Am Donnerstag 05 November 2009 23:02:30 schrieb Erik de Castro Lopo: > Andrew Coppin wrote: > > I'm dissapointed that Haskell doesn't have *more* of a Windows bias. It > > _is_ the platform used by 90% of the desktop computers, after all. (As > > unfortunate as that undeniably is...) > > That is not true in my home and its not true where I work.
Neither is it true in the group of people I know. However, the number of computers in that group which had Windows installed when they were bought may be close 90% - it's close to impossible to buy a completely assembled, ready to go computer without here - which is easily remedied by inserting an openSuse or Ubuntu disk as soon as it's connected to power :) > > In addition, saying "90% of all desktop computers" is misleading; > instead we should be talking about the computers of software developers > and there, the figure is almost certainly well below 90%. > > > In particular, I really wish we could make is so that stuff from Hackage > > actually compiles on Windows. (Disclaimer: Stuff written in Haskell > > compiles just fine. It's FFI bindings that unanimously refuse to > > compile.) It's also somewhat irritating that the I/O libraries have a > > few quirks apparently related to mingw32. But hey, that one at least > > should be fixable... > > The problem here is that window is the odd one out. Still it would be nice if things were easily installable on Windows, so maybe some Windows user should write a tool which makes installing C libraries on windows feasible. > > Stuff written for userspace Linux will usually compile with little > more than minor alterations on OSX and all the other Unix-like > systems. Making that same code build on windows can be a significant > amount of work and that work should not be the responsibility of > the people who write code on Linux and Mac. > > Erik _______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe