A decentralised peer-to-peer system sounds exciting, but does it really work that way when it comes to blockchain in general and cryptocurrencies in particular?

“The beginning of wisdom is the definition of terms.” -- Socrates

In (the largest) CPU we trust
Let’s re-visit the genesis of “purely peer-to-peer” blockchain – Satoshi’s 2009 whitepaper. The foundation of Satoshi’s blockchain – cryptography apart as “digital signatures provide part of the solution” – is… the largest pool of CPU power that acts as the source of truth. Validation of the BTC blockchain is done on the basis of its length, and the longest chain comes from the largest CPU power pool.

Satoshi’s does warn that the proposed system only works “as long as a majority of CPU power is controlled by nodes that are not cooperating to attack the network”. The nodes in question are anonymous. Moreover, their number can fluctuate - “nodes can leave and rejoin the network at will, accepting the longest proof-of-work chain as proof of what happened while they were gone.” 

As for the reliability of that system, it’s SLA terms are simple too – “messages are broadcast on a best effort basis”. Ditto.

Atomic swaps
There are existing proven and sort-of reliable cryptographic methods for true P2P digital exchange of assets. For example, such hardware wallets as Trezor and KeepKey offer “atomic swap” function. Nevertheless, they still required a blockchain – which, as we established, is relying on (presumably decentralised) third parties… Why do we keep coming back to the need for some third party? It’s all about double-spending. 

Double-spending
If the system is P2P and transactions are “sent directly from one party to another”, why do we need “nodes” at all? To prevent double-spending. If the assets were physical and exchanged face-to-face, the transaction would indeed be P2P – I give you a dollar, you give me two apples. You can see my money, I can see your apples. One party can still fool the other, and that could lead to a Mexican standoff – but that’s another story.

Things are different in the digital world. Even if we use “Digital Bearer Instruments” (to borrow an interesting term from Dave Birch), there is no guarantee that DBI owner didn’t create a copy or two. To have tangible and tradeable value, digital assets need to come from a limited pool of supply – if everyone could easily create a copy of the latest Hollywood blockbuster and share it at will on Moviegram (hm, cool idea and name… just kidding), the film industry would have long been dead. 

To wrap it up
Digital assets can be traded in a true P2P way, a digital analogy of physical “money-goods” exchange. Atomic swap is one such mechanism. However, the nature of digital assets – cryptographic or not – does not guarantee asset’s “authenticity”, i.e. value. Assets duplication leads to double-spending. The way to prevent it is to rely on some third party validation, decentralised or not.

Were it not for Satoshi’s "as long as" warning and the recently uncovered Bitcoin bugs, one could argue that blockchain gives the world more freedom and certainty, by removing restrictions imposed by the governments and regulators. The reality is different for now, though.