What language people learn out of own motivation, vs what languages people do for pay.
I'd say based on this graph it, and knowing Haskell myself, that choosing Haskell is still a viable way to attract talent. (that Haskell-spike in the animated chart half through the article is... opportunity I guess)
Just be aware that the animation makes it look like everyone is using Haskell in the evening. That's false and caused by normalization. Other languages are still more used.
That's right: Haskell is disproportionately used in the evenings, but it's not highly used at any time. You can see in the 4th graph that of the 250 languages considered, Haskell is one of the lowest traffic.
This is assuming that SO is a popular destination for Haskell questions.
Github would be the better indicator. Especially if Github could include anonymous stats for private repos too. That would give a good indication about the commercial viability of the various languages.
I'm one of those people who uses Haskell a lot at night time and I rarely use SO for it. I often find answers in documentation, blog posts, IRC, and books. Whereas during my day job I tend to run into a lot of obscure framework errors when I work with Ruby on Rails or the various weird quirks in Javascript where SO is indispensable. I have different kinds of queries with Haskell rather than copy/pasting error msgs, likewise when I use Erlang/Elixir.
I'd say based on this graph it, and knowing Haskell myself, that choosing Haskell is still a viable way to attract talent. (that Haskell-spike in the animated chart half through the article is... opportunity I guess)