Going through all the recent posts, it feels as if everyone is almost on the same page even though some external media would have us believe of endless debates over irreconcilable differences between the devs. It's natural that we feel uneasy about OP_RETURN and its possible consequences if one of our highest priority is to protect the blockchain from spam. If anything, it makes me appreciate the core devs a little more for handling these challenges.
Now think back a few years...the Bitcoin design paper may not even be completed -- let alone a working client -- had Satoshi insisted on including other features such as markets, assets, contracts, and all the possible financial transactions.
Instead Satoshi opted for a simple version of electronic cash and a simple script as a way to add more features in the future. Today, script gives us a simple OP_RETURN that could bring all those features and more. And yet we look for a more complicated solution.
Instead, think of jgarzik's updated OP_RETURN as the simple extension to allow features that would be too limited via script. There is no need to enable dangerous OPs for transactions that can run at the upper layer. Yes, allowing arbitrary data seems dangerous because there will be people who will try to stuff junks into the txns. On the other hand, it opens up a world of possibilities for non-core developers to create unimaginable transactions. Trying to deter OP_RETURN is like trying to hold back the first message boards because of spams.
If it's possible to have an OP_RETURN with variable payloads and appropriate fees in place as advocated by Peter Todd and PhantomPhreak, Bitcoin and its upper layers would soon leave most of the new competitors in the dust. The significant of a simple OP_RETURN can multiply many times beyond current implementations such as XCP and MSC. Yes, we could be building secondary chains, efficiently merged solutions, or even DHTs of DHTs. One day. For today, Bitcoin just needs a simple OP_RETURN.
By the way, thanks both to my mysterious benefactor and to whoever thinks I'm Satoshi. I'm not sure my previous comment merits either of those compliments, but I appreciate them. My SX stealth address is SghSgpVgk7yAFx91DxwfE84Dm6r2hwSpMWcFJbR9wQTAv7pgq61Nx3dNsn if anyone is interested - and if they are, I'm sure they can send me a nonce without too much trouble One would expect
satoshi to be monitoring the XCP/MSC threads closely and may even provide feedbacks as a detached observer.
topynate provided some very insightful posts (in an almost academic-like style), showed us the futility of trying to classify OP_RETURN as carrying non-finanical data, and even quoted satoshi to remind us of the comparable "unstructured simplicity" in an OP_RETURN. That embodies the ideals of satoshi for many. Of course, satoshi would need a stealth address since he could not move those unsteathed coins without being noticed now