed by the security of the main chain?

I don't think the sidechains can use the security of the main chain unless they are merge mined..

If you want to run a POW side chain, not merged, you would need to start from ground ZERO.

I think POS sidechains, with POW checkpoints https://bitcointalk.org/index.php?topic=1170713.0 might be one way to go.

[coalitionfor8mb]

My concern with sidechains isn't about their mathematically sound implementation, but rather with their effect on the incentives structure that has proven to work so well for the original Bitcoin.

If sidechains gain enough traction, they will form new clusters of gravity, which will begin tearing the idea of a single ledger apart fairly quickly with no one being interested in securing the main chain, which in turn would only accelerate the drainage of its value.

If that manifests, the niche that Bitcoin was created to fill will simply be left up for grabs and the competition will surely take the chance.

[brg444]

Sidechains are like altcoins that you can trade bitcoin for without using an exchange, think of shapeshift but if it was decentralized and trustless (no third party), the chains are still pretty much altcoins and will operate so with their own rules, algorithms, miners and what not. The idea of fixing the blocksize with sidechains is ridiculous to me as it just introducing even more altcoins and making it even harder for merchants to accept.

It seems to me you are beyond confused.

Yes sidechains can be used to bootstrap an altcoin but there is little interest in doing that. The point of sidechains is that their units' scarcity can be derived from Bitcoin, essentially pegging their value to Bitcoin. That would allow to extend the properties beyond what is possible on the mainchain while leveraging Bitcoin's liquidity and network effect.

No one is proposing to fix the blocksize issue with sidechains.

[johnyj]

I think it brings more problems than it can solve, due to its complexity. Raised level of complexity will reduce system's robustness and integrity

Suppose that 1000 bitcoins were send to side chain US, and then 1000 USbtc will born. We are talking about a future that each bitcoin worth one million dollars, so USbtc blockchain will have 1 billion dollars worth of coins, enough for a small ecosystem

Then there are several issues:

1. The chain US must be enough secure: You either use other hashing solution to secure the network (and dilute the hash power for bitcoin network at mean time) or merge mine together with bitcoin (risk of double spending attack by other bitcoin hash power)

2. In order to use those USbtc, merchant have to install another client to specifically work with it. There will be other side chains like EUbtc, AUbtc, CNbtc etc...  And since EUbtc is not spendable on USbtc network, vice versa, each merchant/exchange will install and maintain a whole bunch of different clients if they want to do business with its international customers. So unless all the side chains are invisible to end users through a single unique interface, the most likely outcome is that each country use its own side chain due to speed advantage in its domestic network

3. If CNbtc grows a lot and there is a strong demand for CNbtc in china, this will cause its exchange rate to be higher than bitcoins on the main chain. Then there is a motivation to convert main chain bitcoins to CNbtc due to 1:1 peg. There will be arbitraging happening every second on the blockchain to profit from the different exchange rates between different chains, thus the whole bitcoin main chain becomes an exchange platform, dramatically increase its traffic

4. The mainchain is not aware nor responsible for what is happening on the side chain, so if something went wrong on side chain, it might cause the mainchain to fork due to nodes might disagree with the different truth they receive from the same side chain...

In one word, raised level of complexity will bring much more problems that you can never expect

[jonald_fyookball]

could some side chain serve as payment network that could be settled end of day by being included by miners on the main chain as some kind of super block?

[adamstgBit]

could some side chain serve as payment network that could be settled end of day by being included by miners on the main chain as some kind of super block?

you should be able to transfer in and out of a side chain at any time at a fixed ratio. i guess what you're suggesting is that once a day everyone on that chain would be automatically dumped back into BTC?

[jonald_fyookball]

could some side chain serve as payment network that could be settled end of day by being included by miners on the main chain as some kind of super block?

you should be able to transfer in and out of a side chain at any time at a fixed ratio. i guess what you're suggesting is that once a day everyone on that chain would be automatically dumped back into BTC?

precisely.

The sidechain would have less security than the mainchain, because it would be different set of miners/nodes, but the risk would be contained.

[erik777]

On the contrary.  The sidechains would

My concern with sidechains isn't about their mathematically sound implementation, but rather with their effect on the incentives structure that has proven to work so well for the original Bitcoin.

If sidechains gain enough traction, they will form new clusters of gravity, which will begin tearing the idea of a single ledger apart fairly quickly with no one being interested in securing the main chain, which in turn would only accelerate the drainage of its value.

If that manifests, the niche that Bitcoin was created to fill will simply be left up for grabs and the competition will surely take the chance.

I was about to argue against this, but now I think I see how this could happen.  To simplify, CoinX is a sidechain that gains a lot of momentum, so much so, that 90% of bitcoins become "locked" into it. At that point, CoinX has basically replaced Bitcoin.  

While some might view this as evolutionary, what if CoinX sacrificied certain principles, like being semi-anonymous and fungible.  Why would the masses flock to it if it made so many sacrifices?  The same reason most people pay with a credit/debit card today.

That said, if someone can come up with the holy grail of anonymizing transactions, they could create a very beneficial side-chain.  I've tried to think this through, yet found it very challenging because of the difficulty of achieving trustless at the network level.

Here would be an interesting use-case for a side-chain, though.  The first eCommerce driver on the Internet was online gambling (next to maybe p0rn, which is basically free today).  Not surprisingly, there is plenty of bitcoin based online gambling today.  But, can you trust the casino not to cheat?

What if a gambling side-chain published all its bets on a side chain?  You could not only verify that your bets were published, but do statistical analysis of ALL bets to verify the honesty of the casino.  

Contrast that to moving your bitcoins to a casino, where you trust them with your bitcoins AND trust them to be honest, without any transparency or protection of your bitcoins.

To be sure, you could accomplish the bets chain today without changing bitcoin's protocol. The difference might be also protecting your bitcoins in the process.  

  

[coalitionfor8mb]

Bitcoin as a single currency (and single chain) taps into the balance of power that exists on the Planet.
That's what makes it unique, as it sits in a fusion point where "balance of power" becomes "power of balance".

Now, if we Zoom a little bit and look at it from inside of our minds, we will see this.

Western demographic (mostly US and Europe) got better connectivity and higher bandwidth and expressed that 8MB should become a new floor (BIP101).
Eastern demographic (mostly China) got higher hashing capacity but poorer connectivity and lower bandwidth and expressed that 8MB would be a new ceiling.

It doesn't take too much wits to realize what level of bandwidth we should be targeting for the next round in order to keep the system in check.