WelcomeWelcome | FAQFAQ | DownloadsDownloads | WikiWiki

Author Topic: future TC wiki for Wayland  (Read 178 times)

Offline nick65go

  • Wiki Author
  • Hero Member
  • *****
  • Posts: 911
future TC wiki for Wayland
« on: August 06, 2025, 09:59:13 AM »
 Wayland: the word can mean a protocol [1] or a display server program [2].
TC users are more interested in the server, much less in the protocol (useful for the programmers).
 
[1] The protocol describes the communication between a display server and its clients (the user's applications).
[2] The server is called the Wayland composite because it also performs the function of a Compositing Window Manager. Its main task is to rendering windows as bitmap on screens/displays.
 
The goal of Wayland is to provide a (display) server with increase security than the previous X Window server (Xorg). Big corporations (and its developers) try to send X/Xorg server into oblivion, by favorising the Wayland server use. Applications that still depend on the “X server” can still be used under a Wayland, using Xwayland (a layer of libraries). But started to show up native applications which do not request X11/Xwalyand layer.
 
There are few types of compositors: Stacking, Tilling and Dynamic. A list of them can be found in many places, such as in https://wiki.archlinux.org/title/Wayland#Compositors.
The popularity or preference of them varies between users, and not all of them are in TC (tiny-core) repositories. You need just one of them to be installed. But of course, you could use many at the same time: like for example you used Xnest sever (as a window) from Xorg server.
 
In Wayland “world” it seams that the display server is spartan / smaller size than Xorg and Wayland has number less API (Application programing interfaces). It means that not all Xorg functions are implemented in server, some of them are transferred to the compositor (window manager). The API is work in progress for each specific Wayland sever. Yes, there are a lot of Wayland servers’ implementations, instead of a “bloat” Xorg implementation of X windows server. It is work in progress (by programmers) to cumulate more API in specific libraries, so that Wayland implementations to standardize this jungle (started a decade ago).
 
You do not need to use a full DE (desktop environment) to be productive. A DE consists, in principle, of a display server (like Xorg), a display manager (like FLWM) which is common for all applications, a drop-down menu (to list the available installed applications ready to run), or a panel (task bar) like wbar in TC; and maybe a status bar (with icons for clock, volume, WiFi). Of course, you can run a display sever even without a display manager! Like you start Xorg and (Firefox) browser, but if you get out of brower then…maybe you need to reset the machine to kill the Xorg.
 
Similar is for Wayland, the server without auxiliary (panel, task bar, menu) programs is near useless for common users. Next steps from here will be to make some categories with those programs and their configuration files.
 
The resurrection of Xorg, as incarnation of Xlibre could be promising, but its future is not certainty. Better to prepare ourselves for Wayland domination, than to be sorry.
 
From my summary testing, the Wayland server labc has the minim size and speedy indeed.
 
A today search for “Wayland” gave me 134 threads, in 12 topics. Let’s summarize those is a topic/wiki without questions & answers, but only facts & documentation, in this topic (future wiki?). Thank you.
« Last Edit: August 06, 2025, 10:15:25 AM by nick65go »

Offline GNUser

  • Wiki Author
  • Hero Member
  • *****
  • Posts: 1640
Re: future TC wiki for Wayland
« Reply #1 on: August 06, 2025, 11:19:42 AM »
Hi nick65go. Thank you for starting this thread!

My two cents:

Situation in X:
user -> application -> window manager (e.g., marco, fluxbox, etc.) -> display server (X)

Situation in wayland:
user -> application -> compositor (e.g., labwc, sway, etc.)

wayland is just the protocol ("language") that applications and compositor use to talk to each other. wayland's motivation/goal is to get rid of the display server altogether. The wayland creators believe a display server (X in this case) is a mostly-useless middleman doing mostly-pointless IPC in a way that adds considerable complexity to the GUI stack.

Bottomline for me is that I would have been happy sticking with X. But then again, I'm not the one who has to maintain it ;D I get the sense that wayland is the future, so I'm doing my best to prepare.

Offline nick65go

  • Wiki Author
  • Hero Member
  • *****
  • Posts: 911
Re: future TC wiki for Wayland
« Reply #2 on: August 06, 2025, 01:15:20 PM »
I will like to underline that final users do not care about words like "protocol" (Wayland, IPC, D-bus, whatever). Neither about in what "language" (C, Assembler, Rust, Java) an application was written. User are (usually) not programmers, they are mostly "consumers" of already prepared (by others) software. So a word like "compositor", which is a combination of Display Server + Window Manager, may sound strange for them. What a compositor will compose (a symphony?). We (Linux users) were too long trained to have a master program who lead/manage the graphic stack. Some people, who came from Windows-OS to TC, maybe did not even make a difference between a (display) server and a window manager.

Offline nick65go

  • Wiki Author
  • Hero Member
  • *****
  • Posts: 911
Re: future TC wiki for Wayland
« Reply #3 on: August 06, 2025, 02:07:13 PM »
I am pondering what a wiki can contain about "compositors".
Because wiki is aimed at new / common users. I do not think expert users use wiki. I even saw /met  a lot of developers who do not even bother too much to write help or documentation for their software; because they continuously develop / debug / improve. And when the software is "finish" / usable, they consider that is waste of their time to document it, it is under their "ranking". Or maybe they want to still have a job, by making them "indispensable" / not-replaceable, so they can provide training, etc.

I also think this topic should not focus on differences / advantages between Xorg and (Wayland) compositors. Out of curiosity I just "google" for subject "xorg vs wayland" and (oh boy / or OMG for religious people) this damn A.I. is all over the places with deep and clear knowledge!

So I am wonder, who the **** today will read a wiki / help when they could just lazy ask ChatGPT. It is very hard to compete with A,I. (artificial intelligence).

Also, I wonder how a lot of interesting list of links to git or Wikipedia or to specific linux distro (Archlinux is strong on wiki) will help users (if they are wiling to bother to read them).

Offline GNUser

  • Wiki Author
  • Hero Member
  • *****
  • Posts: 1640
Re: future TC wiki for Wayland
« Reply #4 on: August 06, 2025, 02:25:12 PM »
I will like to underline that final users do not care about words like "protocol" (Wayland, IPC, D-bus, whatever).
These users will not be coming to the TC wiki to read about wayland. My assumption is that anything we write about wayland here--whether in the forum or in the wiki--is for the users who want to read, understand, and become familiar with the terminology. TC caters to technical users who are interested in concepts and are okay with terms like protocol and compositor. If they weren't, they wouldn't be using TCL.

The only reason AI can scrape accurate information about wayland (or about any other topic) is because human beings with deep and clear knowledge have put what they know in writing, on the internet. But I don't want to get too far off-topic ;D
« Last Edit: August 06, 2025, 02:54:10 PM by GNUser »