They have been promising the transition to GEGL for so long, it is practically vaporware. Yes, I realize that there are GEGL-based effects in the current builds - but I don't GEGL replacing the current core for a long time - and I have my doubts it ever will.
Maybe the lack of devs? I cannot imagine which commercial company would be interested in supporting GIMP development. Let's face it, financial support for open-source software development is critical (KDE/Gnome may be the exception, but the financial investment into these two may be justified indirectly). ImageMagick fit the niche that the web needs server-side image processing library and made money from it. But GIMP is client-side, semi-professional image processing software, who have the incentive to financially support it? (Maybe some professional graphic design companies, but again, they may not have the top-dev for image processing software, that is a big investment for them). I may miss something here, please correct me if I do so.
There is also a more general problem that Gimp is hard to build. I was doing some JavaFX exporter fixes and getting the environment right took most of the time. Otherwise, I found the code base quite clean.
Let's face it, financial support for open-source software development is critical (KDE/Gnome may be the exception, but the financial investment into these two may be justified indirectly).
Well, in the case of KDE there's Qt licensing. I seem to recall some commercial support for Gnome as well.
I don't disagree with anything that you say. Don't get me wrong. I am not blaming the devs or criticizing them. The work they have done is wonderful. But the state of development and the obstacles to taking this forward are what they are.
That's a fair point, but GEGL is actually a thing now, and you can actually use a cut-down, GEGL only GIMP. I think we're beyond the `vaporware' stage.
