If learning it is no harder than any other programming language that presumes that every other programming language is equally easy to learn. Furthermore it assumes that people have an equally easy job learning whatever language they learn.
Neither one of these conditions are remotely indicated by research on the subject, therefore the statement is false.
Also I did try learning Haskell at one point and I found it harder than some other languages, I would say I found it Erlang level hard which I tried at about the same time( about 2009) but I think Erlang has a little more inclusiveness to their community IMHO
> Neither one of these conditions are remotely indicated by research on the subject, therefore the statement is false.
That does not follow. "Research does not prove it" means "therefore we don't know it's true", not "therefore it's false".
To your wider point: I think (but cannot prove) that some peoples' brains find FP easier to learn and some peoples' brains find procedural easier. And I think (but can prove even less) that the large majority find procedural easier.
> Neither one of these conditions are remotely indicated by research on the subject, therefore the statement is false.
What is the standard unit of measurement for quantifying just how difficult a language is to learn? Is there a standard unit of measurement for quantifying the competence of a person learning a programming language?
I'm pretty sure the field is too young to have developed a standard unit of measurement of either of these things, if there were such units their creation would be one of the stunning intellectual achievements of mankind in recent history and would have such obvious benefits to many other parts of human endeavor that I would expect it would be a field more investment heavy than machine learning at the moment.
However it has been shown that there are people who seem more attuned to different learning styles, styles of programming, there are differences in difficulty between first language acquisition and later (dependent on language similarity, language domain), and many other studies regarding programming language learning that it can be said not everyone learns equally well every language, and not every language is equally as learnable.
> it can be said not everyone learns equally well every language, and not every language is equally as learnable.
I think that’s reasonable, but the question is then at what point does a language become sufficiently difficult that it no longer provides a good return on investment? And to what degree of proficiency must one achieve in a language in order to productively use it?
It doesn’t take a long time to learn all of Elm, and you can be productive in it quickly.
It takes a very long time to learn all of Haskell, but it does not take a very long time to be productive with it.
totally agreed, and I suppose in Paul Graham's concept of a Blub programming language there might be a language that is difficult to become proficient in but allows you to achieve things that would not be reasonably achievable in other languages.
> If learning it is no harder than any other programming language that presumes that every other programming language is equally easy to learn.
Did you click through to the article? The HN title is (currently) "Learning Haskell is no harder than learning any other programming language". The article's actual title is "You are already smart enough to write Haskell".
You appear to be responding to what a zealous mod wrote not what anyone actually believes.
I am responding to what the modded title is because it is wrong.
the article's actual title is wrong in other ways.
1. because someone might not be smart enough to write Haskell (is that different than program in Haskell?)
2. Because someone might be better suited to other language types than Haskell.
language types I prefer - small instruction sets, functional or declarative (that however is my preference) someone might just find object orientation a more natural way of thinking.
Neither one of these conditions are remotely indicated by research on the subject, therefore the statement is false.
Also I did try learning Haskell at one point and I found it harder than some other languages, I would say I found it Erlang level hard which I tried at about the same time( about 2009) but I think Erlang has a little more inclusiveness to their community IMHO