FSF announced a new project: an initiative to reverse-engineer obstacles preventing mobile phone freedom until its goal is achieved.
11 Likes
Maybe Iâm too cynicalâŚbut I have no hope for this project to be helpful.
The first readily-available, commercially-successful smartphone was the Blackberry in 1999. The first iPhone was released in 2007, nearly 20 years ago. Android was first commercially available circa 2008. There have been almost 20 iterations of the iPhone, and thousands of Android phonesâŚand it took this long for the FSF to stop being the old man yelling at the clouds (theyâve made MANY statements over the years saying âGoogle phones bad! Apple phones bad! Donât use them!â, but itâs only now that theyâre saying âwe should maybe make a device running Free Softwareâ.
But letâs move the football down the field a bit, wave our magic wands, and bring their intended outcome into existence - a phone exclusively running FLOSS⌠LineageOS, GrapheneOS, iodeOS, and /e/OS arenât open-source enough for them, so weâre bringing a whole GPLv3 OS into existence for this deviceâŚand then what? It doesnât seem theyâve put any push toward F-Droid, or a similar project. Theyâve made no claims to do what the /e/Foundation does and provide PIM functions like mail/calendar/contact/task/photo syncing, either to an FSF-hosted software or a self-hosted server like /e/OS does. They offer no alternative to Google Drive or Instagram or Spotify or BitWarden, and theyâre not incentivizing other developers to centralize around such things.
The press release talks about trying to reverse-engineer some of the binary blobs LineageOS includes, as if thatâs the single biggest barrier to freedom or FOSS adoption in the mobile space. Even if that were to exist, go ahead and try to pitch the idea of a phone that canât use banking apps, canât use WhatsApp, canât use Instagram or Tiktok or Netflix or offer alternatives to them, canât sync passwords anywhere, and doesnât have an App Store (the FSF isnât listed anywhere on the F-Droid website that I saw). Thereâs no equivalent to iMessage or FaceTime, you canât have a work profile and set up your work phone because that would involve ActiveSync, and if you do want to set up PIM on your phone, itâs a multi-step process involving IMAP/SMTP/CALDAV/CARDDAVâŚbut you can modify your modem firmware!! And you can write your own software that you can then install on your phone! âŚvia a USB cable or typing the code out by hand on your phone with the integrated gcc compiler!
âŚI could have a bad day with a circular saw and still count on my fingers how many people would be interested in such an experience. Even the folks here in the /e/OS forum are going to be completely disinterested, because, by definition, weâre okay with at least some binary blobs, if it means that our Android Auto and our banking apps work. Now, sure, Iâm certain that all of us would love for there to be a day when apps no longer require Micro-G or signature spoofing, ideally because there are viable alternatives to the apps that require themâŚbut we donât get there by starting at a âfully free OS that appeals to an exceedingly extreme minority of an already-niche group and then building out from thereâ, it starts in reverse - get an ecosystem going that will enable users to shift to the ecosystem before ditching their iPhone, then make the phone change just the next logical step.
Maybe the FSF will come up with some solutions to things that can then reduce friction for Ronnie and the Lineage teamâŚbut the press release really reads like a desire for purity from a place that so few people get to experience to begin with. This is why I think /e/OS has the right idea overall - create an ecosystem, ease the data transfer, simplify the installation, make sure that the closed-source apps work, to the extent that can be meaningfully implemented under the circumstances.
Until the FSF sees their goal as a growing series of intermediate steps that will require some compromise in some places on the outset, all that will happen is that users will have to choose between the restriction of the Apple and Google ecosystems, and the restriction of âonly the free-est of free software; lest one taint the purity of the source by daring to view a .mov video!!11â.
4 Likes
What about postmarketOS or Ubuntu Touch ?
But Iâm agree with you. LineageOS, /e/OS, CalyxOS, iodeOS, GrapheneOS make already good work 
1 Like
From FAQ
What is LibrePhone
This is a project to research proprietary files in Android to work towards a long-term goal of free replacements.Are you developing a free mobile phone OS?
NO. There are many projects working in mobile phones, and many of them are largely free software. The FSF doesnât see the need to join these projects, but wants to build upon them and improve on their current state of freedom. While there are many proprietary libraries and other files in Android,this project is focused on binary blobs.
My point is: I do think that /e/os might be able to gain something from this FSF project.
3 Likes
I tried a Pinephone about 3 years ago with Ubuntu touch and Mobian, primiative; but I pefered them to Android. Unfortunatly the hardware was terrible. My opinion for what ever it is worth is: I would like to see an phone/tablet OS that is not Google dominated. As a Linux user Ubuntu Touch and Mobian felt like home.
2 Likes
Thatâs a prety big response/opinion for something that was just announced:
âFSF announces Librephone project
by Free Software Foundation â Published on Oct 14, 2025 03:09 PMâ
How did you form that opinion so fast? Meaning we donât know anything about the project?
Ubuntu Touch is ugly, not really practice in my opinion. I donât know why Canonical does not concern itself with the project any further now. I know they are still helping the UBports foundation, but even so, they could do something much better if they took over the project entirely, at least in terms of design and UX. Linux is mature enough now for them to recommit to the mobile version.
But you have postmarketOS whoâs really better than UT and very pleasant.
the benefit of that project is for what Android users (and postmarketOS folk too) rely on vendor fixes for outside the kernel: bluetooth firmware, audio dsp, sensors - anything that rom builders need to source from the muppets or extract-files.sh/.py themselves if the device is around. That project looks into commonality and will then concentrate on a subset of blobs to re-implement. Letâs quote more of that FAQ:
We will work to analyze common nonfree blobs in current (mostly) free software operating systems. We will use this knowledge and other factors to identify the best device for development, as well as documenting how these proprietary blobs get used by the kernel Linux, in order to understand what it would take to legally reverse-engineer them using clean room techniques.
Besides the FAQ, the FSF has posted the project site with python utils that fill a device research database and an audio interview (skip 3:50 min intro) with Rob Savoye to understand more of the scope and phases of what they want to do.
2 Likes
Thatâs a prety big response/opinion for something that was just announced:
How did you form that opinion so fast? Meaning we donât know anything about the project?
tl;dr: Because Iâve kept track of the FSF for some time, and I read the press releaseâŚand while the project has potential to address barriers to adoption of FOSS mobile operating systems, there are a number of indicators that it will be focusing on principle rather than pragmatic roadblocks that many users have.
For starters, Iâll go back to the timelineâŚthe FSF was around when the Newton came out, and the Blackberry, and Windows Mobile, and the iPhone, and the HTC Dream and other Android phonesâŚand almost 20 years later, now theyâre starting to explore the idea of reverse-engineering some of LineageOSâs binary blobs?
Hereâs an article they wrote from 2022: Ethical Tech | Giving Guide that encourages people not to get iPhonesâŚbut their recommendation as to what to do instead gets murky really quicklyâŚthey say âIf youâre at all able to, and if you find your hardware is compatible with it, we recommend pairing the Replicant distribution of Android with the F-Droid app repositoryâ. They got real direct about avoiding iPhones, but got really shy about endorsing an alternative. Itâs just as well, because the newest phone Replicant supports is the Samsung Galaxy Note 2, which is 13 years old at this point.
In the same article, they encourage ditching Netflix andâŚthe best alternative they provide is Blenderâs showcase, which they do not financially support (nor F-Droid, for that matter, as best as I can tell) and seems to have their selection of films available solely in the browser. No app, no means of showing them on a TV, no âdownloadâ button so the films can be added to a streaming server or added to a USB drive and shown on a device capable of playing them.
The router they recommend - the TPE-R1300 - has 10/100 ethernet ports and 802.11n Wi-Fi. State of the art for 2006, outpaced by even the cheapest Wal-Mart routers in 2025.
And, lest I be accused of pointing to an old article, the very latest iteration makes the same recommendations Ethical Tech | Giving Guide .
Now, its list of high-priority projects seems to be a bit more of a mixed bag. It endorses Signal, which is pretty solid, but their list of Skype alternatives doesnât seem to have much traction, in no small part due to the fact that Skype/Zoom/Hangouts/Teams have a server-component that Google/Microsoft/Teams handles. A FOSS alternative is going to necessarily involve the user also handling the server component, which is a much, MUCH bigger ask of most users than an e-mail address and a password..and none of the projects offer a turnkey appliance. Other high-priority projects center around low-level firmware reverse-engineeringâŚwhich is great, but also device specific, and functionally a constantly-moving target.
Now, donât get me wrong - I very much understand that the FSF is an organization to whom I owe a great deal of gratitude. /e/OS, Debian (and its Proxmox derivative) wouldnât exist without them. The LAMP stack that runs the internet wouldnât be possible if there wasnât a counterbalance to the Microsofts and Autodesks of the world. The GCC compiler made possible more software development than I will ever comprehend. Theyâve been at the forefront of the right-to-repair movement, and were instrumental in bringing the issues to light in the first place. And, of course, Android wouldnât have an AOSP basis in the first place if it wasnât for the GPLâŚmeaning that /e/OS couldnât exist without them. There is good work being done by the organization and I would be remiss to spend the whole time focusing on my discrepancies and not acknowledging their positive contributions.
However, looking specifically at this project in isolation - replacing binary blobs with OSS replacements for Lineage (which is what the press release says theyâre doing) - it truly seems to me like theyâre chasing purity at the expense of pragmatism. If LineageOS is 80% FOSS, thatâs about 70% more Free than Samsungâs implementation of Android, and 79% more Free than iOSâŚbut the announcement states that the goal is to focus on components of LineageOS that will only matter to an exceedingly small subset of developers. From the primary developer on the project, âIâm looking forward to this opportunity to work towards a freedom-supporting phone and help users gain control over their phone hardware.ââŚthatâs great, but it raises the question regarding whether the intended outcome is best served by replacing working binaries, or instead reducing the friction in moving to the imperfect-but-far-more-freedom-respecting projects - like Lineage and iode and of course /e/OS.
So, yeah, I donât think I need to wait until the general release of this project to have concerns about what has already been said in conjunction with the group who is saying it.
Well said.
Every initiatieve towards a privacy respecting and transparent digital world is to be praised.
However when success is defined by a larger number of people adopting a safer, more respectful technology, it means embrassing promissing and commercially viable initiatives that are here already. Read: e/OS and some others.
When you see it as a staged play it also means, for now, cutting corners here and there to become commerially attractive to the âmassesâ. For example adding the voice to text feature.
For those who whish to stay purists, there will always be a little digital corner to be your ideological and technically correct self. Which in turn will be hopefully a boiling pot of ideas and contributions?
In that light I would rather have FSF building bridges instead of launching one more flavor into the arena.
1 Like
to me this is a loosely related rambling, inconsiderate in its length to the forum reader. Judge the project on its technical merits and the devs prior work, the FSF is just an umbrella.
The effort targets low level parts of handsets, not any user-facing parts - and enough info is relased by now to know better what the project is. Their discovery phase has concrete, work-in-progress artifacts that can be inspected (hashing and categorizing blobs across all lineage devices in search of commonality and what firmware to target).
Commonization - even if not reverse engineered - is already helpful for rom maintainers. If the project identifies firmware where re-implementation is worthwhile and feasible, devices using it will have a longer, secure lifetime.
A significant part of Androids security bulletins are about vendor fixes. I donât think the effort is able to replace millions-of-lines-of-code firmwares and write kernel drivers for it, but you got to start somewhere. The alternative is the position of the open hardware camp and be the OEM yourself, but then end-users complain about not getting a device with a modern and fast SoC.
2 Likes
This topic was automatically closed after 15 days. New replies are no longer allowed.