I thought you were referring to an impossible "bearer coin" type of design where PoW wasn't needed when you wrote that 'if only you can mine your own transactions'. Now I understand you meant something else
Yes, something else - I was referring to a POW system with one transaction per block, minable only by the sender of the transaction. I guess Iota is an example of such a system, but it doesn't have to be the only one.
And unfortunately afaics Iota forsaking blocks introduces another flaw which is consensus is not forced, thus there can be (at least in theory and probably not an observed effect until it becomes a phenomenon if ever) divergence of DAG chains.
You forget I pointed out to smooth (in our discussion about Proof-of-Stake versus Proof-of-Work in my vaporcoin's thread last month) that assumption fails when the government + oligarchy is the one taking control, because they can charge the cost to the collective.
I guess this is the plain 51% attack? Again, I'd love to hear the key details to any system which claims to address this, which doesn't create much bigger problems than it solves.
Well I have proposed submitting PoW with each transaction, but I haven't yet detailed how I envision doing the temporal intrablock partitioning. Let me go think more on all that has been enumerated in this thread to see if I have an obvious flaw. And I will also think about when I want to discuss those remaining details of my proposed design. I expect to find a flaw in my design.
I think it is impossible to fix the CAP theorem.