Thanks everyone for your comments - I had no idea this had generated such interest. I've forgotten to update my bio, but these days I'm in fact a PhD. Studying

Thanks everyone for your comments - I had no idea this had generated such interest. I've forgotten to update my bio, but these days I'm in fact a PhD. Studying the impacts of blockchain and other distributed trust technologies has been my day job since September 2016. I freely admit that these are just my interpretations and I may be wrong; that's the point of writing articles like this. That said, I do stand by my analysis and believe blockchain technologies are greatly overhyped relative to what actual value they do add to existing systems. The great majority of users are fairly happy or at least indifferent with the #1 bugbear of most crypto enthusiasts - trusted third parties - and it is far from clear that simply getting rid of them (and substituting hazy, informal, unaccountable governance structures instead) is ever going to be the killer app some people seem to think. However, there will also be many use cases where blockchains and e.g. DAGs can and will be used to create real value very effectively. Just today talked to a company that's doing interesting things to help car rental companies streamline their processes - that's one example of what shared ledgers and smart contracts can do to really decrease transaction costs. However, I do predict that the majority of sustainable use cases will gravitate to more or less private blockchains, sooner or later. This company had chosen that route from the beginning. One thing I would add today to the text is a caution against thinking that existing incumbents can be toppled simply by duplicating the functions of their databases. Databases are ultimately the most valuable asset of many trusted third parties, but their value proposition includes MUCH more than just having a reliable database. One thing coders generally are poor at grokking is that things like user experience do matter, and generally blockchain systems are far from being designed from user experience in mind. Another, probably even more thing entirely, is customer service and error handling when things will inevitably go wrong. Immutable ledgers are by definition poor at handling such things, yet there is considerable evidence that people tend to be very wary of adopting a system if they fear that they can make a mistake that has serious results. Existing third parties have understood this, and offer a lot of services and design features aimed at assuring customers that if something goes wrong, they will not suffer. These are interesting technologies and a lot of people are partying like it's 1999 again. But I do suspect there is going to be a through of disillusionment in our future as well.