There seems to be some confusion about floating rates, sidechains are algorithmically pegged not floating.

Lets try a thought experiment.  Say you can directly move a bitcoin to a sidechain or move it back to the mainchain with either direction taking 10minutes and normal bitcoin fees.

Clearly given that this is the best case confirmation time for bitcoins, and the peg protocol is algorithmic there will be effectively ZERO spread, because the algorithmic peg is an unlimited standing offer at parity (plus per KB fees) and is in direct competition to any market offer, and rational actors take the lowest offer.

Now we introduce the concept of time-preference.  For security reasons (rather similar to coinbase maturity which sees you unable to spend freshly mined coins for 100 blocks) the algorithmic peg has a time-delay.

Now if you planned to hold anyway for that period or longer, then you dont care and the situation is unchanged.

But if you want to do a sidechain BTC transaction faster, you swap it for a small premium with someone who already has BTC on the sidechain and is planning to long term hold, or swap with someone trying to go the other direction.  What you pay them will be small due to the mechanics of arbitrage.  They'll just look for some small fee because to them if they're already long term BTC holders its basically free money, like interest on BTC to move funds back and forth and provide liquidity service for sidechains.  The $ exchange rate is immaterial, the best candidate for sidechain liquidity provider is someone who is anyway holding their own or other peoples BTC for long term storage.

Anyone who tried to sell  BTC on one chain for a lower than time-preference cost on another chain would just lose money.

I think the above logic and economic concept & precedent is extremely simple.  It seems like some people misunderstood Konrad Graf's comments, he's just talking about the mechanics of the low arbitrage spread.

One can look to other bitcoin arbitrage scenarios for a hint at how it works.  Look at the spread between btc-e & bitstamp now that multiple people are systematically arbitraging it.  That is a far riskier arbitrage because you are relying on governance and security management of bitstamp & btc-e in the face of 50% failure rate of bitcoin exchanges.  Ok these ones are survivors and better than full history average no doubt but still there is non zero risk there and yet the spread is basically 0, this is because of competition amongst arbitrators.  Compare to a 2wp, where there is an algorithmic arbitrage.  A bot can take that all-day long at zero risk (using smart-contracts).

The remaining non-imaginary risk is side-chain implementation defect, but the point of side-chains is to allow experimentation on new features to occur outside of bitcoin core.  This is actually a good thing for bitcoin core's risk because it doesnt have to take as much new development risk itself.  Sidechain development will also be rigorous like bitcoin, and you should look at the reputation of the authors of the sidechain you are considering using and have others review it or certify it before you dump your lifesavings into it.  You can still do long term holding on bitcoin if you prefer, and benefit from the even lower than current mainchain risk.

The current pattern in bitcoin infrastructure is most transactions are offchain (in exchanges and other offchain accounting).  Much of that code is not open, or inexpertly written or relying on firewalls and host security and hot wallet ratios plus you're vulnerable to governance failures, operator theft and or blackmail.  Most bitcoin lost to date has been for these reasons.  If, using sidechains, we get more innovation and more onchain transactions, its better to be onchain in a sidechain than offchain from bitcoin.  You dont even own bitcoins offchain, you have an IOU for a bitcoin from a human who typically has no banking governance nor operational security experience, though with better capitalization and management things are improving.

I would think more transactions, more transaction types and more uses for bitcoin, and faster innovation on bitcoin and more onchain transactions and zerotrust/smart-contract based infrastructure are all positive things for bitcoin, and it seems to be that most people with a technical understanding hold the same view.

Its not that a sidechain displaces bitcoin hypothetically and that this is bad; a sidechain is bitcoin, its the mechanism to internetwork bitcoin, to build on it.  Sidechains no more displace bitcoin than HTTP displaces the TCP and IP protocol it is transported on.  Alt-coins and alt-shares are in nominal competition with bitcoin, though it seems highly unlikely they would catchup with bitcoin's network effect nor reach an appreciable real-life usage; but sidechains are not in competition.

Bitcoin has one advantage over sidechains - it has the subsidy, and full node security, so I'm sure it'll be able to defend itself against abandonment or insecurity, and sidechains depend on bitcoin anyway so all bitcoin users on which ever chains have a meta-incentive to see bitcoin main remain secure.   We have decades of subsidy ahead to deal with fee-only security for bitcoin, and sidechains may move forward the ways to do that because the sidechain by default has only fee security from the start.

Anyway one potential use for a sidechain is to host a beta for a major new bitcoin version.  If that version after say $1b value running in it for a year, gets upgraded into bitcoin, that hasnt displaced bitcoin, its facilitated its feature and performance upgrade, which is a good thing for everyone.

The exciting thing about the internet wasnt just the ability to send IP packets, but all the applications and permissionless innovation that could be built using that transport.  Same for bitcoin.

Adam