I have a librem 5.
At first I tried setting it up with a keyboard, mouse & monitor, but it was a little fiddly to get everything usable.
I had a USB-C hub from a macbook with PD + ethernet + hdmi, but I couldn't get the hdmi display to sync.
So I got another usb-c hub with ethernet + dp (+other ports) and that worked.
amazing: my phone desktop showing up on my 4k monitor!
So, that was pretty cool, but I still had a lot of fiddling to do - like which display the app comes up on, and how to switch things around. Also sometimes the display wasn't recognized when I wanted it do - I think I had to reboot with the display plugged in, because hotplug sometimes didn't work.
(There have been software updates since I tried this - it may be better now).
However, at the same time I ordered the second hub, I also got a $15 USB-C + PD + ethernet dongle.
Turns out that was really what I wanted. Let the phone be a phone, sit there charging and ssh/scp into it from my regular system. I can access the entire filesystem.
All of this really drives things home about Apple.
When the iPhone came out, apple was breaking new ground, so people even accepted web apps, because before they had nothing. And with quality/security, the app store made sense.
But after over a decade you realize that the apple ecosystem is more about control.
Why can't I write my own apps (without asking permission/eula). Why can't I run my own apps without either signing up as a developer and paying, or renew some certificate every 7 days? Why can't I plug all the things into USB and use them?
Apple has mindfully denied a lot of utility for their users. I thought maybe the mac usability would trickle over to iOS, but it seems the iOS lockdown has trickled over to the mac instead.
EDIT: wait. Let me say "Thank You" to the purism folks for doing this. It took a few years, but this phone makes me really really happy. I figure in a year or two it will let normal folks happily run their lives ios/android-free.
So, effectivelyâyour phone is your computer? This is awesome. This is what general purpose computing was meant to be.
Marketing caused us to regress to the fragmented landscape of gimmicky special purpose devices that we have today.
On recent droids this works surprisingly well. I have a keyboard in my bag instead of my laptop now.
The issue of display is the only thing that keeps me dragging a laptop around. I've been keeping an eye on monitor replacement AR glasses but looks like I'll be waiting another decade or so for anything consumer ready.
A shame Google's Pixels don't support wired video output.
I sort of agree. It wasn't marketing it was the failure of regulators to enforce the same sort of regulations they used against Microsoft to stop apple et al from becoming monstrous entities in the back of policies that stifle innovation and limit choice.
> I figure in a year or two it will let normal folks happily run their lives ios/android-free.
I hate to be "that guy" - and it seems like I always have to be "that guy" - but this will never allow "normal" folks to happily run their lives iOS / Android free.
The reason is because the collection, categorization, and collation of data on smartphone users is what drives the whole thing forward. It's the gas. The petrol. The solar power. The wind turbines. Without apps collecting and using data about their users, everything ends up having to be monetized differently and people have already shown that's not going to happen. You can't even get people to pay a few dollars a month for the single most important "app" on the Internet - email.
Sorry, but Librem 5 will never be more than a fun developer's toy. Most people are honestly pretty lazy and the sad simple fact is that they do not want to be challenged at all ever. They want to show up somewhere... bang on their keyboard like a monkey, or swing their hammer like a chimp, or smack their buttons like a macaque, and go home eight hours later, and then in 7-14 days, collect their bananas and eat them while watching Ow! My Balls!, or a slightly more advanced version of it like, American Ninja Warrior.
While the rest of nature is in a desperate bid for survival-at-all-costs that is consistently wiping out the slowest and the dumbest, through our struggle, we've managed to regress... we allow the dumbest and the least fit to survive. I know a bunch of wannabe evolutionary biologists here are going to throw around ideas they don't really understand, like, "Fitness doesn't mean being the smartest or the strongest, just the organism most able to reproduce!" No, that's just one single strategy, and not even always an optimal one.
This isn't an argument for killing off the dumb and weak, of course. I'm merely pointing out that most people have zero desire to be truly challenged, on anything, ever. That's exactly why they want their Samsung walled Android experience and their Apple walled iOS experience.
I wish we lived in a world where everyone spent a few dozen to a few hundred hours learning hard stuff to thereby make their lives permanently easier, but we don't. Most would rather spend a few hours doing relatively monotonous stuff over and over than actually learn something new.
OR
Maybe I'm just being an old curmudgeon. I guess we'll see.
> Without apps collecting and using data about their users, everything ends up having to be monetized differently and people have already shown that's not going to happen.
F-Droid and GNU/Linux repositories work flawlessly without any ads and, moreover, have much less malware, if at all.
> much less malware, if at all
Backstabber's Knife Collection: A Review of Open Source Supply Chain Attacks - https://arxiv.org/abs/2005.09535
F-Droid's a great project, but needs to take security more seriously, even if that means accepting some help from adjacent projects.
Youâre a curmudgeon, Linux phones donât need the success of Apple and Google to survive. As Purism is demonstrating, the work just needs to be done and open sourced. People who prefer Apple and Google phones can remain in that ecosystem just as people who would like to live outside of it in their personal PCs can today. You will see a schism in the market. It may not be dramatic enough to âcause disruptionâ but itâs enough of a market to sustain as Pine64 and Purism and System 76 and many others have proven. People want Linux devices.
You don't know whether people want to be challenged or not. Not everyone wants to be challenged at everything. You can be good at something and still both suck and not have any interest in other things. Brilliant programmers might have no idea how to cook anything. Brilliant doctors buy iPhones for the convenience because they have zero mental attention left to focus on things like tinkering with technology even at the slightest. Of course many people don't want to be "challenged" at all, but MOST have something that challenges them constantly, and I would say - a very important aspect you seem to neglect - the majority of people on earth, perhaps even in "developed" nations, have their hands full trying to survive and barely making ends meet, with no energy to spare on other less urgent matters.
> I wish we lived in a world where everyone spent a few dozen to a few hundred hours learning hard stuff to thereby make their lives permanently easier, but we don't. Most would rather spend a few hours doing relatively monotonous stuff over and over than actually learn something new.
I get this, but I'm not sure that using a phone that exposes a full Linux system is the answer here. In fact I think for many people one could make the opposite argument â spend a few hours learning how to use an iPhone and never have to worry about many classes of problems.
Yes, many people use iPhones because they don't have the ability to make an informed decision, but many people also have all the skills to make that decision and still choose them. It's all trade-offs, and for some, walled gardens and EULAs are not a problem, and having a camera and (closed source) software that allows better photos of their kids is far more important.
> I'm not sure that using a phone that exposes a full Linux system is the answer here
or the full list of vulns in the Linux kernel, for that matter.
For a second, I thought, "Man is it really that slow to compile a hello-world Python GTK app?" I completely forgot he was doing this all on a phone.
I'm super impressed with the state of "convergence" here. Dragging a window over from a full desktop monitor to the phone was exciting to watch.
I doubt I would want to do development directly on the phone, but only because the phone is going to be underpowered compared to a desktop. But since it's just a standard linux distro on the phone, it probably wouldn't be too hard to set up the Librem to automatically pull the latest build from the desktop/laptop for testing purposes.
> I'm super impressed with the state of "convergence" here.
Convergence is great for productivity. Ubuntu Touch's main USP was that and I was even able to get it working in UBPorts on my nexus4[1].
It's a shame that convergence is still a hit or miss on android ecosystem, At this point of smartphone compute hardware & USB-C standardization we should be able to plug any smartphone to a monitor. At least Linux smartphone ecosystem is betting high on that.
[1] https://twitter.com/heavyinfo/status/1251048583190609920?s=2...
it is quite ridiculous. as a passionate Samsung DeX defender at first, I dove right into that ecosystem when the s8 came out and am now at a point where my s10 is never getting a software update BECAUSE ANDROID BROKE THE API FOR SAMSUNGS WSL. Termux was a victim too to a much lesser degree but Samsung had literally had Linux on Dex beta and I was absolutely free to just have my phone.
It was incredible going to school, sitting down at a random monitor and just having all of my vscode files and blender assets follow me around
What a shame, I thought at least Samsung have their DeX albeit with special hardware. So no more DeX? I presume some new Android phones at least support mirroring through USB-C.
> "Man is it really that slow to compile a hello-world Python GTK app?"
GNOME Builder uses Flatpak - when you press "run" button it does full flatpak packaging and runs the resulting package inside a sandbox. Of course you could still run such app natively, directly from the terminal, in which case it would be pretty much instantaneous :)
Now that a lot of GTK mindshare seems to be moving towards gtk-rs, I have a feeling that it's about to get a lot less instantaneous.
This isn't a swipe at Rust, per se: if you're using Rust you've probably already decided that build speed and portability aren't as important as memory safety and performance, and I'm sure that's often a valid trade-off.
I do think that "dev builds" should be much faster than release builds, and that means building natively should be default.
I believe this is actively being worked on; there's a compiler backend being built using Cranelift instead of LLVM that will supposedly improve debug built times by a fair amount, and then the full release builds will switch over to the LLVM-optimized version. Last I saw, the improvement between the existing rustc and Cranelift-based one was something like 15%, but that may have changed in the interim.
Indeed - however, I'm usually compiling things straight on the phone, it's not that slow. Rust can be challenging in this regard, but works acceptably well once you get dependency crates already compiled (which can require adding some swap, unfortunately).
I've been making a hobby app with GTK Rust in GNOME Builder for Pinephone and eventually Librem 5.
App runs well on Pinephone but building Rust GTK App on Pinephone takes forever. I did it once out of curiosity.
Luckily qemu cross compiling is easy to do on laptop and SCP it to phone, with native Linux SSH and root on phones it is refreshing.
Cannot wait for Librem 5 to get shipped, it's a lot more powerful than Pinephone, but I'm super happy to be experimenting on the future of Linux phones and convergence, regardless of where it ends up going.
> App runs well on Pinephone but building Rust GTK App on Pinephone takes forever.
Have you tried Lazarus on the Pinephone and the Librem5? It already runs on different ARM boards and apparently it does on the Pinephone too producing fast code.
https://forum.lazarus.freepascal.org/index.php?topic=52217.0
Check also the video on mega.nz linked in that thread.
Thanks for suggestion, haven't tried it before but I'll check it out.
Is it publicly available? If so, may I list it on https://LINMOBapps.frama.io?
Once I get at least close to an MVP I'll get in touch!
It's an exciting idea that as phones get more powerful, plugging the phone into a monitor could replace the desktop entirely for many home users.
This is also true of the pinephone. My entire dev set up for for the pinephone is on my pinephone. All I do is hook up a USB cable and SSH into it (which makes testing a lot easier since mmsd needs to talk to the modem).
What is really neat though is any apps that I develop on my pinephone work on the librem 5 too! Some of the bugs found and fixed on mmsd are from Librem 5 users, and I have been collaborating with the chatty dev (I assume he has a Librem 5, I'm 90% sure he is a Purism employee) with the work I do on the pinephone.
I am looking to purchase either (or both) of those phones. Would you recommend buying one of them now or waiting until a future release? It seems that both companies have extended the release dates a bit.
I would say "Yes", but I also have two Pinephones and have a Librem 5 order in the queue. One pinephone is my "dev" phone (since mmsd tinkers with core mobile networking stuff, I don't want developing to interfere with my normal usage).
I do not think either the Pinephone nor Librem 5 will have any more significant hardware changes, so the Pinephone/Librem 5 you buy now will be the same one you buy in a year.
If you're unsure, I think buying a Pinephone is a safe bet since its $150 for you to try, and if its not for you, you can probably get your money back by selling it.
What do you want to use it for?
For me, Mobian/Pinephone does everything I need it to do for typical daily usage except for Chatty supporting MMS (and hopefully this will be fixed within a month or so).
I care most about 2 android 9 apps I have that run in the background and use Bluetooth. Other than that sometimes I like to tinker and build small projects with the libraries available but am more concerned with the android 9 apps. Other primary usages are phone, text, and browsing.
The version of the librem 5 that is shipping now is the "evergreen" model. Their next model is going to be named "fir" which is supposed to have a 14nm CPU[1] as opposed to the 28nm CPU in the evergreen model. This should mean better battery life, so if that is something that interests you then you might want to wait on the librem 5. Personally, I don't want to wait that long so I'm getting evergreen.
CPU is probably not going to be the only change for Fir: https://source.puri.sm/Librem5/community-wiki/-/wikis/Freque....
I'm personally waiting for the next PinePhone iteration with a better chipset, since the CPU is pretty slow for even budget phone standards.
Combined with software improvements that will hopefully give a relatively smooth experience.
Then again, the PinePhone is so cheap that it's not a big risk either way.
Donât hold your breath on the next PinePhone iteration. On the forums, people expect a new hardware generation to come in only 3â5 years from now. Pine64 seems to assume that the hardware is fine for now, it is only the software that hasnât been developed and optimized enough.
You mention Anbox in another post here. Donât expect that to allow you to run banking apps on Librem/PinePhone, as banking apps increasingly require the Android phone to pass Safety Net, and Anbox doesnât.
You can't compare performance on the pinephone to Android phones because the OSes and app ecosystems are completely different: You just run Linux desktop environments and apps that are designed to run without hardware acceleration.
I run FVWM and Firefox on mine and it works very well.
And why wouldn't you order Librem 5 now?
If you order Librem 5 right now, you get it "in a few months", so the software will be much more mature (while hardware already is). Pine64 currently accepts preorders only for the "Beta-version" of Pinephone.
I see. It seems as though the price tag has gone up compared to the dev versions. With that price I imagine theyâd want things to be stable. I havenât looked to see if they have a vcs or chats to see what types of issues are being addressed.
Pinephoneâs âPreorderâ is mostly just to scare off people that want the software to work, from what I can tell.
If you order a pinephone today, they estimate theyâll ship it to you in late April.
Also, their âbeta phoneâ is the Nth revision of shipping hardware.
The key takeaway to me is that developing mobile apps for the Librem 5 on the Librem 5 is a first class development workflow.
This is great. Not having to worry about setting up toolchains on a separate platform, cross-compiling, and then signing/transferring apps and running remote debugging sessions makes it easy to just build and test apps.
Can anyone explain to me the benefits of staying native, and not switching everything to web apps that use properly implemented web APIs for everything?
No signing also means that the workflow for developing malware for the platform is exactly the same, too. :)
My issue with the Apple ecosystem is not the cryptographic protections - it's with the fact that the keys aren't mine.
I'm not sure I understand. Crypto won't protect you from malware by itself, it just creates a chain of custody. App store apps are not secure because they're signed, they're secure because they're signed by Apple.
For the librem it's just like a desktop computer, don't install software from untrusted sources, keep your stuff updated and you should be fine.
Meh, signing by Apple really doesn't mean more than Apple looked (visually) at a running instance of the app and said "eh it doesn't look like it's breaking the rules."
They don't do any instrumentation and often the developers themselves don't even understand what all the binary dylibs they're including do.
The real reason iPhone apps are safe is because they run in a controlled sandbox with a very limited set of device APIs available to them. You could take away the app store and signing and things would mostly be the same.
Being able to outsource the decision about what is or isn't an "untrusted source" to a security expert is valuable.
Platforms that only run signed code permit that.
You're mixing developing and distribution. You don't need any (meaningful) signing to develop for Android and not even for iOS (although testing is harder there). App signing comes in only when end-users are installing your apps.
As for malware installation on Librem, that is a separate question. If you don't give apps root access and only install from the repos, you're about as secure as on Android or iOS. Once you start installing from unknown sources, however, it's true that you're screwed - but that's because of the permission system, not signing.
I don't understand this line of thinking. There is loads of malware in the Apple Appstore.
https://www.techradar.com/news/apple-app-store-is-apparently...
there are tons of apps on iOS and Android app stores that are malware and they are signed
I'm reading a lot of the comments here and it seems the salient argument is that this will never replace Android or iOS because this or that.
99% sure those comments are correct.
But still, what gets me so hot and bothered is that it looks like purism and pine are making devices that cost almost the same as my other subsidized phone (android) AND might be a viable business EVEN IF there isn't a critical mass.
Linux gets better and better every year even though it will never replace Windows or OSX.
The big deal here is that we might finally have an amazing phone that gives us a complete and awesome developer experience AND respects privacy and the right to tinker fully with our phones.
We've had that on the desktop but NEVER on our phones. This is a sea change that seems worth getting excited about. And it doesn't matter if that doesn't destroy or beat out the alternatives.
This device will never be the marketplace where app developers try to sell my kids shitty apps. Who cares. It doesn't need to be the biggest to be interesting and incredible.
> We've had that on the desktop but NEVER on our phones.
Give it some time. We've had to wait decades to see some people starting the transition from closed desktop systems to open ones, and we aren't nowhere close even to a tenth of the Windows/Mac userbase. Mobile platforms in comparison are newborns, and the industry after learning the PC lesson now is actively fighting against Open Source software and standards by producing software that refuses to run anywhere but their platforms (Whatsapp, banking software, etc) while they spend money to tight close their terminals so they can't be reflashed with Open Source operating systems, a practice that also creates pollution related issues. Time will tell; we probably need at least 5 more years and a couple killer apps that will make a significant number of users move to an open device such as the PinePhone or the Librem 5, or what device will we have by then, so that those figures can't be ignored anymore.
This phone and developer experience looks like it is ready now. My point is that we don't have to use a terrible developer experience to work on an open system now and we don't need to have it be a massive economic success either. This is what we've been waiting for and it's ready to hack on.
That's the right attitude, IMO. Most consumers respond to polish and 'wow!' factor. The Linux desktop doesn't even provide that compared to the competition... Linux on mobile is even further behind both from a hardware and software standpoint. For those who don't care, or are at least willing to overlook the limitations, Linux phones are becoming increasingly viable and provide a number of advantages that the average consumer doesn't value. But there are those who do so there is a market for them.
I don't even use my PinePhone as a phone that much... but it's a fantastic mobile computer. I'll continue to use it and be back for the 2nd gen PinePhone.
I'm a little annoyed it took this long to get reasonable smartphones, but I'm so glad they're finally here.
--A happy pinephone user
You are happy with a phone that won't run an authorized Signal app, takes 5â10 seconds just to open the screen to turn wi-fi on/off on Mobian, and whose map apps are all barebones tech demos compared to OSMAnd on Android? I have a PinePhone too and occasionally hack on it, but I wouldnât call it a "reasonable" smartphone yet when the software is still so half-baked and, unfortunately, it hasnât succeeded in attracting very much of a dev community (as the existing devs sometimes complain).
Maybe Iâm so disappointed because I remember the Nokia N900. Now that was a reasonable Linux smartphone, except for the blobs.
> phone that won't run an authorized Signal app
This is a fault of the Signal developer, not of the Pinephone.
> takes 5â10 seconds just to open the screen to turn wi-fi on/off on Mobian
It's the beta release. It's gonna be faster with GPU acceleration AFAIK.
> Maybe Iâm so disappointed because I remember the Nokia N900. Now that was a reasonable Linux smartphone, except for the blobs.
But how much time passed after the release until it became a reasonable smartphone? Give Pinephone some time.
> It's the beta release. It's gonna be faster with GPU acceleration AFAIK.
No it won't. A (the?) big problem with performance is that we're running desktop applications on a slow-ish mobile device. (all of it: slow CPU, slow GPU, slow RAM, slow flash etc.) To get a good experience you need to throw more (i.e. faster) hardware at it and re-architect and strip down key applications. Or write new ones from the ground up designed for mobile devices and the resource limitations that come with them.
The 'beta' label is an attempt to set expectations re: the state of the distros right now. When they change the label to 'final' people are still going to be complaining about performance.
> This is a fault of the Signal developer, not of the Pinephone.
Whoeverâs fault it is, is immaterial. The PinePhone is still lacking as a phone for at least those who use Signal.
> It's the beta release. It's gonna be faster with GPU acceleration AFAIK.
Can you cite that? The problem is that Mobian's stack is based on a lot of GNOME libs that have not been optimized very much and run slow even on desktop, so GPU acceleration wonât help much. Before you mention the other available OSes on the PinePhone, UBports is obsolete, bitrotting tech.
> how much time passed after the release until it became a reasonable smartphone?
The N900 was a reasonable smartphone immediately upon its release, at least in terms of being responsive and having core apps and functionality that people expected at the time. Nokia had a larger team of developers working on Maemo than there are working on PinePhone OSes.
I mean its two different things: Nokia a mega corp (at the time) and Pine64 (a few hardware enthusiasts + small overworked dev community).
I get your frustrations and as a PinePhone owner myself I would not consider it anywhere near usable as a real Smartphone replacement.
However after installing Arch it is pretty snappy and has certainly sparked some new hope in me. I really recommend you to check it out.
For Signal there is (or will be) Axolotl [-1] which looks promising and when it comes to developer support both KDE and GNOME devs show some support for it. E.g. GNOME recently announced libadwaita [0]. Give it some time and check out the Arch release for PinePhone: [1].
[-1]: https://github.com/nanu-c/axolotl
[0]: https://aplazas.pages.gitlab.gnome.org/blog/blog/2021/03/31/...
> mobian
Itâs hardly fair to blame pine for gtk issues when theyâre shipping kde by default. I can install non-working software (even broken android) on any jailbreakable phone.
Phosh gets way too much attention on the pinephone, it's not a good DE and it's very slow. You can run normal Linux X11 DEs with no compositing and they are very fast. I use FVWM and most things open very quickly, Firefox takes a couple seconds.
All people like me want is something that lets us run tmux and the rest of the apps we run on our thinkpads and the Pinephone does that very well.
It's not the DE that makes things fast or slow (although disabling composition may indeed help with making GTK3 apps run faster on the PinePhone, but that's mostly because GTK3 renders in software and it bottlenecks on PinePhone's RAM bandwidth).
How can you use a smartphone mini screen with tmux? Is it usable?
Out of curiousity, what distro are you running?
Honestly, I have no idea why I'm so surprised. This is what happens when you put desktop dev tools on an (albeit modified) desktop desktop environment on an (albeit modified) desktop operating system on a phone. Can't wait for these devices to mature!
My phone has been at compute parity with my i5 Macbook air for at least the last two phones I've had. I've also been able to plug my phone in to a monitor/keyboard using the same usb-c dock I use for the air and be pretty productive in a pinch. Love the idea of having a real desktop environment instead of Android, though.
It's only a matter of time until we can develop smartwatch apps - on our smartwatch.
Why would you post that when it clearly canât be programmed on device?
It shows the path to that. At least it can be programmed.
It can be programmed, on-device:
https://www.youtube.com/watch?v=tuk9Nmr3Jo8
WaspOS is pretty powerful!
Look at Bangle JS, it has JS interpreter on it. There is no IDE on it, but I don't see it impossible that one connects a Bluetooth keyboard and opens up the console on it.
Check out: https://asteroidos.org/
Get a daily email with the the top stories from Hacker News. No spam, unsubscribe at any time.