I don't know. Having all scientific knowledge written down and being indexed for research seams to scale better. Also I am not sure what point the article is trying to make. It seams a bit vague.
I feel with the sentiment for the "loss of skill" due to convenience tech.
But hey, these days many people have the choice (meaning the time and money), to keep some of the skills alive. The internet gives you the possibility to find any person teaching the skill set you seek. For more common stuff even Youtube is a trove often for free.
* functional style makes it easier to split stuff since you mostly transform immutable data
In the Article IO, State (side-effects) and Data transformation is mixed in both versions. That leads to unnecessary complexity. In that case worse if it hides in sub-functions. But separate it and the right version is better. (you can answer the question about idempotency easily now)
* Comments over blocks of code are harder to keep in sync with the code below, they require more discipline from all team members
So in theory they are nice, but you will never have them unless you enforce them through proper code reviews.
I agree with the Functional style part, but hardly disagree with your second point.
Because of the fact that, keeping comments over blocks of code in sync with the code below requires the same exact amount of effort (arguably less) as keeping function names (and ideally, documentation) in sync with the code inside..
Except that, if the second one is not done, that is a lot more dangerous than the first- exactly because, every time you define a function, you are declaring an abstraction, and if the abstraction changes silently, that's where the real mess begins.
That's what ends up happening in my experience. (See also sibling other comments about the proliferation of flags in a function signature)
(Needless to say, I strongly resonated with the OP as I also love linear code with comments, but in the end it's also a matter of taste..)
Also they note that ChatGPT is familiar with the World up to around mid 2021.
This is the world before ChatGPT. Eventually they will have to update it. How do you separate pristine Information from ChatGPT Information during training? If not addressed properly this will raise the noise level even further.
Yes, but you could create the path just as you plant the grass and amuse yourself to no end on how sheep-like people really are.
Just because there's a path there, it doesn't mean its optimal or good for the public at large.
>it really doesn't necessarily lead to any kind of true decentralized hosting unless someone else has decided to pin your files
You can either pay somebody to pin the files. Or a public Organization like "archive.org" decides that a site is important enough to pin it. The typical surfer will seldom pin something but can keep low freq special interest sites alive.
There is no fee lunch.
With HTTP it has become hard to mirror a modern site. In IPFS it is a built in feature. Makes a better Internet.
"Why" implies to me that there is or should be a rationale. Our culture stores a lot of behavior and spreads it by the means appropriate. Some need rationale most do not.
Killing feels evil and is considered morally wrong. But in reality mankind, like many species, decided at some point, that killing its own should generally be avoided and is therefore allowed only under very specific circumstances.
I feel with the sentiment for the "loss of skill" due to convenience tech.
But hey, these days many people have the choice (meaning the time and money), to keep some of the skills alive. The internet gives you the possibility to find any person teaching the skill set you seek. For more common stuff even Youtube is a trove often for free.