> if something was so unequivocally better then wouldn't this perhaps resulted in a massive adoption rate
That's not what we see happen in life. Just look at how dearly we hold on to coal plants. My country's energy is 90+% coal-based and there is an insurmountable opposition to atom, and great suspicion against solar (government recently started financially deterring against mounting solar panels).
> That's not what we see happen in life. Just look at how dearly we hold on to coal plants.
What a non-sequitur. Adopting a programming paradigm in the code that we write on a daily basis has rigorously zero to do with the infrastructure cost of switching energy sources.
This non sequitur about coal plants is even more absurd and ridiculous once we factor out the fact that the world is already moving away from coal to renewable energy sources, which is quite costly and resistant to change and subjected to an awful lot of special interest groups, and yet during that timeframe people still chose not to bother with functional programming fundamentalisms.
If an idea is good then it holds its own against alternatives. Even in their pet projects people tend to not even bother with pure FP frameworks. While hundreds of millions of euros are being spent on wind farms, people like you and me don't even bother spending a few minutes to get a hello world going with Haskell, even though it's effortless and trivial. Why'd you think that's happenin?
Yeah, over a long time it will certainly happen that a good idea will manifest itself. But that process can take a long time, centuries even.
Let's not forget that there are commercial interests for keeping certain languages down and pushing languages up. There are huge companies like Google, Facebook, ... pushing for languages (and frameworks).
It's often easier to stick to something existing because change requires effort. And I think that's why coal was mentioned - it's easier to stick to it than to switch to something else, so it will take time but eventually it will happen.
He really isn't, and the blatant attempt to poison the well is a clear indicator of the lacking arguments.
I repeat: people don't even bother spending a few minutes trying to get a pure FP hello world going. Zero barriers to entry, zero challenge, zero resistance.
But in spite of the lack of any obstacle or challenge, pure FP fails to present a value proposition that justifies even a five minute effort from pretty much anyone.
> Yeah, over a long time it will certainly happen that a good idea will manifest itself. But that process can take a long time, centuries even.
This baseless assertion holds no water in software development. It does not take a multi-generation epifany to convince someone to use pure FP. All it takes is a single person willing to invest it's time.
But still, those who do invest their time with ivory tower tests don't see value that justifies pursuing any effort, and thus don't bother pushing it anywhere beyond a cursory test.
Why is that?
> It's often easier to stick to something existing because change requires effort.
This assertion holds no water at all, as pure FP frameworks exist for decades and still people do not bother with them. Why is that? And how can you hold such a cognitive dissonance of claiming something is so great and yet it does not exist in practice because no one ever bothered to take advantage of such greatness?
Also, are we supposed to pretend that even Microsoft offers pure FP languages that integrate with their .NET stack and still no one bothers with it at all, even when they can even contam it's use to very small and self contained modules?
The truth of the matter is that pure FP fails to gather at attention beyond navel-gazing ivory tower types because it quite blatantly fails to present a value proposition. That's the core of the issue. I mean, languages like Rust are increasing popularity like wildfire in spite of it's radically different and restrictive take on resource management because it presents a clear and inequivocal value proposition. But pure FP frameworks, in spite of having a head start of decades and a foothold on dark corners of academia, fails to convince the public that it's in their best interests to even peek in that general direction. Why is that? Do we really need to resort to absurd conspiracy theories to get an answer to that question?
First, FP required/requires more resources on average. With time progressing and better hardware coming, this point becomes less important.
The strengths of FP, such as easier concurrency, are becoming more relevant as well.
However, even universities take time to "catch" up. And developers often stick to the style they first learned (or a similar style at least). In addition, they already created and invested into an ecosystem.
But look at all the new programing languages that are getting created. They include more and more FP features, starting with the more easy ones that have a direct impact on productivity (such as lambdas) but it's getting more and more. Even languages such has Java start to slowly move into this direction.
Or do you want to deny that?
Also, I have no idea what you mean by "FP frameworks".
That's not what we see happen in life. Just look at how dearly we hold on to coal plants. My country's energy is 90+% coal-based and there is an insurmountable opposition to atom, and great suspicion against solar (government recently started financially deterring against mounting solar panels).