Re: [Haskell-cafe] who's in charge?

Hi Greg, Am 27.10.10 17:56, schrieb Gregory Collins:

Günther Schmidt writes:

...

Hi Greg,

busy no, merely incompetent. If I was capable of writing a good library I wouldn't have bothered, I'd just rolled it myself and published it. I am capable, giving sufficient time, to write code that will get my email for me. But it's gonna be bad, hackish and ugly.

As we are 10+ years now still without one of the most essential libraries any programming language needs I guess it's not that easy. It has just been recently that I wanted to do email via haskell. I was very surprised not find one in place already. As Michael pointed out, there are a couple of basic packages, of varying levels of completeness. Part of the reason there isn't a "complete" solution on par with that of Python's is that sending& receiving email isn't actually as easy as it looks.

No argument there :)

You'd have to deal with several protocols (SMTP, POP, IMAP at a minimum, each with SSL variants and half a dozen authentication schemes), dozens and dozens of RFC standards (see the full list at http://www.imc.org/rfcs.html and allow your mind to boggle), plus all of the weird exceptions and workarounds you need to add in to deal with all of the broken software that's been in use over the years.

Python has excellent email support *now*, but it wasn't always so, and the reason things have improved is that there are a whole lot of engineers employed *in industry* who have been paid to work on these things for years until they got it right. Guido works for Google, you know. I'm not sure how many people are being paid to write Haskell code right now (not including graduate students), but I'd wager the number is less than 200 *worldwide*.

Basically that is my question: If there is someone at the top who has an eye on this. That essential libraries come about. I am aware that Haskell is a project without industry backing so some things will have to happen a lot slower. I'm just wondering if there is someone who steers Haskell in that direction.

There are areas in which we simply can't compete right now and this happens to be one of them.

G Günther