Conclusions

Digital assets detractors usually say that there is no proven demand for cryptocurrencies, but it has been demonstrated that demand not only can be measured but that crypto-economies and their driving variables can be ranked as demand evolves \cite{Venegas}. Perhaps the exercise of comparing Bitcoin and BitcoinCash is not entirely fair (after all BTC had the first mover advantage, by several years), but the heuristics that we have learned from the data have relevant implications nonetheless. For instance, one could identify what are the sources of systemic importance, or what traffic is overpriced or underpriced. And since in blockchains transaction count and exchange volume can be manipulated by batching transactions and other artifacts, one of the viable measures of value might be actual supply and demand of attention.
Furthermore, if crypto assets defy the “Efficient Market Hypothesis” and the idea that all available information is encoded in prices, something more profound may be going on here: beyond any of the traditional definitions of utility, disintermediation of trust by itself might entail a premium. In that case, the value of the chain may reside on the chain itself: the nodes running the software are simply an expression of people’s beliefs — being that the belief that the market can be manipulated for personal gain; that it is about time to challenge the government monopoly on money; that algorithmic money might be the more convenient utilitarian artifact to conduct transactions if you have already digitized a large part of your day-to-day activities; or else. This belief consensus is a human-machine construct, and perhaps this is why economists who are not trained as technologists have a hard time grasping the implications of a blockchain financial system.
But what is more intriguing is that what the quantitative analysis reveals is not conflicting at all with the definition of intrinsic value — value is, after all, a matter of perception. So the argument that cryptocurrencies have no intrinsic value is without merit, and as we have demonstrated, not backed by data. Furthermore, even regulators stances are evolving; according to FinCEN, a digital currency can represent a “value” that “substitutes for currency" \cite{watch} -- this value representation is what is encoded in the off-chain network flows that we have quantified as trust metrics builders. And a more fundamental question about value arises: as the trust asymmetries between crypto economies reveal a structural divergence in value perception, could this paradigm provide incontestable proof of value in digital assets, including those with enhanced privacy features which by default make key transactional data opaque or unavailable?    
Immediate applications of this research include Discreet Log Contracts \cite{dryja}, which have the potential to enhance the use cases of Bitcoin and other cryptographic currency networks by allowing users to discreetly enter into futures contracts for a wide variety of assets, trusting oracles only to sign the correct price. Possible next steps include the formalization of evaluation frameworks for the trust metrics and trust models. For instance, the share of flows is in principle a probability, therefore it could also be analyzed using formalisms from logic. Subjective logic \cite{J_sang_2016a} is a type of probabilistic logic that explicitly takes uncertainty and source trust into account, and could be used for this purpose. Also, topology concepts such as persistent homology could be implemented to study the robustness of the trust metrics obtained \cite{Buchet_2016}. Finally, one may argue that in essence, trust asymmetries are a particular case of information asymmetries. In this view, we could use the rich literature of information theory, signal processing, complex networks, and, econophysics to develop on the methods here described.