Tech opinion: we need a Linux of phones
And I don't mean like android or like an actual Linux distribution for phones, I mean the same approach to mobile development that's taken to Linux development, which is make shit that works on the worst of devices, have a million forks for users to pick from, try wacky stuff and weird configurations, and let me touch every single part of my device
@gattogateaux yeah I'm not touching anything purism for several reasons and I didn't even know it was founded by a rapist
Don't they run the KDE thing?
@xenofem an actual Linux isn't likely to work well for phones, and yeah partly for those reasons and partly because it was never designed for that kind of interface
If I was better @ coding I'd absolutely set out to write a new operating system for mobile devices optimized for touch from the get go
@dragon the closest thing might be lineageos, i guess, since it has builds for a lot of devices
i've yet to come across a super, i guess, conventionally hacker-friendly mobile OS. maybe mobian or something, but i haven't tried that at all
@dragon Still waiting for a Pinephone.
@CyReVolt not what I'm talking about
@dragon What I am sayig is that I intend to work on such projects. A suitable platform would be necessary as a start.
Some related projects:
- KaiOS, a fork of FxOS, formerly B2G
Are you also starting somethig? I'm interested in exchange. \o/
@CyReVolt unfortunately I don't have the skills or time required right now to do anything, but I've sorta explained what I meant in my reply to xenofem earlier, Linux was never designed for touch interfaces and I don't think a Linux based project would work, if I could I'd absolutely start a project for a fully free touch-focused operating system, something made for phones and tablets from the start and that's built on stable technologies as opposed to like java
@dragon Alright. Aren't most POS and other touch devices in the embedded space based on Linux? What would be wrong with it? It wouldn't have to be Java of course, so as I wrote, Hellaphone, FxOS etc are very viable options. The problem is really just with vendors not publishing sources. You'd have to reverse engineer many things.
@dragon When I looked into PostmarketOS, they support multiple desktops targetted towards phones! One of which, ofcourse, is a commandline.
Apparantly the default is a simple Weston one which I might try first, though there's also ones based on KDE, XFCE, & GNOME.
The issue is that many of those devices are more locked-down than computers were when Linux started. I'm not sure how to correct that...
@alcinnz I don't know how to say this anymore I don't mean a literal Linux distribution for cellphones, a terminal on a phone is cool but absolutely not the way that anyone would want to use a phone, and neither is a desktop!! Linux is not and never was designed to be touch optimized, we need something similar in mindset but absolutely not linux
@dragon Yeah, heavy modifications are required. We can reuse Wayland, etc and many of the same standards, but the desktops need to be heavily customized if not outright rewritten.
Most of the infrastructure now supports touch, the FreeDesktop.Org standards still match up to what we need, but the UIs aren't designed for phones.
I'll give a review when I get my PinePhone to what extent your fear is accurate...
@alcinnz we can use something else than Linux, a microkernel architecture would be a much smarter approach considering the overwhelming diversity in hardware and architectures, Linux already struggles to handle WiFi
@dragon I'm by no means against substituting the kernel... I care about software freedom, not a particular kernel!
I would want to keep the FreeDesktop.Org standards though. I see no reason to replace them for phones, and it'd help me develop novel browsers for them.
@alcinnz maybe but it'd need a lot of fundamental changes, we don't deal with data on phones the same way we do with computers at all
@dragon The changes I can see are:
1) Reduced if any need for the MIME info database, and several other specs, due to lack of a user-visible filesystem.
2) Standardize elementary OS's "Contractor" protocol.
Can't see any changes being needed to the very relevant .desktop (app entries), icon themes, Media Player Remote Interfacing Spec, notifications, & AppStream (app repository) specs. Dispatching MIMEtypes/URI schemes to apps is less needed but still useful.
@dragon Hardware, software, firmware, carriers, ... ?
All of the above?
Every problem has lesbian solutions! join in on the fun!