i’m not a developer, i’m just a user, but reading some comments make me worry about it because of workforce and money
in example something wrote by someone who seems to know things better than me
my little suggestion is to be carefull, because as you already know you have a budget and if what this guy say is true doesn’t seem a good investment
I think you don’t need to worry. They don’t do it for the fun, but because it simplifies things. The simplest way to get started is probably to fork a Wayland compositor and remove XWayland. Compare the diagrams here (X vs Wayland) and here (Wayland + X).
Also the comment there about “I don’t see upstream happen” - so what? He doesn’t see that. I don’t care. No prove given.
Also, even if it would not get accepted by upstream (which is a truly silly assumption, given that it was seemingly suggested by upstream …), so what? If it does the job, it does the job. It’s not like you constantly have to add features to keep up with technology.
Don’t give in to all the doubters and pessimists out there. Maybe ditch reddit 
Also the comment there about “I don’t see upstream happen” - so what? He doesn’t see that. I don’t care. No prove given.
Maybe you don’t care, but that doesn’t mean that others don’t. During the funding they said that they would be working towards upstream and that’s what I’ve expected (until now). There are multiple reasons why having it upstream, here are a few:
Many of the arguments in the reddit thread are valid. To regain my trust in Purism I would love if they gave us a simple and clear list of pros and cons of developing a new compositor because that is a huge task which should not be underestimated. As someone who has written a wayland tiling window manager (which just barely works and took a lot of effort), I would not recommend them to take this path.
If they would choose to develop their own compositor a year or two because they find some dealbreaker which they cannot solve with Mutter I would respect their decision, but as of now when the only sentence in their blog regarding why is “for performance issues and for better security” I am very sceptical and concerned about their priorities.
The GNOME team has outlined requirements for the compositor they will use in Gnome Shell 4, so if they talked with the GNOME team and made sure that it fits their requirements specified regarding the architecture that could also be a solution however but I see no mention of that either.
https://wiki.gnome.org/Initiatives/Wayland/GnomeShell/GnomeShell4
Sincerely, a concerned Librem 5 backer which wishes the Librem Software Development Team the best of luck!
I back that it is very unlikely as Purism has partnered with the GNOME foundation
Well, Reddit is a diverse community but I believe “ditching” it is a bit harsh. Spreading the awareness around the product and getting feedback about people’s concerns is still valuable 
Maybe losing faith in Purism because of his decision is a bit harsh? So far they have been quite successful in leading their projects to success. A bit of patience from us, and Purism keeping its Progress Reports might be the key to a wider community engagement 
I know I’m joining the “bash Purism” bandwagon here, but didn’t Canonical try the same thing with Ubuntu Touch? And didn’t it fail? A new compositor seems like we are heading down the Mir path again. I don’t want to see that.
i’m reading some blogpost on the web and some forums comments about it, and onestly i’m pretty worried, because alot of people say other project failed because of this kind of wrong step
as i told i’m not a technical guy, but a normal user, just a privacy/linux entusiast, so i will share my thought about it from my prospective
i don’t really care to have a product who have the same interface on mobile and desktop, and even the market say it, windows phone is the big example, people use windows/macosx/linux on desktop and ios/android on mobile, they are different, and people does not care about it, so i don’t like the idea to see a fail there because of this kind of purpose, as other backer there, we just wanna a secure, privacy phone who respect our freedom, gnome, kde, or something else doesn’t really care, people need apps, functionality, stability and security
if the job is done by gnome there is no problem, because the budget is safe, but if the job is done by purism, i onestly feel to be better use the most ready DE and plasma mobile seems the most advanced, we can still download and use gnome mobile DE when is ready from gnome community
i know i’m a normal user who know nothing about business and technology, but keep in mind, even bigger company like microsoft or canonical and even jolla (former nokia employee) made big mistakes who compromise all the project
Actually Mir worked rather well on the Ubuntu Touch (tested it on my Nexus 5 back in the day), but the issue with it was that it took rather long to develop and delayed the project slightly (which is what I’m worried about here aswell).
In the end mir became what it ought to be in the very beginning - a wayland compositor. But hell it took too much time for Canonical to grope the path towards this obvious solution.
@johan-bjareholt, sorry but that was meant to say I don’t care about his unsubstantiated doubt. Of course I care about upstreaming. I just have no doubt that Purism will just do that.
Especially I don’t really get how people get all worked up about “oh, they will do their own damn thing” as a reply on a blog post that lays out how they plan to work together with Gnome and KDE (iow: upstream)
@blendergeek: No, that is totally not the same thing. Mir is not just a compsitor, it’s the whole thing. They stole a lot from Wayland, but wanted to have something under their own control in order to NOT have to cooperate with others. Canonical has a well known NIV Syndrome. Canonical neither partnered with both GNOME and KDE nor did they discuss with the community to arrive at a conclusion.
From our discussions with GNOME maintainers of existing compositors and shells, we may be better off igniting a new compositor (upstreamed and backed by GNOME)
(emphasis mine)
Excuse me, but if you don’t trust that they mean to upsteam when they literally write just that - what exactly could they even do for you? You could as well just assume that they will send you a clay brick next January, in which case you really should not fund the project.
alot of people say other project failed because of this kind of wrong step
@eagle well… I hope I could at least partially explain that such speculation is not backed by facts. People tend to misread, misinterpret, or not read at all or lack the knowledge to understand what they read.
You also might misunderstand some of the parts here. The compositor is not the DE.
While not explicitly stated, it seems that Purism plans to
That is implied by the fact that Gnome is not as mobile-ready as Plasma. So, for the initial release, we will likely see Plasma as main DE, maybe Gnome as preview. But as the stated goal is to avoid the X11 legacy, I assume that they would like to have just that one compositor (PureMutter). Otherwise, KWin (from Plasma) would pull in all the X11 dependencies.
I really hope that this is the case, but the fact that they are not specific enough scares me since I believe not going with the vision of the compositor being a part of desktop gnome in the future and unite would be a huge mistake in my opinion. I really hope this is all just bad communication from their side.
But seriously… they explicitly state that it will be upstreamed and backed by Gnome and the possibility that Gnome will back it, but not use it, even though it is exactly what they have already planned to develop anyway, scares you?