I think owning the Framework OS is a bad idea. You might imagine the case of System76 with Pop!_OS. System76 is providing open source hardware, firmware. However I feel that System76’s idea is they decide what’s the best, and sell products to users.
But in my opinion Framework’s unique idea is that they provide a “framework” for users to select and decide their best content. This approach covers the “long tail” market. I would like to see that Framework is in the neutral position between Linux distributions and *BSD. As a result, we are gathering in this community beyond the difference of each Linux distribution, and contributing as a collective knowledge. This is beneficial for each of us.
Yeah, it turns out that my idea was not so good at all, but that’s the beauty of discussing it: everybody shared their point of wiew, and I’m not offended at all; when you’re wrong, you’re wrong and there’s nothing to be ashamed of.
Anyway, my idea came from the enthusiasm for this product and the wish for it to stand out in the crowded market of consumer laptops so it can succeed beyond the expectations and set an example for the rest of the consumer electronic industry. It already stands out, of course, and perhaps the “frills” brought in by a “brand idendity” campaing (such has delivering a custom-branded OS) are not necessary at all.
I have also realised something on my own, which didn’t occur to me while writing about this idea: Framework is not Apple, and it should never be. Therefore, there is not much sense to strictly integrate a custom OS for their hardware, or to push the marketing towards the idea of a close ecosystem. On the contrary: Framework has to be the most open ecosystem in the market!
As bad as my idea is, then, I’m happy that we’ve discussed it, and that some good thoughts came off of it.
I think time will tell. I think some of these linux distro companies with Hardware or Hardware companies with a linux distro are the ones who come out on top. Give it a decade you might be right. I mean look at POP.
I agree. Apple and Framework are very opposite. Apple for “short head” market, and Framework for “long tail” market. Apple decides what the best is, and sell. Framework sell “framework” for users to select and decide what the best is. Both strategies work in a different way. Apple can select the best combination of hardware and software to give users the best experience they think.
There are some other companies the open sourcing hardware and software. But open sourcing the software and hardware is not enough to thrive open ecosystem. Framework just has one product, and the firmware is still proprietary BIOS. However Framework is trying to create systems for users and developers to contribute, and co-create hardware parts. This Framework’s forum system is one of the systems to empower users. I don’t know any other laptop companies that have their own Discourse forum. Framework people’s behaviors on internet is very open. I feel it is sensible for the purpose.
Thanks for providing this topic. It is a nice discussion.
Framework should 100% do this, similar to what System76 has done with PopOS. Yes! Or they should have Ubuntu certify on its hardware. Either way, Linux will need a much better experience. Otherwise, users are going to continue to be frustrated, fill the forums and Framework support channels, etc.
Pray tell, how would they do this? The easiest course of action would to basically take Debian, Arch, OpenSUSE, or whatever and slap their own DE on it. At that point it would be easier to just upstream drivers or whatever else to the project that they would be basing their OS off of. Linux needs no more fragmentation in the market, this is why Linux has the problems it does.
Better idea but involves cost that could be spent elsewhere and would still require the cost of dedicating funds to supporting Ubuntu.
As the OP himself said
Re-read the thread and I think you will understand the momentous task you would be asking Framework to undertake. Not only that but I personally and others on the forum would rather they dev a Free firmware for their product, that would be worth it for all users, instead of a select few (since no small number of users use Windows)
@Trex182 and @GhostLegion it seems to me that the main issue with Linux is hardware compatibility, which is almost a non-issue because it appears that after version 5.15 all the hardware in the Framework laptop is fully supported; the remaining pieces of the puzzle would be the Iris XE graphic and the AX201 wireless drivers, but this is clearly Intel’s job.
To guarantee that every Linux user has a good experience, save for the desktop environment and package management, the easiest course of action for Framework would be to submit all the drivers for its motherboard to the upstream Linux codebase, which would then propagate eventually to all distros downstream. They have already opensourced the Firmware, which is a huge and helpful step, and maybe there is little else to do to complete the task and give an “official” support for Linux with a very minimum effort.
They released source on their EC (Embedded Controller) but the BIOS/UEFI remains proprietary, and even then, the EC was already open-source since they used Chromium-EC (which is fine)
I want Coreboot, preferably as blob-free as possible and the ME neutered and disabled, hang the loss of sleep functionality
You are correct that it is up to Intel to add support in the kernel for their products and to release open-source drivers
I understand what you mean by brand identity. Its a good way of thinking however I think for this situation it would he a negative. Let me explain there are already many distributions out there using the gnu system or another that has included some sort of modified linux kernel. Fundamentally they are all similar in a regard and the laptop can work amazingly well on all of them provided one were to be able to actually have the knowledge to make it work on some of the ‘unique’ ones. Open source software in the end is about you taking the system modifying it ricing it to your needs and your use cases and your preferences and to what you exactly need. A totally free system is a system that you own and is something that is unique to you in a sense. So making a distribution or basing it off another will put a huge toll on the already small framework team causing them to spend more money and spend less time on improving the hardware (which in this case is the number one priority to see more devices like this one in the future). There are a plethora of more reasons why this would be a terrible idea but I admire and appreciate the support you gave the framework community and I hope all our contributions will help change the future and give rise to more free and open hardware such as this one.
The will of Richard Stallman is still alive and will never cease to be.
As most know, the no. of people using Linux compared to windows is more than that of the general world, meaning there are many Linux users of the framework laptop.
As most also know, this is for really good reasons.
And many want to have a framework distro. (See YT comments and reddit and other places)
Please discuss the features, and if it will really be made, in this Topic.
(Note: This Distro mostly won’t happen, so we’re just discussing the features of the hypothetical framework distro)
The Linux kernel is good, but there are some 'microkernel’s that have more modern security and technology, with more efficiency. They just don’t have drivers and are still pre-release.
Using them as optional for the distro would be good.
I find it bizarre that this ever gets requested.
Aside from the bottom line “how could this be profitable” question,
who would even use this?
Most Linux users are pretty satisfied with whatever they already have.
Given that perfectly functional useable Linux distros already exist, there really is nothing that framework could meaningfully add in the shape of a distro without committing some serious System76 levels of resources into the ecosystem.
And while Linux is supported and has more of a voice than usual, that doesn’t mean it’s worth it for framework to hire and sink that much time and money into something whose benefits seem really unclear. You’d be literally better off writing a flatpak/snap helper gui than maintaining your own distro.
