discuss-gnustep
[Top][All Lists]
Advanced

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

Re: A Critique: Getting Started with GNUstep on Windows


From: Doc O'Leary
Subject: Re: A Critique: Getting Started with GNUstep on Windows
Date: Sat, 20 Feb 2016 16:22:45 -0000 (UTC)
User-agent: com.subsume.NNTP/1.0.0

For your reference, records indicate that 
Dan Hitt <address@hidden> wrote:

> Much better if the valuable parts of the stack (Gorm/IB,
> ProjectCenter, copy-paste apparatus, dock, window manager)
> were tuned to a particular OS that the gnustep developers control.

Absolutely.  But I question if that is a realistic aim, given the
resources currently available.  I don’t even see a documented reference
platform (closest thing appears to be what’s on the downloads page),
which would be a reasonable starting place.  Then on top of that I
could see building an umbrella package that had all the necessary
dependencies.  After that you could probably put together a custom OS
distribution.  And that’s just on the Linux side, where you actually
*could* control all the pieces.

> They should never be in a position of having to respond to
> a change in the kernel, or a change in the compiler,
> or a change in the XYZ utility.  Instead, they should choose
> the kernel, the compiler, the utilities, and accept updates
> only when it makes sense for the top of the stack.

I think that’s wishful thinking.  External dependencies will always
exist, and bugs and security issues should *not* be ignored just because
someone wrote some code higher up on the stack that wasn’t encapsulated
properly.

> I think Apple has only rarely put itself in a position
> of producing software on a stack it didn't control.
> And when it did (quicktime) it was for a very definite
> reason, and of course they had enormous resoucres
> to do it with.

Apple has enormous resources to work with regardless.  GNUstep can’t
match them; I’m not sure any company can.  It is an endless source
of amusement for me as a developer when a client asks how Apple does
something, because I get to answer with some variant of “They build
a second campus for $5 billion and fill it with people.”

> I sure agree with the approach --- "modest and calculated" --- but
> i think running on somebody else's stack is neither of those :)

I’d love to see the calculations that show that managing the full stack
is more modest.  :-)  Apps already exist, OSes already exist.  Let
other people do that heavy lifting.  The task of getting Cocoa to
function on *any* non-Apple platform is burden enough for one project.

> Well, they do --- i mean, the page you see when you visit gnustep.org
> is very clear (imvho) about what you're getting into: you're
> getting into Middleware City.

I don’t think it’s clear at all.  Especially not to a newcomer, who has
absolutely no idea what the history of the technology is.  The web site
remains a muddled mess, because there is no vision that can actually be
communicated to a visitor.

> Maybe i should pose this as a question: if you were writing
> a "specific stated vision/plan" for gnustep, what would it say
> (just briefly, as an overview, not detailed)?

GNUstep is a project to bring great software from the Mac and iOS to
other platforms, like Linux, Windows, and Android.  It is also a way
to portably develop software for those platforms using the same
technologies that have turned Apple into a juggernaut over the
past 10 years.

> What would it say if it were to reflect what you want?

What I want is measured by what can reasonably be achieved.  There may
be some future where GNUstep is bigger and can do more, but it has to
attract a substantial number of developers if it wants to shoot for
the moon.

> What would it say if it were to reflect the reality of gnustep today?
> (Would it be different than what's there today?)

GNUstep is a project that mimics something from the 90s that you’ve never
heard of.  Here’s some code; you figure out what to do with it.

-- 
"Also . . . I can kill you with my brain."
River Tam, Trash, Firefly




reply via email to

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