[ragnese]

There are multiple angles. As the stewards of GTK, they should, IMO, try to keep it flexible and customizable to whatever extent is manageable and reasonable. This post is about Mutter, which is a window manager, which should have very little to do with the app "ecosystem". They can, and should, do whatever the hell they want with Mutter, GNOME Shell, Nautilus/Files, etc.

Even in the link you posted, they're talking about GNOME, not GTK.

[marginalia_nu]

When I complain about Gnome driving away users with hostility, it's mainly their GTK stewardship I talk of.

That, and things like primarily designing the interface for a touch screen, despite PC touch screens not really taking off. Very out of touch.

[xethos]

> things like primarily designing the interface for a touch screen, despite PC touch screens not really taking off.

That was actually an absolute godsend using the Pinephone, and IMO laid the groundwork for the Librem 5 (and modern Linux-on-Mobile interfaces) to take root. I do not believe PostmarketOS would be doing as well as it is if they didn't have apps that play nicely with touch.

You don't use it, and you don't appreciate it, and that's fine. I'd say it most defintitely has a place though, without even touching on the chicken-and-egg bit about touchscreen / mobile Linux not taking off vs Gnome pushing for touchscreen / adaptability before it goes mainstream

[marginalia_nu]

I really don't understand why we need to absolutely ruin desktop UIs in order to have mobile interfaces. For web UIs it may be argued as a necessary evil as designing multiple front-ends is expensive and reactive UIs can theoretically be made to exist and shown in small demos to be decent, but when designing desktop applications?

[xethos]

Having a framework that can be adaptable, like GTK, allows for padded, but IMO reasonably-sized touch targets. Designing an adaptive desktop app means the effort is only spent once, but can kickstart the virtuous cycle of "Mobile Linux is less trash than it used to be" -> More users are willing to use Mobile Linux -> More effort is spent making it less trash.

Though if you insist on click-targets that are exclusively for the mouse, I've found most KDE apps less mobile-optimized. The elderly and mobile users can appreciate larger touch-targets, and you can avoid GTK, which seems like a perfect compromise

[ragnese]

Point taken on GTK, and I can't really disagree since I haven't even poked at writing a GTK GUI in many years.

But, you still couldn't resist complaining about the UI implementations, which sounds more like complaints about GNOME apps and GNOME Shell. Who cares if you think that GNOME Shell looks like it accommodates touch screens? Firefox, for example, uses GTK and doesn't seem to look like a touch screen UI to me as I'm typing into this text box.

[marginalia_nu]

The problem isn't that they accommodate touch screens, but that they do so at the expense of keyboard and mouse users, and then they push these changes to GTK in a way where keyboard-and-mouse interfaces become clunkier and GTK-developed UIs become very hard to integrate with other desktop environments.

Firefox has definitely been affected by this. The hamburger button is a touch paradigm which makes no sense on a large desktop screen with a mouse and keyboard-control scheme. It only serves to add more clicks to every interaction. Likewise the reduction of the scrollbar to a scroll indicator.

I was sad when Gnome 2 became Gnome 3 because I really liked Gnome 2 and Gnome 3 was broken. Then I moved on, but where ever I went insanity from the Gnome project kept leaking and making UIs worse.

[pseudalopex]

Would you like to read what an Xfce developer said about GTK stewardship?[1][2]

[1] https://news.ycombinator.com/item?id=40568184

[2] https://news.ycombinator.com/item?id=40568042

[ragnese]

I read them, but I'm not sure what point you're trying to make, or why it's directed at my comment. I mean that genuinely.

This Xfce dev says that GTK4 is less capable than GTK3, and they feel that GTK5 will continue in that direction. They also acknowledge certain things in the first comment:

> [0] Full disclosure: I'm an Xfce developer, and have been disappointed with the direction GTK has been taking for some time. I don't begrudge them their prerogative to do what they need/want to achieve their own goals with the toolkit they've built and maintain. But it really is making life more difficult for me.

>

> [1] Part of the argument is that Wayland doesn't natively support things like cross-process embedding, so a cross-platform toolkit shouldn't have these types of widgets (the classic problem of only being able to support the lowest common denominator). But a) you can absolutely build something like that for Wayland (something I've been working on, though it requires tens of thousands of lines of code to do), and b) with other changes, it's incredibly difficult and possibly impossible to even implement the XEMBED protocol on GTK4, for people who do only care about X11.

If the GNOME guys took out stuff from GTK4 or 5 for bad reasons, then I don't like that, either. Which is basically exactly what I said. However, it sounds like some of these changes would be hard to do and maintain well, such as cross-process embedding. Perhaps the GNOME devs made a decision to focus their surely limited resources toward things they think will be long-lasting. And, perhaps, by their estimation, trying to support Wayland and X11 by adding (and maintaining) tens of thousands of lines of code would be a big burden--especially if they believe that X11 is not going to be super-relevant in the near future. I don't agree with that estimation, and I assume that it'll be a very long time before X11 isn't necessary anymore, but so be it.

All that said, it still has nothing to do with Mutter, which is why I replied to the comment that I did. Because GTK, and Mutter, and GNOME Shell, and GNOME apps, and non-GNOME GTK apps, are all different things, and this post was about Mutter.

[pseudalopex]

You raised GTK stewardship. I replied about GTK stewardship. Why raise GTK stewardship and complain replies are not about Mutter exclusively? Why raise GTK stewardship and dismiss it saying so be it?

The 2 paragraphs you quoted did not represent the 10 you did not.

An Xfce developer saying they can't recommend GTK for new projects outside the GNOME umbrella had information your comment did not. It was not basically exactly what you said.

[ragnese]

Unfortunately, the context starts getting lost as we get deeper into discussion threads like this, but originally, I brought up GTK stewardship because I felt that the top few comments in this thread started to conflate the various projects developed by the GNOME organization. The original HN post was about Mutter, and the first few comments in this reply chain were about software being customizable, etc. Those could've been about whether it's okay or not for Mutter to lose flexibility. But, the one I replied to started complaining about software "imposing limitations on the rest of the ecosystem".

That's when and why I decided to point out that there are different kinds of software projects, and they have different goals and priorities. It's like the old "library vs. application" code: libraries are generic and reusable, and should be written as such, whereas applications are specific and focused.

I brought up GTK simply as an example of a "library project", for which critique of its reusability is warranted, as a counter-example to Mutter, which is an application. Complaining about Mutter's effect on "the ecosystem" is silly. It wouldn't make any less sense to complain about XTerm's effect on the ecosystem by it not supporting Wayland. Anybody in their right mind would just say "So, use one of the other 10,000 terminal emulators in Wayland instead of XTerm"--and rightly so. But, because Mutter is a GNOME project, and GTK is also a GNOME project, I think that people lose focus on what they're talking about.

I did engage with you about GTK because it's interesting, but my point in bringing up GTK was specifically to say "Yeah, those complaints might make sense if we were talking about GTK, but since we're talking about Mutter, they do not." to the comment I replied to.

[pseudalopex]

Your perspective is more clear now. But I disagree.

The article was not about Mutter exclusively. It explained Mutter dropping X11 made GNOME strictly focused on Wayland-based environments. And GNOME Shell would be tied to Mutter even if the article didn't mention it.

Mutter is a library. GNOME is not the only desktop environment which uses it. I don't know if the Pantheon developers wanted to drop X11. But Mutter dropping X11 imposes this limitation on Pantheon.

Your claim the Transmission discussion was not about GTK was incorrect by the way. The GNOME developer said they hadn't decided if GTK would deprecate GtkStatusIcon. The Transmission developer requested GTK make an abstraction. The last comments were a GNOME developer recommending Transmission change architecture so it could support GNOME and in his words whatever odd desktop people want to use after GTK deprecated GtkStatusIcon.

Libraries and applications are not separate inherently. Apple's application scripting architecture was a great strength when it was more supported for example.