I'm not interested in either of the "Satoshi willed it", and "but I want it sooooo bad" non-arguments.
So you're only interested in what code Satoshi put in. You've arbitrarily decided that the plan he communicated in writing is irrelevant, but not the code he wrote 5 years ago.
In the context of this debate, what Satoshi communicated in writing as to his intention for Bitcoin's future block size, is more important than what he coded 5 years ago as a temporary anti-spam measure.
I believe they are not as secure or distributed as Bitcoin main.
Please, explain to me how merged mining is insecure. Take your time.
Bitcoin miners can attack sidechains at very low cost to themselves, as they can earn BTC mining Bitcoin main chain WHILE doing the attack on the sidechain. Furthermore, a successful >50% attack on a sidechain is much more destructive than one on Bitcoin main, allowing the attacker to steal all of the BTC backing the sidechain. A sidechain is also likely to not have 100% of the network hashrate of Bitcoin merge-mining it, since it requires extra effort for a Bitcoin miner to merge-mine another chain, so it will be easier to >50% attack.
For all of these reasons, sidechains are not as secure. I recommend you read the sidechain whitepaper published by Blockstream to see the risks that sidechains face.
My argument is perfectly fine, and you actually add to it.
For how could we come up with a sensible solution if we haven't actually experienced the issue?
Your argument is terrible. A large subset of the population can continue to host full nodes without upgrading their consumer broadband, with the block size limit that Gavin has proposed (16.8 MB with 40% growth per year). Given this fact, there's no reason to gamble Bitcoin's entire future on an experiment to see what happens when blocks fill up at a low 1 MB limit for an extended period of time. I don't want to give any altcoins an opportunity to take market share from Bitcoin.