FSF announces Librephone project
Why can't they just partner with postmarketOS here?
Why do we have to have /e/OS instead of a better supported LineageOS, because /e/ is a 1:1 copy anyways?
Why do we have to have a Librephone project now instead of partnering with say, Fairphone and the Pine64 people?
Open source loses this war because proprietary devices are streamlined. The only thing that comes close to this is GrapheneOS, LineageOS, and postmarketOS.
LineageOS has huge problems since the mandatory eBPF requirements of late Android versions, which postmarketOS and its upstreamed kernel drivers could fix. GrapheneOS has huge problems because of Pixel devices, which LineageOS could help with.
We need a unification of this ecosystem because each on their own is hardly surviving on their own against the megacorporations.
Finally! It took the FSF long enough to catch up with the overwhelming usage of mobile devices, but it's better late than never.
I like that this project is trying to tackle something much more challenging that can't be done with just software: reverse engineering device firmware and binary blobs, the pieces of software that actually make hardware components interface with an OS. Understanding how this stuff functions is key to being able to write replacement software, so we may have less non-free software to deal with. I don't have any experience in trying to reverse engineer software, so the best I can do for now is cheer on from outside, unless I want to try my hands at this stuff later.
I also like that this project is not intending to produce an Android-based distro, but focusing more on reverse engineering. Although I read that the results are targeted at helping developers of Android-compatible OSes, the results can hopefully be used by non-Android [GNU/]Linux distros and perhaps other *nix stuff, like the BSD distros. The FSF (by way of developer Rob Savoye) recognizing that a project like this is not going to be quick, easy, or cheap, and is a long term effort is good, as that likely means this project isn't going to be easily abandoned just because of not being able to produce quick results.
I hope that this whole effort can eventually let us break free of the Apple-Google mobile device duopoly, as it sure is getting tiring for me to stick with one of these two companies for my mobile computing needs.
> Practically, Librephone aims to close the last gaps between existing distributions of the Android operating system and software freedom. The FSF has hired experienced developer Rob Savoye (DejaGNU, Gnash, OpenStreetMap, and more) to lead the technical project. He is currently investigating the state of device firmware and binary blobs in other mobile phone freedom projects, prioritizing the free software work done by the not entirely free software mobile phone operating system LineageOS.
The time is right for this project I hope they succeed.
Well… mixed feelings here. I spent a lot of time dealing with early smartphones and hacking away at Android, Tizen, FirefoxOS (remember that?) and several variations on that theme back when manufacturers were vying for differentiation, and I get that the FSF has a mission, but I don’t see this panning out.
Like many folk who’ve been watching Google’s gradual shutdown of AOSP and alignment with Apple in terms of platform lockdown, I think the days of fully open devices are actually coming to a close. Again, I applaud the FSF’s initiative, but you need to get a lot of buy-in for this kind of thing to work—-manufacturers, developers (both OS and app devs), and, of course, users, who will never accept anything that doesn’t let them do things like banking, shopping, mainstream social apps, etc.
And you can’t do a lot of those on an unlocked boot loader (which I think is going to be the logical consequence of replacing bits of the OS) without more hacking. It’s like XML and violence—-it will only lead to more of the same.
I expect the usual amount of “you can do that with web apps” pushback, but let’s be real. Except in markets like India where simpler and vastly cheaper platforms make sense, you either use iOS, Android, or… nothing but voice calls, and I don’t see enough here to make me think this will be something for everyone.
> The FSF has been supporting earlier free software mobile phone projects such as Replicant,
Hopefully this project will go better than Replicant. Here are my notes on running Replicant on the (then already very old) flagship Samsung GT-I9300:
https://www.neilvandyke.org/replicant/
The hardware was a little difficult to obtain in the US, and WiFi worked only with a blob of questionable provenance.
It looks like Replicant has been stuck for several years, and they recognize that they need to find a new device, funding, etc.
(After Replicant, I spent some time on PostmarketOS with various devices, and then gave up and bought iPhones, and then got ticked off and moved to GrapheneOS.)
I wonder whether the FSF is already collaborating with Purism on this, to leverage their work on the Librem 5 and PureOS, which I believe the FSF is well aware of. If the FSF manages to muster a lot more open source volunteers on a more affordable hardware, but that work is also usable for Librem 5, then it could be a win-win. (And Purism also has something called Liberty Phone, which is a made-in-USA Librem 5 phone, so their lawyers should talk about trademarks in any case.)
Ultimately, I don't think the most important challenge is in binary firmware blobs, but the software which people depend upon to run their lives. What does it matter if you can run a completely free software stack on your phone, if your bank software (or your required government ID, as is looking depressingly likely) requires you to run a Big Tech approved phone OS? Perhaps the FSF can't do much about that, but that is where I feel they could truly make the biggest difference for freedom for the average user.
Unfortunately, even if you could completely de-blob the kernel itself (and for many chipsets, that would require a considerable amount of reverse engineering work!), smartphones bear the Curse of the Modem.
In a modern smartphone, modem is often a part of the SoC itself - and it runs some of the biggest and fattest blobs you've ever seen.
It is very inspiring to see a project announced like this with the developer’s name attached to it. As someone who has always struggled with the confidence to be open about my work, let alone work openly in public, it feels extremely inspiring to see Rob Savoye (and Zoe and John behind him) nail their plans to the door like this.
My thrill is matched in strength by the loathing I have for this Apple device on which I type, whose entire boot process is miserably locked down from the very start. It is like a bicycle made from Mickey Mouse logo bolts where the spanners are proprietary and not for sale. The situation is just as ludicrous.
The two major phone OS companies both stand on the shoulders of IBM PC, openly bootable hardware, and the fantastic software systems nurtured and built on top of these platforms — the BSDs, GNU, Linux, and the long tail of all that run on them. It is very troubling that their own platforms are the antithesis of being openly hackable.
Librephone could be successful in a few ways. Outright, as a device, but also as a carrot to bring open handheld hardware to enough people to drive political change (with a small-p, the politics of society, as well as politics of the big-p kind) such that iOS and Android would have to follow suit. With actual public policy Librephone could also end up being a stick: bringing about legislation that requires computers of any kind to be able to boot software of our choosing. Right-to-repair plus plus, if you will.
With enough Librephone devices in the right hands, either the market or the law will demand that we have the same openness and freedom to use our devices the same way we do commodity x86 hardware today. The same freedom imprisoned and exploited in the core of mine and your phone, right now.
The phone is the critical root identity anchor for most of the world now. And many countries outside of the west has already made the Sim card a root identity. Additionally to make it trustworthy (think Google wallet and digital wallets and so on) to work they cannot trust the end user because effectively you the user don't own your own identity. So that's why the phone has to be proprietary - so that it's secure element can be trusted in interactions with the state-big-tech nexus. I talked about my experience with this while attempting to cross borders in SEA. https://polykey.com/blog/architecting-anti-fragile-trust-at-...
There is a lot of work to do to reverse the trend of increasingly locked down computing devices, particularly on mobile.
But from scanning through this press release, this seems nothing more than the FSF doubling down on their failed RYF approach, which does absolutely nothing for user freedom. In fact it's a big negative for freedom, as it ties down resources that could be spent doing something useful in doing something completely pointless like putting firmwares in ROM and adding another chip to load the firmware.
The thing is, firmwares are here to stay. And firmwares that can be stored on the filesystem and loaded by the OS during driver initialization increases flexibility and reduces BOM cost. So that's what device manufacturers are going to do, and RYF will not have any effect on that.
Ugh, I don't know. From a practical standpoint, I can see why basing on Android makes sense. But I really wish we can "somehow" extend an existing Linux distribution (or an Android kernel, even) with a user space reworked to function well on small screens. Maybe that's a pipe dream.
What I'd really, really prefer is to be able to program the device with the same ease as developing a local Linux application. If I need a UI, I'd rather that be a web front end, and not something that needs GBs and GBs of special IDEs and other bloatware. That way, we don't need specialised "apps" for each and every thing: any service that already has website, should work as it is. Just point a browser at it.
And how do I tweak an "app's" UI if I must, rather than beholden to it? That's right: web extensions.
Meta-commentary: At least within the HN community there seems to be a strong interest in a pursuit such as this, given that this is at the top of the front page, and has been for a little while, plus the first page has simultaneously contained these two stories:
https://news.ycombinator.com/item?id=45584498
https://news.ycombinator.com/item?id=45585869
It's heartening.
This seems pretty relevant on the heels of yesterday's popular discussion on how "Free software Hasn't Won" [0] in terms of tools available to the average consumer.
Just because pieces are open-source (or "free software") doesn't mean the autonomy and capabilities we want are necessarily present in the overall system.
There are several comments about Linux on phones años PostMarketOS.
Apparently an usable daily driver already, FuriLabs' FLX1s. Nobody seems to have mentioned it yet.
Interesting that they chose Android as a base and not one of the desktop-Linux-for-mobile ports like postmarketOS.
It's a great idea. Why not join forces with the PinePhone and Librem folks? They're building the hardware and I'm sure they could use more software folk to help out with the firmware and OS.
Librephone is reverse engineering project that attempts to remove remaining proprietary binary modules, not a competing project.
> Triaging existing packages and device compatibility to find a phone with the fewest, most fixable freedom problems is the first step. From there, the FSF and Savoye aim to reverse-engineer and replace the remaining nonfree software. Librephone will serve existing developers and projects who aim to build a fully functioning and free (as in freedom) Android-compatible OS.
I think this mission is doomed to obscurity. FSF is way late to the party and is not bringing the type of resources needed to accomplish this in an impactful way. Reverse engineering device binary blobs will take a long time regardless and by then devices will be several generations behind.
Their efforts would be better spent on building up maintainers of existing custom ROMs and partnering with open source phone hardware initiatives like Pinephone. Maybe even go the route of HarmonyOS and run a completely different kernel and support Android app emulation. They should be looking ahead and not trying to match functionality today and it changes around them.
https://librephone.fsf.org/FAQ.html
Currently scope only seems to go as far as the operating system
Great idea. I'd love to hear something realistic like, "Practically everyone who uses a phone or computer is under surveillance. While this project may seem late relative to the damage done thru unlawful surveillance over previous decades, it represents a start that could lead to better privacy for some early adopters and gradual shift away from the blobs and proprietary technology that place control of our hardware in other hands than our own."
What I'd like to see is open standards for smart phone software: standard ways to have drivers with a stable ABI and API (the goal here is to decouple specific OSes or kernels from specific hardware, if the drivers are free that's even better but is not needed for this goal), a standard way to write data to the internal drive and to boot from internal and external drives (UEFI could work here), and bootloader locking being a thing of the past. We have this or similiar for desktops and laptops. The result of this is that I can run any of hundreds of options for my OS (any major Linux distribution, various BSD options, Windows, and more) even on very old hardware.
I'm currently hacking a toy OS in Zig on the PinePhone, and I have to say the documentation is a bit painful, or sometimes just missing, for parts of these complex SoCs, and that is meant to be a fairly open platform.
But the modem binary blob is a whole other world, and I am not sure how they could tackle that, since my understanding is that this is partly done for carrier licensing reasons? ie. to avoid abuse on the cellular networks. So isn't an open source radio driver also going to have to be licensed in the same way, and then ultimately shared as another binary blob?
The PinePhone compromise seems to be 'isolating' the modem & blob at the end of a USB link. Although I'm not 100% sure how that works yet, since I only just got the graphics & fonts working.
But even that is a bit of a puzzler, since I'm currently framebuffer-to-lcd based, but I know there is a Mali GPU hiding somewhere. I suspect that will also involve another blob. Anyway, the framebuffer approach seems fine for now, it is booting in ~2s, and the less binary blobs involved the better.
It will be interesting to see what FSF can achieve. But, personally I think they would be better focusing on a fully open-hardware dumb phone, and build upon that.
Does anyone know where the Linaro wiki has disappeared to?
An initiative that supports a (quasi) mainline linux kernel and driver support for smartphones seems like the more logical initiative before ironing in a smartphone linux distribution.
The way I read this is that the FSF has gotten funding for one guy to work on this project. Great but as soon as you're doing anything hardware related you need to expend a lot of development effort just keeping up with new releases from hardware manufacturers. It's a never ending treadmill.
With phone hardware lifetime so short, would it be possible to catch-up with hardware update cycle? I guess each new version of a phone can ship with new versions of binary blob drivers. As mentioned in the announcement, reverse engineering the blobs is a huge effort, when it is done, hardware may already be out of sale and the effort would need to be repeated for new versions.
Cool idea, but I’m skeptical. I just want a phone that works—calls, texts, banking apps, and a good camera. If this Librephone can’t run my usual apps or needs me to wait years for it to work with new phones, I’ll stick to my current one. Why not team up with projects that already exist instead of starting over? Hope it works out, but I won’t hold my breath.
> Triaging existing packages and device compatibility to find a phone with the fewest, most fixable freedom problems is the first step
Maybe this is my PM brain but why go after the most compatible devices first vs the most popular devices today?
Surely you’d maximize software freedom by targeting the most used devices so those can switch first
Took them long enough... The free software movement was still stuck on PC despite the fact the whole world moved to mobile. Glad to see they're finally starting to catch up.
They should probably prepare themselves to make ideological concessions... The situation is very ugly here in mobile land. Treacherous computing, remote attestation, DRM, all ubiquitous and normalized...
Two phones might be our sad reality, one for freedom, one for compliance.
Why aren't they sending representatives to 6G standardization bodies? It's too late for 5G and under.
Great. But IMHO better regulation is still needed: force makers to have unlockable bootloader and provided libre drivers for their device (for the OS that they originally ship with); force makers to provide alternatives - for example using alternative "play services" by only providing general API that others can provide pluggable implementaions…
Is there any other way than going through reverse engineering? Projects like LineageOS and others have shown this is really hard.
Why not simply start from scratch and make a truly open source phone? That is, design and build the electronics and the OS that goes on top of it. A bit like an iPhone+iOS but fully open source. Is this dream really unreachable?
All the best for the efforts, however I am bin long enough around this planet to not have big hopes how this will turn out.
Phones aren't x86 hardware, which only got open due to a lucky event, regretted by IBM.
This is exciting, exceptionally the firmware & binary blob foundations that are the biggest roadblock.
Concerning the UI, I wish we had another attempt at a web-based mobile OS. FirefoxOS was too early, but APIs are much more mature now, and WASM offers great performance for low-level stuff. I might work on this full time when I retire.
I applaud the move, but it's going to be really hard if manufacturers aren't willing to document their chipsets and keep bootloaders locked. The folks at Pine64 were forced to waste resources to develop their own platform, which after the enormous effort ant time invested resulted outdated the day it came out of the factory, because of that.
Getting governments, banks and other big entities to play along here is going to be the main challenge.
It's too late when we are moving away from phones and shifting towards wearable technology.
I highly doubt this will takeoff. I'm betting it never works beyond a couple outdated phones.
The fact that there is proprietary software running in "open source" mobile phone OSes may not be addressing the source of the problem. Because it seems that by funding a project like this it almost implies that the parties funding it don't necessarily trust the people who own and thus could open source the proprietary blobs tomorrow.
The leap I seem to have trouble getting to is this. If you can't trust the people responsible for the proprietary software, how can you be sure that they won't turn around and start using new chips or software once the existing ones are reverse engineered? Perhaps it's about patents and the patent holders could be using this IP as a cash cow?
I’m not saying this shouldn’t exist, because it should, but does anyone actually have any faith that the FSF can actually do anything here? They’re like 15 years late to the party
undefined
Thank you John Gilmore.
I guess it will be as successful as GNU Hurd and others.
dear FSF, let's discuss your https://librephone.fsf.org/ web site.
It does the job but it's not easy on the eye.
Full-width line of text. Readability nightmare. Here is how it looks with just a link to a CSS (I closed my eyes on the cssbed.com and picked one at random).
The world could have been very different today if Nintendo or Sony had put phone functionality in the DS and Vita.
Any reason that can't happen now in something like the Steam Deck?
For it to succeed, they must also help put pressure on governments (countries like Brazil or Italy) and banks to stop depending on "Play Integrity" because only Google has the keys (and blocks leaked ones) so we can't count on bypasses being available (it's not just a matter of obfuscation).
This needs to be done before age verification apps become universal..
I am a fan! I have missed this for years.
Good to see someone fighting the fight
I want this, even if it means we have to pay some of the people who work on this.
> Librephone will serve existing developers and projects who aim to build a fully functioning and free (as in freedom) Android-compatible OS.
It may well be that Google will not rest until "Android-compatible" means that they can put their foot down on this. We should be prepared for that eventuality.
undefined
FSF is a joke
I'd like to see an android auto replacement, and them partnering with existing free phone approaches.
Two many xkcd already about creating new standards.
Let's hope the phone's ui won't look like FSF's website.
I suppose my PC's BIOS is a binary blob, yet I run open source Linux on that machine.
>Librephone aims to close the last gaps between existing distributions of the Android operating system and software freedom
I am so happy they are focusing on Android, one of the most popular operating systems widely used by every day people. This is important work for providing user friendly, free software to users.
Let's just hope they don't fall into the trap of disqualifying binary blobs sent as part of drivers vs opting for hardware that harcodes the blob.
FSF never does and never will understand good software. The problem they have is they don't care about the user as much as they care about the developer. They want everything to be easy for the developer and they put the user second.
thank god.
To me this feels like blah blah blah, but I’d love to be very wrong, of course.
Tom is a liar
[dead]
[dead]
[flagged]
How will this phone comply with child safety laws?
*Edit* Because Idiots are Downvoting me, look at the texas law SB 2420 as an example. These phones will essentially be illegal in texas unless they comply with already passed laws.
Looks like we will have to wait forever.
I can't take these jokers seriously.
Years after mobile phones came onto the market they are now planning to create their own phone.