Path: csiph.com!eternal-september.org!feeder.eternal-september.org!nntp.eternal-september.org!.POSTED!not-for-mail From: John Ames Newsgroups: comp.os.linux.misc Subject: Re: Wayland Makes Progress Date: Mon, 4 May 2026 09:43:07 -0700 Organization: A place where nothing fits quite right Lines: 44 Message-ID: <20260504094307.00000494@gmail.com> References: <10sv1kk$gjtg$1@dont-email.me> <10t001q$pcmf$5@dont-email.me> <20260430095306.0000330b@gmail.com> <10t01vb$2p94u$1@dont-email.me> <20260430101559.00000fe5@gmail.com> <10t0kv8$10lf6$8@dont-email.me> <20260501135638.000014d3@gmail.com> <10t3a7f$1og72$5@dont-email.me> <20260501155425.00007474@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Injection-Date: Mon, 04 May 2026 16:43:17 +0000 (UTC) Injection-Info: dont-email.me; logging-data="4014044"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX19Ur72zw/Sgk0PIZv+An4KAhzM8cOhXNdU="; posting-host="76dbb82330bc37174905376d955f9084" Cancel-Lock: sha1:uk5Fgey/uvFE+lTy5/QifYACfFY= sha256:NMf1AaKwe2MtO0kBluvSdT9AXMSJ6jbtHZr+hQsFlZM= sha1:4/fZACpId/onRpb/2f3wtKqAzjY= X-Newsreader: Claws Mail 4.3.0 (GTK 3.24.42; x86_64-w64-mingw32) Xref: csiph.com comp.os.linux.misc:86125 On Sat, 02 May 2026 10:25:32 +0100 Richard Kettlewell wrote: > If a user moves a window to a particular location, or configures a > particular rule in their window manager about how menus behave, or > where application windows go, etc etc etc then the application should > not be allowed to override that. In the general case, I'd agree - but UI conventions on the *nix desktop have evolved in a few different directions over the last 42 years, and "multiple sub-windows in a predefined layout" is 100% a valid design that multiple different applications have adopted for their workflow. If their concern were really for preventing rogue/nuisance applications from disrupting the user's desktop arrangement, they could've had some kind of application-specific permissions system (as, say, OSX does) so users could configure this themselves - could even opt-out by default. But flatly disallowing it altogether - to the point where it's not even provided for in the protocol - is an *incredibly* blinkered view of UI design that more suggests a developer mindset of disinterest in anyone else's way of doing things. Which would be fine for someone's personal project - but emphatically *not* fine for a project that proposes to replace a huge piece of infrastructure relied on by many millions of users across thousands of different applications/use cases. It's like if the IETF decided to exclude port 6667 going forward 'cause they personally didn't care about IRC. > This is _not_ inconsistent with applications that want to create > multiple windows in a particular spatial relationship to one another, > but it does mean that just specifying absolute coordinates isn=E2=80=99t = going > to do the job: rather it means that such applications need a way to > communicate the desired relationship. AIUI, MR264 is an approach to > implementing exactly that. Maybe so - but Wayland is 17 years old at this point, while MR264 is ~2 and if you look at the comments they were still trying to argue that You Don't Need That as of a year ago. Whether or not MR264 becomes an acceptable solution, it's incredibly clear that the developers do not prioritize user needs over their own ideas of How Things Should Work 'til 800-lb. gorillas like KiCad force their hand.