discuss-gnustep
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: GNUstep.sh / env sanity patches


From: Armando Di Cianno
Subject: Re: GNUstep.sh / env sanity patches
Date: Thu, 19 Aug 2004 10:05:41 -0400

On 2004-08-19 07:50:16 -0400 Dennis Leeuw <address@hidden> wrote:
What I don't understand is why should you or any developer be forced to put a GNUstep based server in a standard un*x location? One writes a GNUstep-based server and one thus needs the entire GNUstep environment (in the case of a server atleast -make and -base). So it could just live within the GNUstep environment with e.g. just a init-script to start and stop the so clearly written server.

What am I missing?

Dennis

I for one will probably never install any bit of GNUstep into /usr/lib. Pretty much right when I started the package management Gentoo thing wrt to GNUstep, I had convenient models to follow. On Gentoo, KDE is installed into /usr/kde/<version>.<num>/, for example. (Disregard any notion that I think Gentoo is the "one true way" ... I'm just using examples from my own background/context).

So I see it, and I think most see that what your saying is true. X11 itself decided to pretty much hardcode itself into /usr/X11R6, at least in a "it's the right way" mentality kind of way, if not hardcoded-ness. However, dynamic libraries are meant to be dynamic. That's kind of the point: they can be shared, they can be updated, they can be moved.

Now, for GNUstep's sake, your argument is true. However as users/developers of GNUstep, who are on this mailing list, having this discussion, we are also inherently _evangelists_ of GNUstep. We need to "woo" people to GNUstep. We know it's easy to woo developers, once they grok the power of Obj-C + Foundation/AppKit, but pretend we need to woo project managers or (egad!) manager-manager types. The ability to say "This will fully integrate into our UNIX solution" is a HUGE selling point, and furthers the growth of GNUstep. But of course I see your point "JUST PUT IT IN /usr/GNUstep AND IT JUST WORKS!!", which makes sense to any of us. But to expend the energy to make things slightly more reconfigurable wrt to the underlying system is worth it, I believe. Consider it the cost of "woo". If the cost is simply too high, we should say "No", however, it sounds like it is not.

__armando





reply via email to

[Prev in Thread] Current Thread [Next in Thread]