On 15/08/2008, at 2:21 PM, Dean Landolt wrote:
Speaking as a developer that can't get a *nix box stood up at my company if my life (or project) depended on it, I'd say the population is certainly greater than zero. Of course, as an Ubuntu user and all around FOSS- lover, I'm not for lack of a couch sandbox -- but as far as developing somethingfor the 9-5, couch is totally off limits without a windows build.
I stand corrected. I won't bother asking why open source application software is apparently A-OK but an OS is verboten .. think I can guess the answer : (
That said, I totally agree with the below sentiment. Couch is still in alpha -- as long as nothing's done to permanently preclude a multi- platform build down the road, there's not much sense in gumming up the works during heavy development. Sure, couch is *server* software, but as we move on down theroad to a more distributed net, localhost will likely be the server ofchoice for more than a few -- look at what ajatus is doing without even a whiff of a server-side. Once a 1.0 hits, with a Windows build Couch will beone hell of an embed option for desktop developers.I was more referring to the next phase, when somebody would actuallydecide to use CouchDB.In that case i would try to rely on software stacks in their 'native'habitat. But that's just my preference.Geeze, I can *kind of* understand someone wanting to use Windows fordevelopment if they absolutely have to have DreamWeaver or something, but on a server?! Of course you are right that people should run server software, on servers, in their "native" environment. One doesn't use WINE to run IIS on Linux, one should not use cygwin or whatever to run Erlang/ CouchDB onWindows.If you couldn't escape Windows on the server for some unbelievably good reason (like the aliens are threatening to blow up the world if you don't), then it would probably be easier/preferable to run it in Linux under VMWareanyway.
smime.p7s
Description: S/MIME cryptographic signature
