Until it has produced as many sites, I consider any such statements as anecdotal.
That is, I'd rather measure a system's productivity with actual production in the wild than with any of the systems "inherent" capabilities.
It might be productive for the author, but I don't see the general web programming public finding it more productive. People have learned Ruby to use Rails, but not many have ventured to learn Haskell to use its frameworks.
>It's not as popular therefore it's not as good. Your logic is definitely valid!
And your logic is definitely faulty. I've never used the word "good"
What I said is "more productive" (what the author claims) can only be measured in actual PRODUCTION.
That is, the important thing is not:
(a) "If I were to use X framework/language, how productive would I be over Y framework/language?",
but:
(b) "In an actual empirical observation, what framework/language is actually responsible for the largest volume of production?"
The author talks about (a), and advocates Haskell. But that is not an empirical, scientific, measurable observation, it's just his personal opinions, feelings and anecdotes. Only (b) gives an actual overall metric of the productivity of two frameworks/languages combos.
Even having the same person doing the exact same project with both X and Y framework/languages and comparing the speed with which each was done, would tell us very little. Maybe someone he was more comfortable with one or the other, maybe that particular project fitted especially X over Y, maybe it didn't need to communicate with legacy stuff with neither X nor Y do well, etc.
The only way to tell what generally was for production for the majority of people, is to, DUH, see what the majority of people have used for their productions.
If we are empiricists, yes. And I say that as a Django man.
If we are to held an idealist view, no. But it would just be ideology making up for a lack of the same volume of production being done with Ruby compared to PHP.
Remember the "worse is better" motto? Worse could also be more productive.
Now, I don't care why PHP is more productive in actual volume of production --instead of more productive as in "it makes you more efficient". It could be because of "stupid" programmers that cannot adapt to Ruby, because of inertia, because it is fast to start with, because it has a more vibrant ecosystem than Ruby/RoR, because of large amounts of code already built with it used to bootstrap newer projects, because of lack of RoR publicity, because of just being there first, etc etc. Thing is: by usage and number-of-sites metrics, it is.
That is, I'd rather measure a system's productivity with actual production in the wild than with any of the systems "inherent" capabilities.
It might be productive for the author, but I don't see the general web programming public finding it more productive. People have learned Ruby to use Rails, but not many have ventured to learn Haskell to use its frameworks.