I started reading with that article but it doesn't really seem to answer the questions it's claiming to answer. It's talking about how Bitcoin et al. have a shared consensus that's updated once every 10 minutes, and how everyone needs to validate everything which is inefficient (I agree). But while trying to explain how holochain does it different I don't read how except for insect analogies, for example:
> ## Order of Operation Matters
> It turns out when you focus on distributing process first you end up with even more greatly distributed data. This is because each participant holds only their own data. In contrast, blockchains store everybody’s transactions in a single database that every node verifies and copies. The underlying foundation of distributed process enables deeper distribution of data and parallel architectures. Ethereum and smart contracts, are doing this the other way around — layering processes on top of a global ledger.
> By distributing process at the foundation, and leveraging Intrinsic Data Integrity, our approach results in massive improvements in throughput (from parallel simultaneous independent processing), speed, latency, efficiency, and cost of hardware. There is no need to wait 10 minutes to see if your transaction gets committed. This architecture can facilitate huge volumes of even extremely “low value” transactions permitting creative uses for coordinating shared activities that wouldn’t merit the cost or energy of implementing on a blockchain. You also don’t need to incent people to hold their own record — they already want it.
This is all true but the section heading "order of operation matters" - and how holochain deals with this - is simply not addressed. Given HN is a technical community feel free to link to actual research papers or more technical content that actually explains how Holochain solves the problems. This piece looks like a PR piece to me, one that doesn't really have any answers.