On 27 Nov 2019, at 12:29, David Chisnall wrote:
I am no longer actively working on Foundation or AppKit, so I
shouldn't get much of a say in what those projects do, but if GNUstep
is going to be tested with only GCC then my advice would be:
- Actively market GNUstep as only an OpenStep implementation, drop
all Cocoa references.
- Drop support for Clang.
Currently, the project is setting itself up to fail by advertising
features that no one tests.
Ok, this is the third thread on this topic but I can only support it
from a user’s and newbie’s perspective. It was quite disappointing
when I discovered that this Cocoa compatibility thing is very limited
taking into account the current feature set of Cocoa.
Thinking about it a while and seeing Sergii’s Nextspace project one
could consider you can build a fully functional desktop experience using
the OpenStep API.
So it might be a less overambitious goal to polish the existing API,
keeping GCC compatibility and thus preventing a fork of the community.
(The clang people might distribute their ARC enabled apps as they are
trying currently and thus testing and improving clang runtime quality.
I’d like to do that.)
Even using a GCC compatible API/libs there are many things that could be
improved to target a wider user audience:
1. Polish the look and feel of AppKit and make it integrate well with
current GTK themes (same for Gorm).
2. Work on the integration of ObjC with f.e. gnome-builder.
3. Upgrade xib integration to work with interfaces created with recent
Xcode releases.
4. Provide a plugin for Xcode that warns the user when he uses classes
or API calls that are not part of the OpenStep API or implemented
GNUstep API so that cross platform developing and porting apps becomes
more easy.
Some unfinished thoughts…
The clang/ng runtime environment distribution keeps important to me,
though.
Johannes
