Yes, I believe the course of the history of the Linux desktop would have been very different had GNUstep took off, especially around the crucial years of 1996-1

Yes, I believe the course of the history of the Linux desktop would have been very different had GNUstep took off, especially around the crucial years of 1996-1998. In our timeline KDE started in 1996 and GNOME started in 1997, and neither desktop adopted GNUstep, instead preferring the Qt and GTK+ GUI frameworks, respectively. However, suppose instead of KDE and GNOME being developed, the community rallied behind GNUstep, speeding up its development. Had GNUstep been completed by 1998 or 1999, this would've been revolutionary, especially if some type of Wine-like compatibility layer was built to support NEXTSTEP/OPENSTEP binaries, thus immediately providing the Linux desktop high-quality commercial applications such as those from Lighthouse Design (https://en.wikipedia.org/wiki/Lighthouse_Design https://en.wikipedia.org/wiki/Lighthouse_Design), as well as providing a path for companies with legacy NEXTSTEP/OPENSTEP codebases to bring their code over to GNUstep running on Linux or one of the open source BSDs (IIRC, NEXTSTEP was built on top of a 4.2BSD userland running on the Mach microkernel). Why did we end up with KDE and GNOME instead? I wrote about it in more detail here (https://mmcthrow-musings.blogspot.com/2019/06/some-thoughts-about-gnustep.html https://mmcthrow-musings.blogspot.com/2019/06/some-thoughts-...), but in short, my opinion is that (1) GNUstep wasn't ready when KDE and GNOME were under development; the developers involved preferred to use existing GUI toolkits, and (2) there might have been a reluctance to using Objective-C in the mid-to-late 1990's. Another fascinating consequence of a world where GNUstep took off is two competing implementations of the OpenStep API: Apple's (Cocoa) versus GNUstep's. Would Cocoa and GNUstep have diverged from each other, or would GNUstep have followed Cocoa like in our actual timeline? Would a GNUstep desktop have (1) stuck to OPENSTEP's UI conventions, (2) moved to Mac OS X's UI conventions, or (3) chartered a new path? We'll never know, but it's fascinating to think about what could have been.