Nice example of how good programming is often about meeting systems where they are:
Whatever your feelings are about OOP and memory management, the reality is that if you choose GTK, you're forced into interfacing in some way with the GObject type system. You can't avoid it.
Well you can avoid it and we did avoid it. And it leads to a mess trying to tie the lifetimes of your non-reference-counted objects to the reference counted ones. There was an entire class of bug that kept popping up in the Ghostty GTK application that could basically be summed up as: the Zig memory or the GTK memory has been freed, but not both.
GTK having this kind of scheme is why Vala was made and inspired by C# but uses GObject and all its joys, then transpiles your code to C.
This is why a surprising number of GTK applications (more than I realized) are coded in Vala. I lowkey wish they had just adopted D instead of building up Vala. D is basically compiled C# to me in its own ways.
The GTK C API definitely feels like it's begging for a domain-specific language. I guess that language turned out to be Vala, and from what I could see it was a pretty nice language.
My takeaway from all this is a little bit different though: I feel this whole situation ultimately some credence to the Scheme/Racket philosophy, where you can effectively make yourself a domain-specific language without ever leaving the base language. In an alternate universe, we don't choose between adopting D or inventing Vala. We can just stay in Racket.
The problem is that D community keeps pivoting all the time, so now you get compiled C# with the breath of .NET ecosystem and IDE tooling, and is almost impossible to sell D, even if still has the edge in a few language features.
I'm not talking about merging .NET into D I'm talking about instead of building Vala which transpiles to C, the creators of Vala could have just made a D library instead. D does not need the .NET ecosystem to survive.
D needs a usable ecosystem, the point is that since Andrei Alexandrescu's book came out, the incomplete pivots have allowed the competition to get most of D features where it had an edge back 15 years ago.
Even Andrei eventually got back to C++, now working at NVidia.
Native AOT or .NET Native, plus all the Midori related improvements, and everything else C# has access to, was one example, there are others from other ecosystems.
Careful what you wish for. These days one has to be thankful if a badly designed old system isn't replaced with a new system that's even worse despite the best intentions ;)
If you believe everything around you must be good before you can do good, then you must trick yourself into believing imperfect things are good to do anything, or withdraw completely from the world. Your radical philosophy leads either to delusion or isolation and inaction. Just do what you can.
Only in a dictionary. In a sentence that has the context of a poorly designed system, they are because you explicitly have to do worse programming to meet the system at a worse point
Being pragmatic about the design of an ecosystem and working with it, could be seen as good programming vs the alternative of fighting that ecosystem and thus creating entirely new classes of bugs.
And this was what the GP was responding to. They were was that the pragmatism of working with that ecosystem showed good judgement.
