I think the
point about trustelssness of smart contracts stands out.He is not wrong when he says that there is a lot of trust being put into smart contracts to function at various stages like:
1.) The developers are trustworthy and not hide/ ignore bugs in the software.
2.) The publicly auditable software is actually being audited by enough people to be considered reliable.
3.) Even the most robust software will continue to have points of break.
Bitcoin code seems to have been above such failings till now. Though we cannot know a scenario where a fast enough computer would be available to alter the history. You cannot rely on the fact that some secret government somewhere will not fund such a project. All these are plausible scenarios one should consider, if we talk about making the world rely on it.
The solution to this, in my opinion, isn't to shrug shoulders and say, well, blockchain doesn't work because it's too complicated for normal people to understand.
The solution, in fact, lies in making the software more accessible and easier to read.Programming languages have evolved from
Binary bits to mnemonics to syntax in English language. Could we, possibly, evolve it into a simple conversational Q&A which will be easy for everyone to understand?
The underlying compiler can be the robust point of trust which could establish itself as a sort of oracle that can do no wrong and will always give you the right answer. Once you have such a compiler in place, trusted through history of usage and public audit (like bitcoin), you could implement easily understandable smart-contracts on top of it.
The other statement that clearly shows lack of understanding and imagination on part of the author is:
People who actually care about food safety do not adopt blockchain because trusted is better than trustless. Blockchain’s technology mess exposes its metaphor mess — a software engineer pointing out that storing the data a sequence of small hashed files won’t get the mango-pickers to accurately report whether they sprayed pesticides is also pointing out why peer-to-peer interaction with no regulations, norms, middlemen, or trusted parties is actually a bad way to empower people.
The examples are completely unrelated. There is no metaphor-ness in relying on blockchain records to trust whether something actually happened on a product's lifecycle. You are not trusting the farmer's manual record on spraying pesticide. That is what
Internet of Things is for. You
rely on sensors to upload data onto the blockchain without any human intervention.
If implemented, such a scenario can easily work. Very less human oversight is required to confirm periodically that the systems are indeed being followed to the point.
IOT and Blockchain together can really remove trust/ middle parties from logistics and manufacturing records. It's not like it cannot happen if people like the author find it hard to wrap their heads around it.