I see, the pesky block format is the issue in my opinion. Because it's standardized. If the transactions were each put arbitrarly in blocks, it would be enough to just verify their hashes to see that they are original,and then there would be no need for the block format.
It would be a very big deviation from the protocol , but it would be interesting to see an altcoin try this.
Well, we can't really state that Bitcoin is very well designed. I'm certain that the developers who are working on it currently, would do a lot of things differently if they could. The only 'good' thing about altcoins is that they can be used to test out some 'difficult' ideas. I don't think that we are going to see such radical re-design (what are the limitations of a HF?), so there's not much 'point' in discussing it in, at least not in this thread. I do wonder if they can get RC1 released before the halving (even though Segwit still has no activation parameters). However, rushing is usually not the right thing to do.
AFAIK the only thing implemented so far is segwit... which only (temporarily) avoids a hardfork, and still increases bandwidth, which is the supposed boogieman of "government control".
Compact blocks aim to improve bandwidth requirements, i.e. lower them. Additionally, have we already forgotten all the excellent upgrades that they've delivered (e.g. libsecp256k1)?