RE: The concept of an exocortex is similar to the invention of glasses
Let's not forget that a vast sum of technological advancements come from military developments for specific use cases. I think a more frequent case is when a new piece of technology gets developed for a specific use case, to solve a specific problem, and then, when we can collectively take a step back and look at its (potential) impact, the idea of this advancement can evolve into a less specific technology absent of the cultural/contextual bias you are speaking of.
Take a look at cryptocurrencies. Bitcoin was born as a reaction to the financial crisis, it solved a specific problem, and when the dust started to settle, people realized that this technology can be applied in a wide range of other industries, solving a wide range of other problems.
Can you cite some examples of lasting technologies which are highly specialized? Bitcoin solved a 2008 problem. It's 10 years later and the problem space has evolved. Has Bitcoin evolved with it?
The CISC vs RISC vs ASIC architecture debate. CISC is much more generalized compared to ASIC, but ASIC is application specific. If Bitcoin is like an ASIC and Ethereum is like RISC, then perhaps Ethereum has an advantage. On the other hand if Ethereum is like RISC but something else is even more generalized, why would we stick with ASICs?
Same could be said for mining. If we had a universal chip which can mine everything efficiently then why would we ever need ASICs when we can simply reprogram that universal chip?
You might have misunderstood my because my point exactly was that they start out specialized, and then evolve into generalized iterations. Inventions are rarely made for "general" use cases because a specific problem is the catalyst of the invention itself. We don't use the bitcoin protocol for smart contracts, but the idea of smart contracts on the blockchain wouldn't exist without the bitcoin protocol.
Unfortunately I do not see Bitcoin moving into a general purpose direction. While I don't know what it will become, it's not in my opinion currently evolving toward that. Also I would say smart contracts were thought up long before Bitcoin. It is true that Bitcoin was the first implementation of many ideas but the main idea was Nakamoto consensus.
For 2008 Bitcoin was at the time the state of the art. Even in 2012 Bitcoin was still state of the art. I cannot say Bitcoin is state of the art in 2018.
This is exactly my point. :) Bitcoin is not state of the art anymore because the innovation it brought to a specific problem evolved into something more generic, namely smart contract platforms. However, the implementation of such smart contact platforms (which arguably won't even use Nakamoto consensus, the core innovation, upon their mainstream adoption) wouldn't be possible without the initial innovation of bitcoin.
By the way, why do you say "unfortunately"? Isn't it a good thing that new protocols are being built for new ideas? It would be sisyphean to keep mending old protocols that were designed for a different purpose.