> Alright, so there is the LineageOS community, which provides weekly updates for an impressive number of smartphones. They provide a practical solution today, and I am very grateful for that. However, such Android based projects will always run behind Google and the phone industry, fixing only symptoms but never the root-cause.
I think it's very ambitious to not base this off of the work being done on Android but of course would love a proper GNU/Linux distro on my phone as well.
Lineage OS (formerly CyanogenMod) will run on a Samsung S II which is 5 years old. It might support some older devices that I am not aware of. I could see it still being supported 5 years down the line.
Yeah, what raised a red flag when reading this project's description was that having a single kernel for all devices and no compatibility layer like libhybris is unfortunately an unrealistic goal in the world of ARM android devices.
An approach that is less likely to result in vaporware would be to maintain ports of Halium to older devices based on LineageOS kernels/drivers, with an optional Alpine Linux userspace.
Otherwise this project is little more than README.md Hacker News bait.
If there are enough helping hands, we could do this as a community.
The bootstrap program, pmbootstrap, is a working base component that makes development much easier (alone the automatically set up cross-compiling with distcc, ccache, armhf and native chroot), so it is definitely more than a README.md.
Also the title says, that the project is "aiming" - which does not say it has reached any of its goals yet. But at some point it needs to be announced to the community, so I might as well do that now, at a point, where other hackers could join in and parallelize development.
The biggest issue in trying to have a long running constantly updating OS on a Android phone is the device drivers.
Instead of trying to reverse engineer and make open source drivers, I think you should try a more generic approach something like a rump kernel or some other generic layer for each Linux kernel version which can simulate the interface needed for the particular kernel driver. That way you don't have to keep porting or reverse engineering device drivers.
Your approach of trying to make an open source driver is a losing one, it requires a lot of work and you can't reasonably keep up with it.
> Of course, the big if there is: Can you reach the critical mass of drivers without the project exhausting itself.
If that were the case, we wouldn't find ourselves in the current quagmire that we're in. Linux kernel already had a huge number of devices supporting it.
The truth is, there is nothing stopping device vendors from releasing a device driver for a specific version, maintaining it for an year or two and then abandoning it. The better approach would be to figure out a stable interface layer for kernel drivers so that you can simply plug and play the vendor supplied device driver.
Could you help me understand your reasoning behind not pairing up with or taking advantage of existing projects like Halium or LineageOS?
I'd love to run mainline Linux without any Android cruft on my old phones, but realise that this would take a lot of effort per each device. Each Android device is different, with its own kernel patches, proprietary modules and sometimes Android-specific userspace HALs. The Tegra in the Nexus 7 was designed to run non-Android Linux, there are native X11 drivers for that thing - that's not the case for devices like my current phone.
> Could you help me understand your reasoning behind not pairing up with or taking advantage of existing projects like Halium or LineageOS?
From the article:
> Of course I am not the only one, that came to this conclusion - especially in the last few weeks with the Halium project rising (greetings!). I am all-in for working together — sharing udev rules, merging Android kernels together, whatever makes sense!
> [...] postmarketOS does not fit the Halium model, as it avoids the Android build system entirely and does not run Android next to GNU/Linux.
> Thanks to Replicant, LineageOS, Halium. Together, we can make the vision of long-lasting, open source smartphone operating systems a reality!
So I see postmarketOS as part of the community. We have multiple projects, that want to provide a more open alternative to Android, on Android devices. This will only work out, if we work together and share, what we can.
postmarketOS differs from the other projects, that it tries to completely cut-off the Android parts. This will be harder for hardware compatibility, but then again, it does not depend on Google's upstream Android code (which does not get developed as a true community project, they only put out the source after they are done developing behind close curtains), and it uses less resources than running Android as second OS next to your regular GNU/Linux.
Regarding kernels and drivers, postmarketOS directly packages LineageOS kernels. Proprietary 3d acceleration will be avoided for now, some tests with "weston-smoke" and other demos showed, that it works fast enough without them. You won't be able to play 3d games, but that's not really in the scope of the project for now.
Very few mobile phone arm devices support device trees. Only Windows Mobile phones provided UEFI (with locked bootloaders). Most ARM devices just attach random shit to random pins and have totally non-upstreamable kernels.
For the most common apps, no, it's not slower in my experience on Nexus 5, 7, 10 and some Samsung tablets.
For older devices, there are performance improvements in the ahead-of-time compilation with upgrades in major versions of the OS.
That helps. Lack of multimedia hardware support for older devices in newer apps can still be an issue.
Running Lineage OS 13 on a nexus 10, I see great performance for all the default apps and google apps. The main source of performance issues I see is apps that do not support the multimedia hardware (eg. arm neon on Nexus 10). Some video apps such as Directv now, HBO GO don't use neon, overheat with high cpu use as a result. Those that do support neon work great, including Netflix and Hulu. So it also depends on app support for hardware.
I don't know if you count it as an old phone but I use it on a Samsung SIII (Wikipedia says that launched 2012) and it's snappy as heck.
The current images are nightlies so I have encountered a few bugs. Previously was running CyanogenMod 13 on a stable image and that was fine (but the CyanogenMod company folded so it wasn't going to get security updates). Looking forward to a stable release of Lineage.
~I don't think Lineage OS 13 exists.~ (edit: it seems it does for some models). CyanogenMod 13 was Android 6 I believe and Lineage 14 nightlies are currently 7.1.2.
It exists. But I guess I get why you might be thinking this: with CM13 there were nightly builds for lots and lots of devices, but with LOS13 they heavily reduced the amount of devices for which they build nightlies.
I'm not sure if this means that they ceased to support LOS13 for these devices entirely. But it does seem that way, on account of how long it's been since the device tree ('DT') and kernel for my phone have seen any update. (HTC One M9, a.k.a. himaul)
Nowadays I just build LOS13 myself every once in a while, because it still receives generic updates and fixes, and luckily the current kernel and DT still work fine. It's just that, while the DT also wasn't updated that often back on CM13, iirc the kernel did receive updates and fixes fairly regularly. With that no longer being the case is why I suspect they halted supporting it at all.
My reason for staying on LOS13 rather than to upgrade to LOS14 is that the Xposed Framework does not exist for LOS14 / Android N. So that's what I'm waiting for. I wish the LOS13 would recognize how many people want to use 13 on the M9 and provide updates even if it's just once a month. Or maybe they do recognise this, but they just don't have the manpower, resources, or heck maybe there were too many problems when users switched from CM to LOS... I don't know.
There is just one really annoying error which I don't know how to fix, which is that the default camera app crashes, I suspect because of face detection. I thought they fixed that on CM13 ages ago. I guess he bug snuck back in. The OpenCamera does not crash, thank heavens, but it's still a nasty issue I wish would be addressed.
Anyway. My point was that it's still possible to build LOS13 for many devices.
If anything newer android versions seem to get faster and use less battery on my nexus 4. I'm trying to avoid resource hungry apps like spotify, google play and Facebook though so ymmv.
Why is "phone interface" buried down at the end of the page under "future goals"?
I mean, if a project wanted to put a "10 year life-cycle" on an IoT blender by porting Gnu/Linux to it, you'd think the first order of business would be writing a Gnu/Linux program to, you know, blend things. Getting 10+ years of running performant Weston demos would be a very distant 2nd to that.
> Why is "phone interface" buried down at the end of the page under "future goals"?
Well, right now there is no phone interface available yet, and that is why it is a future goal (which does not necessarily mean, that it is in distant future).
> I mean, if a project wanted to put a "10 year life-cycle" on an IoT blender by porting Gnu/Linux to it, you'd think the first order of business would be writing a Gnu/Linux program to, you know, blend things. Getting 10+ years of running performant Weston demos would be a very distant 2nd to that.
I'm not selling you a blender here. I wanted to create a solid base to run GNU/Linux natively on smartphones, and now there is this pmbootstrap component, which works quite well and seems worthy of a release. This enables other people to jump in and contribute, if they want to, and accelerate development even more, because multiple persons can work in parallel on the project.
This is much more important to me, than silently working a few months or however long it would take me in my free time to package Plasma Mobile and make a release after that, just to hear that another crucial component is missing from the OS.
> This is much more important to me, than silently working a few months or however long it would take me in my free time to package Plasma Mobile and make a release after that, just to hear that another crucial component is missing from the OS.
So, imagine sharing a video that shows the fruits of your labor in order to drum up some interest in your project. Strauss' "Thus Spake Zarathustra" is playing in the background.
The video shows you booting the phone into Gnu/Linux. As the music builds, you touch the screen of the phone to make it start to do a task. As you enter data into the phone to complete this task, the music builds toward a climax. Finally, at the climax of the music the task on the phone is complete, and the phone reacts in the most dramatic way a phone possibly can.
What do you imagine the phone doing in such a video?
Read his comment again, you are missing the point. He is not offering you a phone, he is offering lower level components and showing you how it can be used to build a phone. Feel free to jump in and help him!
Congrats to OP, hope this takes off in a major way! We could use a free (as in freedom) OS for phones!
Are you deliberately trying to discourage this person to the point where they give up and quit? The negativity throughout this thread is making me want to just leave HN.
While I dont agree with the tone of the post above, I agree with the content. While its a noble thing to try to build a new phone OS, I dont see the slightest chance that it will take off. Look at others who tried and failed:
- Ubuntu trying basically the same with an estabilished brand, mobile experts, a huge community,..
- Mozilla. Here same as Ubuntu
- Microsoft. Yeah, they were closed source, but had almost endless $$$ to throw at their project. Heck they even
bought one of the biggest mobile brands (Nokia) and failed.
"I'm doing a (free) operating system (just a hobby, won't be big and professional like gnu)..."
Now Linux is the most widely used OS in the world, and the "big and professional" GNU still doesn't have a production-ready kernel release.
We need this project, or something like it. We need a Free OS for phones. I actually find the HN negativity on display here encouraging; it worked for Dropbox.
That's is what I think is one of the huge problems with GNU/Linux and OSS projects, design being an afterthought rather than one of the core features.
I mean there is a ton of linux distributions that run better on my PC, but I still rather Hackintosh it due to the linux distros interface design and I bet others do the same.
I know this is a new project, but in order to ever hit the mainstream, it needs to have a interface at least as good stock android/TouchWiz. Otherwise people will be driven away, I ask people about Ubuntu and they won't ever say "the design is bad". But they said they switched back to Windows because "they couldn't get a hang of it", because the interface made them feel confused.
I agree with the "interface at least as good as stock Android" part. I wouldn't even mind if someone re-created the Android UI with native GNU/Linux libraries instead of the ones from Android.
As long as the user choose the interface, this would surely make it easier for people to switch from Android to postmarketOS and users who'd like to use something different could freely do so.
Wait, I did not say that it needs to be compatible with anything. Just recreating the top bar, that you can drag down to connect to your Wi-Fi etc. and the three buttons at the bottom working in a similar way would make it very easy to migrate without any compatibility requirements.
But as ocdtrekkie has noted:
> At the point where the phone OS can't make calls yet, worrying about style is a bit premature.
This project looks awesome btw and keep up the work. When I read through it, I saw it more as just running plain Linux on your old Android devices; turning them into what you would normally use a Pi/Beagle Board for. I'd love to be able to turn old Android phones into small Linux machines, and not have all the Android rubbish.
For me personally, I want it to do what I need my phone to do, privately. I don't care if the entire UI is black and white and the only font is Courier New if it does that.
Design is second to having a rock solid platform to build on top of.
At the point where the phone OS can't make calls yet, worrying about style is a bit premature.
I know we're still calling them "phones", but I make and answer voice calls with mine far less frequently than I do with my desktop computer.
It really is just a pocket-sized computing device and I use it almost exclusively for its web browser and internet connection. So this didn't strike me as even a little bit odd.
There is one small fallacy in the decade old PC versus the decade old smarphone comparison: PCs are not status symbols but smartphones are and fashion is a huge component in why smartphones are replaced.
Another driving force is that a smartphone made a decade ago would simply not be able to use many of todays apps because it is missing certain sensors.
so while I hope this will take off I see some obstacles. Happy owner of a 10 year old Nokia here that serves me well, I've tried quite a few smartphones over the years but I never found anything that I really needed that would make me give up 5 day battery life and being 100% drop proof.
I think there is a community that would like this. Not sure how big it is.
I drive old cars (newest is a 2004 year) and use old computers (my laptop is a 2009 Macbook Air) because they are a fantastic value and they do what I need.
I use a phone for making calls, texting, light web browsing, as a portable music player, and for maps/navigation. That's really about it. Maybe a 10yo phone would struggle with maps but a 5 year old phone generally would not.
> There is one small fallacy in the decade old PC versus the decade old smarphone comparison: PCs are not status symbols but smartphones are and fashion is a huge component in why smartphones are replaced.
Think about Apple's Macbooks, I'm pretty sure that they are status symbols as well.
People who buy devices as status symbols, to show off how much money they have, are not the target audience of this project. They usually do not care about the benefits of open source and sustainability anyway. This project is about recycling old phones, which would be insecure/useless without an alternative operating system.
> Another driving force is that a smartphone made a decade ago would simply not be able to use many of todays apps because it is missing certain sensors.
That is right, but it is theoretically possible to get the devices from a few years ago running for five more years, and the devices created today for the next ten years.
Macbooks will already run Linux, sometimes with some difficulty but usually not much more than a 'how-to' or a readme is required to get most stuff working. Brand new laptops of any brand find themselves stripped of their default install and turned into Linux machines.
Of course there's a market for it. Do you not have at least a few friends who are holding out with their candybar dumbphones or antique iPhones to this day, citing reasons that include but are not limited to 1) I'm not wasting money on a new phone that will have to be replaced every year 2) I hate how much trash electronics generate 3) I want to be more engaged with the people around me so I want my phone to be less capable/less distracting 4) I don't like facebook and the NSA spying on me and I won't get a smartphone unless I'm sure they can't.
I have a good handful of such people in my social circle, ranging in age from 23 to 58, usually higher education levels, and usually non-technical. The biggest common factor amongst them is that Android doesn't work for them.
It exists. Desktop Linux exists, works well and maybe has some single digit market share but in a market basically consisting of everybody and their dog this is still an impressive number. It may not be enough for global megacorps but enough to be sustainable.
The most successful business created industries for themselves. In ollieparanoid's case, he's creating a secure and supported OS that runs on any phone, and when you consider that 90% of phones are missing a secure and supported OS, then that market is pretty huge. If this works then his OS dominates any old, unsupported phone.
People don't care about open source and sustainability but they care a lot about not being hacked and having a phone that runs, and runs fast.
>People don't care about open source and sustainability but they care a lot about not being hacked and having a phone that runs, and runs fast.
You overestimate non-technical people. They're not afraid of getting hacked, they don't understand why they should be, and they don't get why unmaintained software is bad.
People don't give a shit to the extent that the device works well. Security is of course a huge part of that but you don't understand people well enough if you think there's a significant group that will sacrifice good UX and reliability for security and rapid updates. That's what nerds like us do.
Not so sure about phones being status symbols. Maybe in some circles, but i have trouble telling brands, models and makes. Smartphones look very alike, small black, no buttons, the same layout.
It's easier to spot the difference when somebody is actively using their device. But its not obvious. I think that vast majority of the population have no clue which model is "the hottest". Maybe geeks, maybe Apple fans.
To me it looks as we all have the same phones, using them the same ways.
I agree that trying to use a 2007 smartphone wouldn't be a pleasant experience. However I think that running a 2017 phone in 2027 might not be that far fetched. I can't really tell a difference between my previous phone and my current one. I only got a replacement because the old one broke down. The pace of mobile hardware development has slowed down and phones two years from now won't be significantly better than the current generation.
I don't think that smartphones are a status symbol any more. Everyone has got one and they're all pretty similar. It's just a commodity these days.
The planned obsolescence cycle of replacing your phone every two years can't die quickly enough. We need serviceable hardware and long term software security updates and we should have had that a long time ago. The current situation is not sustainable.
> PCs are not status symbols but smartphones are and fashion is a huge component in why smartphones are replaced.
For those who buy the newest iPhone or Samsung Galaxy every year this might be the case but the rest of the world either doesn't care and see it as a tool or simply can not afford a shiny new flagship device and may be grateful if their existing device lasts longer.
> Another driving force is that a smartphone made a decade ago would simply not be able to use many of todays apps because it is missing certain sensors.
I don't see substantial innovation in the hardware space anymore besides bigger numbers on the spec sheet. Chances are high that phones from today will crack the 10 year mark of usefulness. We are already at a point where 4-5 year old high end devices are still perfectly usable for most daily tasks.
Yes, with the deceleration of Moore's law, small players probably have a card to play. When the spec sheet doubles every 6-18 months, only the big players can follow and update their product lines. Small businesses and startups now have access to hardware that is good enough for viable products.
The other factor is if the phone doesn't have a user-replaceable battery then it has a short self-life regardless of the software situation. Batteries don't last.
Maybe Linux phones have that problem. Apple has battery conditioning drivers on MacOS and iOS to ensure batteries last super long. I'm almost 3 years into my my 6 plus end battery life has not degraded substantially. And replacing it is pretty cheap.
Apple is affected. It depends on how you use your device, you can easily kill you iphone battery (as well as any other phone) in two years from normal use. It's one of the main complaints I hear from iphone folks.
That is why "lasts whole day" isn't good enough for a new phone. In a year it will not last a whole day. And it will seriously hurt the longevity of it, but noone cares.
They don't care because Apples batteries degrade much more slowly than others because of that conditioning software. And eventually you always have to replace a battery. Removable makes that a little easier, and a little cheaper, but the tradeoffs (lower capacity, more frequent replacement, worse build quality and packaging) aren't worth it for something an iPhine only need to do every 3 years in normal use.
Anecdote: I'm getting ~9 days to a charge with my Nexus 5. Battery life improved after I moved from CM to LineageOS. Max was about 7 days on CM. I know batteries don't last forever, but the hardware is still perfectly good. Google doesn't care about Nexus 5 owners anymore. It's sad really. They should at least open source the drivers if they don't want to support it anymore. Even if I purchase a new phone, this one could still make someone happy for many more years. It doesn't belong on a rubbish heap.
Is that 9 days idling? That's fantastic for a nexus 5! Mine is running stock Android, it's over 2 years old, and the battery lasts maybe 16 hours with normal daily usage. It's still a very capable phone, but it's a shame it's becoming unsupported and slowing down as makes their apps more sluggish. I switched to an iPhone SE recently, Google's apps run much better on iOS in my opinion.
In most cases, a more powerful inexpensive phone would cost less than trying to replace the battery. Heck, if you value your time a new phone is probably less expensive that trying to salvage an old one.
They also told me on reddit, that it is probably not such a good idea to buy a cheap battery for a device, that has a battery that is not meant to be removed, because the resulting cheap quality may make your phone go up in flames.
Anyway, there are still so many devices where you can remove the battery out there.
And not everyone has money as highest priority, so I'd rather take my time and buy a new battery for an old phone than throwing it away and creating unnecessary electronic waste, just to save a few bucks.
Plus when you buy a new device, you will need some time to transfer your data, and set it up again, possibly experience new bugs and what not. So that also costs some time.
I agree to many their phone seems to be a status symbol and a big part of their connected identity. More than a couple times I have been made fun of for my "ancient" 5 year old smart phone, but I only use it for basic texting and email, so why bother upgrading when it still works fine?
I think brand new shiny mac laptops are status symbols these days (says the guy who lugs around an 8 pound GPU laptop to do machine learning on it, oh excuse me, I mean "Edge Computing").
I hope this takes off. I was happily using a note 4 for years, occasionally buying a new battery because it allowed you to change it by removing the back (remember all the phones that used to do that?). The only reason I had to retire it was because I needed security patches but each update made the phone progressively slower. My new phone has exactly the same features and if I factory restored the note 4 it would run as fast as my new one but phone manufacturers don't exactly like that, and it's something that should change.
This is fabulous! There are millions of old phones out there with decaying versions of iOS and Android. Google and Apple are ignoring them, so these devices are just plummeting in value. This is the perfect market for an actually-open-source mobile OS!
If somebody can get React Native to run on PostmarketOS, then we could start building up an ecosystem of non-walled-garden apps that are compatible with iOS and Android!
> If somebody can get React Native to run on PostmarketOS, then we could start building up an ecosystem of non-walled-garden apps that are compatible with iOS and Android!
What leads you to believe this would actually happen? PhoneGap/Cordova have been around for ages, and they could have already enabled such an ecosystem to have formed, but we haven't seen that happen. Of course, there are other similar technologies out there, too. Even Firefox OS failed, and that was with the backing of Mozilla and its eager community. What's different about this case?
Good question. I think there are two big reasons this is more feasible now than it has been in the past.
1) Proliferation of compatible devices. Firefox OS attempted to launch with zero compatible phones. (The dev device I bought would crash on boot, every time). Postmarket OS is launching with hundreds of thousands of compatible devices already in the wild, many of which are sitting unused in people’s drawers. If it gets much contributor momentum, there will be millions of devices ready to switch.
2) As a React Native contributor, I am totally biased, but I think it represents a big shift in mobile development. Personally I don’t think ionic/cordova apps feel very good, especially on older devices. RN is the first open-source xplat mobile app environment that allows you to produce high-quality apps. Or at least, by the standards of FB, Airbnb, and Microsoft.
It would require react-native because then phones that are not running gnu/linux can also use those react-native apps. So, if you build an app with react-native, then it runs on Postmarket OS and the commercial phone OSes. Perhaps I'm missing the reason why another tool is just as cross-platform?
EDIT: Looks like there are plenty of options for cross-platform mobile UI libraries including QT and Kirigami. Like others have said, why use React then.
For longest time I used Samsung Galaxy with CyanogenMod, while it was usable. It was pretty much a stretch. I'm also guessing using that old ass phone for so long is one of the huge factors why smartphone isn't a big part of my life (as it seems to be for most of people of my age). Since my phone was so slow and barely worked I always opted for other means of communication and face-to-face socialising.
> postmarketOS is a GNU/Linux distribution, so there's no problem in having multiple phone interfaces (just like KDE/Gnome/XFCE/...) and let the user choose.
IMO, there is a big problem with that. The WHOLE reason the smartphone space finally took off was because Apple came in and rationalized the interaction model and implemented a robust, responsive UI. Android is successful largely to the extent it faithfully aped that model. Treating this as a secondary thing your project doesn't need to be opinionated on is a recipe for failure.
But the whole point of this project is to appeal to the mass market, and give smartphones a ten-year life cycle. They're explicitly aiming at a much wider userbase than hackers alone.
No, the whole point of this project is to be able to use your 10 year old phone for other things, like a mini-computer, or a Pi.
You run a newer build of Lineage or CarbonRom or whatever on an old phone and it will most likely run like crap (current LineageOS on an old Sony Xperia Z 1st gen for example, is a terrible experience).
But that phone still has a ton of power. It's more than capable of being better than an early 2000s laptop. If you can stick Linux on it, it can be an embedded device. Get Wayland working and you get a little screen you can use for projects.
This has the potential to turn a lot of obsolete phones into powerful little hobby machines.
Do you think that hackers as a market aren't, by near definition, early adopters and want new technology? I don't know who else a 10 year old phone would appeal to but parts of the mass market?
(Some) Hackers are also the kind of people who keep their hardware around for ages and run strange OSes if it fits their needs, because they want daily tools to work, reliably and like they always did. Tinkering can happen on top of that or with less important things.
Yup. I'm running a four year old phone because it does what I want and newer phones don't. Money isn't even an issue, there just isn't choice out there today.
About the question whether Apple has invented a better UI and whether someone else did or could have done the same:
Regardless what you think about UI-paradigms or vendor lock-in, there is something that Apple spends far more time on than others to get right:
Papercuts [0]. Every interface comes with a thousand little annoyances that distract and increase cognitive load. Apple tries to bring this number to zero.
You have to find managers and developers willing to reserve a LOT of time for these "trivialities", but ultimately that summed-up effort is what makes their computers so convenient to use.
> The WHOLE reason the smartphone space finally took off was because Apple came in and rationalized the interaction model and implemented a robust, responsive UI.
I see this "fact" thrown about all the time by under-30s who only ever knew of the iPhone since they were teens.
The iPhone OS interface, as it first existed, faithfully aped the Palm Garnet OS, which was the latest iteration of Palm's highly successful OS going all the way back to the late 90s. Before there were smartphones there were PDAs, and while it's debatable that Apple invented the PDA with the Newton, the platform didn't become a commercial success until US Robotics released the PalmPilot and PalmPilot Professional. When Handspring successfully melded the Palm OS with cellphone hardware creating the Treo, the touch controlled, icon based, modal fullscreen app "smartphone interface" as we know it was created. Palm bought Handspring and pivoted to improving the Treo line of smartphones, culminating in the Treo 600 and 700 series.
Speaking of responsiveness, while Palm's own devices were fairly standard and could be sluggish, Sony's Clie line made Palm OS feel almost as fast and fluid as the modern iOS interface. The hardware was expensive but it was the absolute best handheld computer you could buy in the early to mid 2000s. The first time I used the original iPhone I still owned a Clie NX70V, and the Clie was less jittery than the iPhone if I had a lot of apps open on both devices.
That's not even touching on Nokia's Symbian, arguably the best selling smartphone OS in history, but Symbian never resembled Palm OS, Android, or iPhone OS until after those systems became widespread.
So no, Apple never invented anything in this space; they took what already existed, polished the hardware, correctly implemented multitouch, and applied Jobs' marketing magic. I drank the Kool-aid and bought the original iPhone soon after its release. I ended up promptly selling it and going back to my Treo to get real work done when I realized the iPhone was nothing more than a (arguably excellent) mobile web browser. It wasn't until Apple finally started allowing apps (something Palm supported from the beginning) that the "modern" smartphone revolution really began. Even then, hardware improvements were driven by Android based devices, with Apple always playing catch-up.
Do you think that both interfaces are the same simply because they are "touch controlled" and "icon based"?
Also, saying there were other worse phones before the iPhone came out does not refute the point that the reason the space "finally took off" was because of Apple.
I feel its my duty to report that after looking at that image of Palm Garnet OS it shows an app called "DiddleBug", which is apparently a note taking app.
I am certainly glad Steve Jobs applied his sinister marketing magic to whatever this is so that no one has to be inserting IntelliBoogers into their DiddleBugs on their PalmPilots just to take a note.
Yes, things like these remind me to take our Product and Marketing guys more seriously, because, left alone in a cellar, I too probably would call things DiddleBugs and IntelliBoogers.
You can of course take notes out of the box with a more informatively named program. I'm sure there are apps with weird names in the iPhone app store as well.
It was single threaded with no memory protection either, so you could write into other programs memory space and easily crash your device. :)
I still loved PalmOS. I had two Visors, a Treo, a Centro and even a Palm Pre with WebOS. HP hasn't done anything with all their Palm IP for a while .. sad to see it totally gone now.
That's still plenty to spot major differences between the stylus-based PalmOS paradigm and modern touch-based smartphone interaction. There's a drop down menu on that PalmOS home screen, and the app launcher has to be scrolled by interacting with the scrollbar on the edge of the screen (or using the hardware scroll buttons). Looking at a screenshot of one of the applications would show even bigger differences. Really, the similarities don't go much further than a tile-based launcher and a status bar at the top of the screen. The PalmOS UI had more in common with the Newton UI than iOS.
I was under impression that Apple regularly beats Android with hardware. CPU, GPU, screen, touch hardware (not sure lately for these two) and others I'm sure.
Well, like any other group there are always outliers. :-)
I've heard many strange things from folks born in 1990 and later; I've actually been told by someone with a straight face that Apple invented the MP3 player as well as the first touchscreen.
Well that [mp3] is because they don't know much about anything techology wise :) not that many 90s kids remeber midi or MD players despite they were quite popular (lots of geocities sites would have a midi music playing in the background for example.) but I get your point, it is fair to say most people can hardly remember much anything they generally don't care.
The ability to choose between multiple desktops does not mean, that each desktop itself is inconsistent.
For example, you have the consistent Apple interface. And you have the consistent Android interface. If the user could choose between both on the same device, why would that be a problem?
Focus. Apple and Google have hundreds or thousands of engineers working on these operating systems, full time, for years. To get this project to even 1/10th of the quality would be a massive undertaking. Open source mobile OSes do not have a great track record right now. It'd be amazing to have a real option with a solid development and user base behind it.
Yep. I have not fully decided yet on which part I will work next (that sort of depends on the feedback by the community). But I definitely want a solid base.
>The WHOLE reason the smartphone space finally took off was because Apple came in and rationalized the interaction model and implemented a robust, responsive UI
I don't think that's the whole truth. Apple became successful because efficient and powerful mobile hardware arrived to support Apple.
Just because one paradigm worked, should we stop and treat it as gospel. Perhaps there might be others waiting to be discovered.
Apple waited to enter the market until it could practicably make a mobile device with a web browser that didn't require mobile versions of web sites. (Sure, mobile optimized is better, but the iPhone could at least load and zoom in on regular sites.)
Whether this was entirely intentional or not, I can't say for sure, but they certainly emphasized it at launch, and they designed a device with a screen that was huge enough, and it worked out great. Turns out the web was the killer app for the internet, and it was the killer app for smartphones too.
99% of the time when it's not a game a curses GUI or a readline UI is just fine.
Really, I'm tired of all this UI "inovation." just give me a TTY already, it works fine on my laptop and it would work fine on my phone if everything weren't all tied up in closed blobs.
This only works if other people can use KDE if they want all the flashy stuff.
My phone is basically just Termux at this point. I still use the gui for making and receiving calls and for web browsing but for all other purposes it is just a portable TTY.
lmao is this satire? how would you accurately type commands with such a tiny keyboard without tactile feedback? and how would that be faster than tapping on an icon?
Touch keyboards are quite good? I don't see what's even surprising about this. Yes, "tapping on an icon" is faster when the icon exists, but Android/iOS does not favor doing things nearly as powerful as a full Unix. Dumb example: file transfer. I have photos on my phone and don't want to go through a cloud service just to get them onto my laptop. I have yet to find an app that even comes close to rsync over SSH. Could that app exist? Sure. But until somebody writes it I'm going to keep using the last 20 years worth of good tools.
[CLARIFICATION EDIT: this places a FaceTime Audio call. The number you're calling would need to be a contact who can accept that call. Not exactly general-purpose.]
Hear, hear. I would add that Android has suffered to the extent it enabled OEMs to make choices and modify Android. There are plenty of places, like camera software, where OEMs should go crazy trying to make differentiation in their products. Core UI isn't one of them.
This is awesome and I've been waiting for something like this to come along for a long time. It's crazy how easy it is to get Linux on an ARM based Chromebook, but nearly impossible to get it on a phone.
Android is a security nightmare and most of us are aware of that at this point. On top of that, Android has been moving functionality off the device and into their services for years making their AOSP offering weaker and weaker. Keep up the good work!
Edit: At some point it'd be nice to use a GuixSD or NixOS configuration file as your "one custom package" instead of an Alpine package. Any Linux on bare metal though would be welcome of course.
No, real security nightmares are reserved for things like WannaCry and all of the compromised Linux based IoT devices. You know, things that actually do real world damage at massive scale. As for the state of Android security, unfortunately the analysis and numbers don't really back up your baseless claims.
OK, there are millions of Android users and most users put their personal information on their phone. I think any breach here would be considered gravely significant.
Non-Google branded phones get updates far later than their Nexus/Pixel counterparts if at all. Google and carriers drop devices from updates eventually (2ish years) so if you continue to use your device after that, you're playing with fire.
That's just covering the delivery mechanism, not the inability to set a encryption password separate from your pin, or the vauge permissions groups, or the fact that all apps can see your global clipboard, etc. If users can't get the newest version of their OS software, then you kind of fail at security 101. Any bug fix only fixes a small subset of your users. At least Windows XP device users knew how long they'd get security updates and had a clear upgrade path afterward.
Linux IoT suffers a lot of the the same problems. No one updates embedded devices.
You can draw the "nightmare" semantic line wherever you want, of course. IMHO it's not even remotely secure unless you only and always buy the newest Pixel phone directly from Google. Then we can talk about modern Android security issues.
There are over 2 Billion Android devices that access the Google Play Store each month alone - and I'm not even factoring in all of the Android based devices that don't even have Google Play Services like China. Where are all of these nightmares? When you have the #1 OS in the world, by a large margin, every nefarious organization is targeting you. And how much damage have they inflicted on all of those devices compared to the damage inflicted by Windows and Linux IoT devices?
Incidentally, Apple just released 40 security updates recently. 21% of iOS devices aren't even using iOS 10. Is there going to be a security nightmare for them too?
I don't like Android. I prefer iOS because there is a single vendor for managing the sales and the development of both the hardware snd the software. This pro does come at a premium cost but I don't want my phone to stay on Android 2.X because AT&T does not want to release upgrade, then the only option is only going to be flashing and installing a new version on my own. That wipes my data and I probably will break my warranty worh ATT by flashing (I never checked).
iOS has bugs and performance issues too. But overall it is quite solid. The app store experience as a consumer is nice too.
If this project is aimed to bring freedom software to users on mobile with thr goal of running on phones for 10-years, great. But for mass adoption I am going to be bold: it won't happen. Firefox OS failed because (1) device spec is poor (Mozilla wanted to market it in countries where smartphone was not common), (2) not enough apps, (3) the influence and trending look of Android and iPhone.
> I prefer iOS because there is a single vendor for managing the sales and the development of both the hardware snd the software
If that's really your only reason for preferring iOS, Google has been making phones for over 7 years. There are obviously valid reasons for that preference, but the single one you've mentioned isn't.
My chief problem with both iOS and Android is far less the OS versions and far more the app update treadmill of despair that has eventually effectively bricked every single smartphone and tablet I've owned.
For if one opts out of updating those apps, either future versions require an OS update, they stop communicating with underlying cloud services and break, or they require current HW not to make the phone unusable.
If this trend continues into IoT, IMO it will be a dystopic fail whale. At some point these gadgets become appliances (with apologies to Cory Doctorow) wherein IMO their behavior should be well-defined and consistent over time.
Example: do you really want a Smart TV that has to boot and which by default communicates analytics about you into the cloud? I don't. But maybe I'm just an old fogie who remembers how awesome it was when TVs starting turning on immediately instead of having to warm up the picture tube.
The carriers do in fact have a say in the upgrade schedule. I know that (at least for a little while) they stopped being able to release updates quite as quickly once they released the iPhone for Verizon in the US.
And yes, if you buy your phone directly from Google, you'll get your updates directly from Google with no delay (other than their staged rollout process).
> then the only option is only going to be flashing and installing a new version on my own. That wipes my data
Make a backup. When you restore your phone, you also loose the data.
> and I probably will break my warranty worh ATT by flashing (I never checked).
You don't. First of all, you don't have any warranty in regards to your phone with AT&T. Second of all, you can always flash back your original firmware. Third, there are exceptions to allow these kinds of things.
> (3) the influence and trending look of Android and iPhone
Are you talking physical or on the software side? If it's the software side, it's just software and it can improve. iOS and android both have.
Most Android vendors now test and release android versions within 4-8 months of a new Android version and generally commit to 2 android upgrades for any given model.
This seems stupid. Android is hard to update because of closed source drivers - especially really proprietary stuff like Bluetooth, the RIL and graphics. That combined with the lack of a stable Linux driver interface means that only the manufacturers can really update the kernel.
I'm guessing this "fixes" that by just disabling everything.
As it says in the article, the project is no-where complete in a sense where you can just use it as your day to day smartphone OS, so it does not "fix" this yet. The project has just started, but there is a solid base with "pmbootstrap" now.
The Linux community is already past the problems of proprietary drivers. We are talking about people that reverse engineered a full network stack, graphic cards, audio devices, you name it.
This statement is horribly wrong. The sad state of even wireless ac drivers in Linux today show that there is a severe lack of driver development. What makes you think that a more proprietary and more guarded cellular modem will have better driver support than the current state of driver support? Isn't this just wishful thinking?
This is true for a subset of desktop/laptop/server hardware, but is nowhere near correct for other Linux devices like smartphones, routers, or SBCs.
And for the most part the mature Linux Desktop/Server ecosystem isnt full of reversed engineered drivers, its full of open source drivers contributed by manufacturers. My Linux Laptop works well without binary drivers (a firmware blob or two excepted) because Intel spends the time and money to employ people to write drivers for their GPUs, WiFi, etc.
It's even worse than that. The device manufacturers often can't update the kernel on their own; they need cooperation from their peripheral vendors. I see a lot of Android customers criticizing device vendors for fragmentation and failing to release OS updates. But Google really created the problem by failing to build in stable APIs for a hardware abstraction layer and loadable device drivers. Unfortunately the only ways they could really fix the problem would be by forking Linux or building a new kernel from scratch and those would be huge efforts.
> Unfortunately the only ways they could really fix the problem would be by forking Linux or building a new kernel from scratch and those would be huge efforts.
Or they could make their user mode pieces work with multiple kernel versions. In, say, a desktop Linux distro, it's often possible to leave the kernel at one version and upgrade other components.
I didn't say that. But the kernel (and by that I mean ring 0 only) is a small percentage of android compared to user mode code. Privilege escalations in the kernel are more rare than other issues.
So saying you cannot patch user-mode binaries (again, a majority of code in the system) because the kernel cannot be patched sounds an awful lot like throwing out the baby with the bathwater. And on Linux, the user to kernel ABI is fairly stable (see Linus's rants on the subject), which makes patching these pieces independently much more feasible.
Put another way, if someone says a vulnerability in libstagefright.so can't be patched because kernel modules, that person is lazy and making excuses, or doesn't know what a kernel is.
One thing though: it doesn't state if it's possible to make/receive calls/sms using a phone. You know, I'd appreciate my phone to be able to do such things.
EDIT: SMS also do not work. In other words, what works is the touchscreen with Weston so far. But the underlying tooling (pmbootstrap) is pretty far, so that's why I have released it now.
Good idea, I hope it gets developed much further. To cover the lack of apps you should concentrate on providing native internet browser with good PWA (progressive web apps) support. That way only a few native apps will be needed by user, e.g. dialer, SMS messenger.
In reaction to people who are confused, that this is not a finished product, I have added a note saying "(Not usable for most people yet!)" at the beginning. The summary at the beginning did not make that clear before.
How does "Project Treble" affect this "issue"? The extra layer of APIs that Android will be built on should solve the "Fork for Every Device and Every Build" problem, yes?
As I see it, project treble will improve the update process, but not completely resolve the endless forking. You will still have the "vendor implementation", which is different for every device. It will contain at least the kernel and drivers (kernel/userspace), for which you, as a user, depend on the manufacturer to keep it updated, and which will probably not be mainlined. This means, that after two years or so, when the support runs out, you will still have a device, that does not get updates anymore. Only for a smaller component of the operating system.
Also keep in mind, that this will work for newer Android O phones only. We already have more than a billion phones (the number could be much higher, I did not google it right now), which will never get that improvement. They can be saved from being electronic waste with projects like postmarketOS.
I was thinking the same thing, Treble sound like the perfect solution for the kind of problem this project aims to fix, but I realised that the scope of this project is different. There's no way they can back-port Treble to older devices, and there's no pressure from consumers or manufacturers to do so. Project Treble aims to fix the fragmentation problem, but it does from this point on, older devices are not included.
There's also the fact that Android will keep on rolling, new features will be added and manufacturers will keep on using them as they see fit, so Treble can help users with a security update or two while that version of Android is supported, even if the manufacturer stops supporting their "vendor layer", but what will happen when Google drops support of Android O?
All the new versions of Android are very much similar to an average user and my gf is always surprised when I'm so happy one of my devices is getting an update because getting "more fine grained permissions" or "customizable quick settings" are things about which an average user doesn't ever think about anyways.
It's mostly for us power users and people who just want to show off how they got something only cool/rich/smart people have. There were big differences between older versions of Android but apart from split screen which I used 3 times so far I don't see a reason to update.
I'm pretty sure the reason they don't port back the features to older androids is either because the old devices are lacking the specs or because they only want to have something to show as "new".
I know my friends avoided updating their iPhones for years because the new OSs only slowed the devices down.
There are enough people around the world, who can not even afford a smartphone but would like to have one if they could.
It had Wi-Fi already, you could write messages make phone calls, play some casual games, read e-books, listen to music.
So they probably would be happy to have the original iPhone today, but with software and an open source software eco-system that allows it to still function like the day it came out.
Assuming, we had reverse engineered all drivers, we could even improve it and give it state of the art cryptography for messaging and phone calls.
Expect the rate of improvement in smartphone hardware to gradually decline over time, just like it already has for desktop and laptop computers. At some point even low-end smartphones will be "good enough" for the vast majority of users, and then device vendors will shift most of their innovation efforts to whatever is the next new hot form factor like AR glasses or whatever.
> At some point even low-end smartphones will be "good enough" for the vast majority of users
I think we are more or less already there, maybe not the lowest of all ends but buying high end already feels like buying that more shiny expensive car with more horsepower which is hardly more useful because you are unable to leverage it.
I'm still perfectly happy with a 5 year old MacBook from mid 2012. It has a 2.7 GHz i7 CPU, and 16GB of RAM. I think I'll still be happy with this laptop 5 years from now (unless it breaks). I played with the new touch bar at an Apple store, and you can do some cool stuff, but not enough to make me buy a new laptop.
I think I would still be happy with my iPhone 6S 10 years from now, but there are many reasons I would upgrade. Here's a short list:
* Waterproof (Come on Apple, it can't be that hard. Samsung did it.)
* Longer battery life (I'm looking forward to >= 1 week)
* OLED screen with deep blacks and vibrant colors
* The dual camera focus hack in a regular-sized iPhone
* Some breakthrough in camera technology that gives a proper depth of field without the dual cameras. (I'm starting to get really disappointed with my 6S photos, but I don't want to carry around a 7 plus or a huge DSLR.)
I don't care about the processing power or memory any more. The iPhone 6S has a perfectly smooth UI and all of my apps run perfectly fast.
There are plenty of people using very old devices by Western standards in emerging markets: for them, it's a question of whether you want internet access or not: modern devices are totally unaffordable (heck, charging can be expensive, and involve walking several miles to your uncle's farm where there's a generator).
That's the first of a kind device. 10 years from that is not a good model for predicting the change 10 years into the future. In the last 5 years the rate of change has clearly plateaued.
I like the ambition. I've had a tinfoil-hat theory for a while that phone hardware is only built to last roughly 1.5 - 2 years gnerally before they physically begin to degrade. I've flashed new roms to a Galaxy S3 but it tended to reboot (i.e. hard crash) a lot still.
I would very much like to find a way to use something like this on old devices that have locked bootloaders that prevent third-party kernels from booting. Is anyone looking at a way to bootstrap a system like this underneath the manufacturer's Android kernel?
That'd be nice. I bought a MOTO X last year, don't do much on it besides using the camera and FB (hence my choosing a lower end model) and I can't use it for more than a couple of minutes without hiccups and app restarts.
It means: remove the chroot folder (safely, with umounting everything first).
You can do it with "pmbootstrap zap", it will ask interactively, which chroots you want to delete.
Disclaimer: I'm fully behind this idea, but I think there are some issues to be ironed out.
I don't think the OS is really the bottleneck here. I think getting a mobile device's hardware to run for this long is. There are a few things that make this difficult:
* Batteries - These currently only have so many cycles. A 10 year phone would need to provide a largish battery holding area allowing for two wires of varying voltages to be supplied. It would then in turn need to be able to charge at these various voltages through those two wires. I think voltages ranging from 1V to 24V would be a reasonable guess at where future battery tech would stay. Any person could then feasibly change their battery for some future tech.
* Connectivity - In just a few years, we've gone from only really having 2G to having just 3G/4G. In 10 years time, I think it wouldn't be unreasonable to think that maybe 4G will be switched off. I think it's unlikely that WiFi will remain for the next 10 years too. You need some robust connectivity module that hackers could connect to an arbitrary home-brew hacker's module. Something like UART/Serial/I2C/SPI.
Other things could burn out over time, but replacing those things becomes way too difficult. The processor, GPU, RAM, case, buttons and screen are just things I think that need to be the core phone.
Another place that could be interesting to explore is converting old, arbitrary phones into useful everyday devices. For example, I'm looking at converting my old HTC wild fire phone (the screen is small) into a permanent low-power alarm clock. It'll sit on the WiFi and if it sees my laptop/phone on the WiFi it will schedule the alarm, otherwise not.
I have another slightly newer phone with a larger display, so I was thinking or either making some notification board for my office or some very cheap VR headset for watching YouTube videos in bed without holding my phone up.
That said, there's no reason why all of these new "smart" devices couldn't reuse old technology. I think a small technology company that recycled old mobile devices into new smart devices would be pretty awesome. People pay loads for a recycled pencil for example, why not a recycled electronics device? On the far end of the scale I've seen some awesome coasters made from old motherboards.
I would love to see offices and schools buying more second hand kit too. There's no reason why thin client software can't run on an old mini-intel board, or even a semi-old smart phone. Kids don't need this year's laptops to write a report in Word. I think there's a big space for reuse here.
>>In just a few years, we've gone from only really having 2G to having just 3G/4G. In 10 years time, I think it wouldn't be unreasonable to think that maybe 4G will be switched off. I think it's unlikely that WiFi will remain for the next 10 years too.
I'm using a Nexus 5, which was launched almost 4 years ago, that uses 4G. A reliable, widespread 5G network is at least 2 years distant, probably more. Barring breakage, and changing the battery, the Nexus 5 could almost certainly be used 6 years after launch, and still work well. The barrier for that sort of lifetime is software, not hardware.
I think you're right that 10 years is pushing it.
But, firstly, bear in mind the range of smartphone users there are. Really large percentages of people have never installed an app for instance. There are grades of users on their technology demands, and just as there will be a percentage that always want a brand new phone, there will be a percentage that just want something that is stable. There are a substantial percentage of people that are more worried about the UI changing underneath them than whether or not they're using 3G or 4G.
Secondly, even if 10 years is unlikely, the prospect of being able to run a functioning device indefinite ly allows the lifetimes to increase by more realistic but still significant amounts, 10 years is niche but 4 or 5 isn't.
>I'm using a Nexus 5, which was launched almost 4 years ago, that uses 4G. A reliable, widespread 5G network is at least 2 years distant, probably more. Barring breakage, and changing the battery, the Nexus 5 could almost certainly be used 6 years after launch, and still work well. The barrier for that sort of lifetime is software, not hardware.
My prediction for the communications modules not lasting is also trying to take into account that the rate of developments is likely to speed up - not stay the same or slow down.
>There are a substantial percentage of people that are more worried about the UI changing underneath them than whether or not they're using 3G or 4G.
I'm talking about simply being cut off from the rest of the world. Many Countries have now fully retired 2G networks. In less than 5 years, I can see the same likely to happen for 3G.
>Secondly, even if 10 years is unlikely, the prospect of being able to run a functioning device indefinite ly allows the lifetimes to increase by more realistic but still significant amounts, 10 years is niche but 4 or 5 isn't.
I think phones should at least match laptops - there's no reason why not. I think we should encourage a few things:
* Standardization of battery sizes - why is every LiPo pack completely different? If think this might even be arbitrarily pushing up the market value of the packs.
* Expansion ports - why is there no expansion port on a phone? Something just big enough to plug a small PCB with a standard connector. That way you could add something onto the internal bus to add functionality. For even common users, this could be slightly more battery, secondary processor, increased RAM, additional GPU, AI chip, VR chip, better camera, louder speaker, new communications, etc. Looking at the Intel Edison, if you can put an entire computer into an SD card [1], you surely could add value to your phone, either immediately or in the future.
* Unlocked - you should be able to easily load your own OS onto your phone with no consequences. This is the same with most laptops. Maybe somebody can't make a phone OS to last for 10 years, but I'm sure there will be some OS in the next ten years. Is it unreasonable to be able to swap your OS?
Completely unrelated, but another thing I would like to see is the use of colour e-ink displays for phones [2]. Video is something that needs to still be solved [3], but I think we pay too much energy for viewing static content on a phone display.
I'm still using an iPhone 4S 64GB from 2011, which my dad bought for me as a graduation present.
Partly because I'm too poor to upgrade, and partly because it still meets most of my needs.
The biggest problem I face is software rot in iOS 6. That would be the same regardless of OS. I can't use WhatsApp any more, or Kakaotalk. Facebook used to forward my messages to email, but now they demand that I use a Messenger app. Instead, I just check Facebook's website using the browser.
There haven't been any really significant apps that I needed that have been released in the last 5 years. I've made a list in case I ever update my OS. I don't need BBM or Pokémon Go. The messenger apps are the worst for forcing software updates, but I hope that people will eventually move back to email and SMS, which just work.
I got my first smartphone in June 2009. Here's the list of what's happened (in no particular order):
* 2 phones had their 2.4GHz radio die (no bluetooth, no wifi)
* 2 phones were damaged by water
* 1 phone had screen cracked
* 1 phone was lost
I'm now on my 7th phone in 8 years.
Ironically it was my first phone that lasted the longest (about 2.5 years before the 2.4GHz radio died). Since then I've gone through about 1 phone per year. Phones have gotten much more fragile as they've gotten thinner, though water-resistance is now more common so that could solve at least some of my problems.
I do get what you're saying though because my wife's smart phone lineup is:
Palm Centro
iPhone 4S
iPhone SE
Each upgraded because software that had previously worked stopped working.
Your experience is such a compelling argument for libre software and hardware, that I'd stop using Facebook and Google right away, if I still were using them.
But from reading through the comments, no one has analyzed it yet, if you mean that. It has less than 2000 lines of code, and it is Python, so it shouldn't be hard to read through.
Oh and there is room for improvement. Automatic Travis CI builds and what not. Alpine Linux already does this for every package, so with some effort postmarketOS could also do that.
I think it's very ambitious to not base this off of the work being done on Android but of course would love a proper GNU/Linux distro on my phone as well.
Lineage OS (formerly CyanogenMod) will run on a Samsung S II which is 5 years old. It might support some older devices that I am not aware of. I could see it still being supported 5 years down the line.
https://download.lineageos.org/