Specifically, though, this comment is a pretty good example of what this article (and I, now), am talking about:
> Certain problems, like working with databases in the principle Haskell way, are still open questions (e.g. see effect systems). But to call a mere difference in approach "arrogant" is extremely arrogant in itself
Which points to a problem very specific to haskell, brought up in the article, which is "How do I actually get things done?" Which, according to that comment (supposedly in support of Haskell) even points out that something as obvious and boring as "using a database" isn't clearly defined in Haskell. Most programmers want to use a programming language to solve a problem. The haskeller's argument, I guess, is that Haskell tries to do that while also applying very strict constraints on how problems are solved. Great, right? Except that those constraints are so strict that even problems that aren't significant or meaningful are difficult/not well defined (like using a database).
So if the answer to "How do I get things done?" isn't "Like this" but instead "Haskell doesn't work that way", most programmers will consider this a nonstarter.