As someone who backed them in 2017 I'm so happy to finally see the specs. It's amazing to follow their progress, I love what they are doing with Gnome/Phosh, that gives me hope that there will be many native apps for the most important tasks and not just some web-app mess.
For the people complaining about the price/specs ratio: this phone is no standard off-the-shelve system, it had to be designed with user privacy and openness in mind while having a ridiculously small market. I'm amazed they did it so cheaply.
Personally I hope that they will be able to sell enough to continue development. Maybe this will lead to a second generation with multiple models for different market segments. I'd love a librem phone with a good old blackberry keyboard in the <1500$ price range with a little more RAM/CPU power and screws holding the back plate.
I'd go even a step further: Since they open-sourced all their work, maybe other vendors can pick up / fork where they left off and we may actually have more than one Linux phone in the future!
Two years ago that seemed like wishful thinking to me, but I am very excited about how far Purism got in an even shorter timespan.
Like PinePhone? They're pushing for a ~$200 phone in Q3 this year, same as Purism. The difference is that Pine64 doesn't seem to be putting nearly as many resources into development, just enough to get functioning hardware out there and then leaving it up to the community to build the ecosystem.
It's an exciting time! While I'm waiting, I'm setting up my own PBX to hopefully take control of my phone service, and I'm going to experiment with using VOIP over data instead of going over the "regular" voice channels.
I'm super excited to see what's possible with an open platform like this.
Having set up a PBX before, what does it give you that allows you to "take control"? You've still got to connect to some phone provider. It's basically just a highly-configurable alternative.
GNOME and Linux are under the GPL. So are phosh (Purism's shell), Chatty (their messaging app) and probably a lot of other components. So you can't make a closed fork.
It's usually quite easy, they're often on Github though some companies host them themselves. There are a few companies (looking at you Lenovo) who release partial/non-working sources but most vendors do comply with the GPL.
It often seems to take some persistence, but typically it is possible to get the kernel sources. The manufacturers can drag their feet for a while but ultimately they have to give it to you.
Compare that to the rest of the software on an Android device; you'll have no luck convincing Samsung to give you the source to any of the UI bits on their phones, for example.
Linux specifically does not use the GPLv3 for exactly this reason. rms et al. complained about "tivoization", which came from tivo shipping boxes that contained the code but were locked down. Linus didn't really care as much, or at least not enough to re-license. He didn't like the restriction, and new that such a provision would prevent exactly what MS is now doing: allowing more freedom for the users.
That's orthogonal though. WSL would be fine even with GPLv3. It runs on a hypervisor. WSL poses no restrictions on modifying the source of the kernel. You can fork, rebuild and run. You can't on a tivo device.
Not exactly. Microsoft is trying to expand functionality in certain places to allow things like CUDA acceleration, AF_PACKET support, and other important hardware-y things that WSL can't handle. You're right that you can run a different kernel, but Microsoft has modified theirs for WSL. For that reason, they have to release modifications.
Have you read the GPL? It makes very little sense for hardware (it’s actually a pretty weird license full of all kinds of software specific things. “Linking” to “operating system components” for example doesn’t make them a derivative work, whatever all that means. There’s quite a lot of other stuff.)
The argument for using the GPL when creating copyleft software is that most copyleft software uses the GPL and you want everything to have a compatible license, that’s not really the case for hardware.
"For the people complaining about the price/specs ratio: this phone is no standard off-the-shelve system, it had to be designed with user privacy and openness in mind while having a ridiculously small market. I'm amazed they did it so cheaply."
How does that affect the price tag? Would it be less secure if they charged less?
Well, using an off-the shelf system like many of the cheap smartphone companies do allow you to keep development costs down, and leverage economies of scale even for small batches.
For one, they are bringing up a stack that's quite different from the one usually found on smartphones, which sounds quite costly.
From the security point of view, I am not sure it would be less secure on the sw side, but it would be less open. On the hw side, they have kill switches, isolated the firmware on ROM chips (RYF-compliant), and have a pretty strict security-first approach.
Unfortunately that places it out of reach for me, financially-speaking. I'll continue with postmarketos, and keep an eye on the pinephone.
If you constrain yourself to non-NDA hardware, rather than lowest-cost vendor, you've just notched up price quite a bit. Take out mass manufacturing economics, and the price seems quite economical, actually.
Isn't it mostly all off the shelf hardware and software though?
I think they're going to discover that a lot of people say they want security but when push comes to shove when they're buying a phone they first and foremost want a good phone. If you were really bothered about security and privacy on a phone you'd just use tracking blockers, a disposable (google) account, not log into facebook, turn off location, use tor, or existing encrypted chat software to talk to other people who are also using disposable (ie not their main email accounts to sign up to google) accounts etc. But do all this on a regular phone. I'd always been concerned about timely security updates, software that's been written by people with no experience with security and who have no reputation to damage by cutting corners.
I love that there are 3 hardware kill switches, and that turning all 3 off "additionally disable[s] IMU+compass & GNSS, ambient light and proximity sensors." There are so many ways in which clever hackers have teased out information from thin streams of always-on sensor data.
It would also be kind of cool if they added physical LED lights reflecting hardware activity, normally covered by some physical blinder that you could remove if you're curious--for example, a light indicating WIFI traffic, one for each sensor, etc.
Exactly. I absolutely detest the fact that my carrier can remotely activate my phone's baseband processor to conduct surveillance. Without a hardware kill switch there's basically no way to prevent this other than, perhaps, using a faraday bag.
@RandomBacon, what mistake? It's clear you were saying, a spy app can record all the audio, gps location etc in a file, and then when the phone regains cell signal, the spy app can send out the recorded audio, gps location etc.
Not the stochastic swine, but their mistake was saying GPS signals would be logged while inside a Faraday bag. GPS signals are RF, thus blocked by any proper Faraday cage.
Faraday bags don't block a full spectrum. In fact they tend to be rated only for specific frequencies. GPS and GSM/CMDA/LTE are different enough that you should never assume a bag that is designed to block one will also get the other. Same with WiFi, AM/FM, Shortwave, WWVB, etc...
I wonder if these are pure hardware switches, like in distinct on/off positions where the off position is wired directly to the sensor's power or similar, or more like buttons with software behind them which then does whatever action (and hence can be changed)?
So for me, this is roughly what I want in a phone. But the thing is, I bought a phone over two years ago, with better specs for $100 less. That same phone today is priced at $200-$300.
The way I see it the real value add for this phone is the software, and this whole mainline Linux, FOSS, run whatever you want approach is very attractive to me.
But is it worth $700? Is it worth $700 when my current phone is in great condition and most likely will be usable for several more years?
I feel bad, because I do want to support this kind of thing, I'm just not sure that buying __this__ phone is the right decision for me.
If you want good software you simply have to pay more. These are small teams with no prior experience trying to do something huge, it makes sense to expect to pay even 5 times more if the direction of the project is good. Hopefully future versions will be cheaper as they learn and gain experience.
That was my motivation as well when I prepurchased the phone. Even though you might not get a $600 phone, Purism and the entire Linux community gain a lot of experience about the difficulties of designing hand-held hardware, applications and operating systems.
In my opinion their progress has been fantastic, and I am quite excited for the future of Linux smartphones.
So I checked, and there are no "official" current LineageOS builds for any Nokia phones (unless I'm missing something on https://download.lineageos.org)
There don't seem to be any builds anywhere for the 4.1 (is it an uncommon phone?)
For argument's sake, let's say you bought the 6.1, now your option is to download an "unofficial" ROM from some pseudo-anonymous person with an anime avatar from XDA, hosted on whatever free file hosting they can find today.
Don't get me wrong, I used Lineage when it was called Cyanogenmod, and it saved one phone from landfill which is great. But I'm older now, and I need my phone to work and work well - I really don't have time for problems like https://forum.xda-developers.com/showpost.php?s=76595163ac85.... For some people, that's worth the money.
Similarly, I've pretty much given up on LineageOS ever getting the Nexus 5X to the 16.0 version. Their list of supported phones looks decent, but when you realize they can't keep even those phones up to date the project starts to look like a failure.
I'm hoping to find an alternative for friends and family who still have the 5X - it's been dropped from Google's security updates for a while, because their support ends abysmally quickly.
You hit the 5X jackpot - up to a certain manufacturing date, those phones were notorious for bootlooping and never working again.
More on topic, the LineageOS support requires a volunteer maintaining a buildable tree for the phone, and testing will require the hardware itself. I'm not sure how much CI/CD helps here.
You're right: of the four 5X phones owned by close friends or family, only 1 bootlooped (and one was stolen). It would be great to have continued support for the remaining phones, but planned obsolescence is too strong.
The Nexus 5X supposedly has a maintainer with LineageOS, but maybe they're busy or no longer own the phone.
Unfortunately, I believe the supported phone is the Nexus 5. The 5X is not a different version of the 5, it's the next generation of the smallest Nexus phone line, since the "6" was already taken. The numbering went Nexus 4, 5, 5X.
Yes, who complain that the software might not be that great, but who also want freedom, your option would be to buy without expectations as a way of donating to the project or to simply donate money directly to the project.
That's the great thing about open source! People form groups and spend their own time/resources to build what they want instead of waiting around for the corporate big brother to build it.
You're correct, it's not a donation. However Purism right now is important in a way Comcast is not. Their work on the software side is making the GNU/Linux phone a reality and it will help both vendors and makers a like in the future. Supporting the company is indirectly supporting the GNU/Linux phone ecosystem and may be one of the best things to throw your money at right now to push development forward.
On the Comcast side, they can get fucked. My whole city pays them a crazy high fee for internet and more a month with very little improvements being made in infrastructure. Sure they now a have a remote with voice commands, but I and most people I know are almost never near the advertised speeds. At least the latency is good.
B corps are allowed to prioritize specific values enumerated in their corporate charter, even at the expense financial gains. Regular corporations have a fiduciary duty to their shareholders, which in recent decades has been interpreted as requiring companies to maximize short-term shareholder value
There is no fiduciary requirement to put profits above all else. In fact there is no fiduciary duty to make profits, just basically to do what's best for the company. There is no short term time frame listed. If there is a requirement to make profits it's not because it's a responsibility. Shareholders may want it but no one gets punished if a bad decision for the long term success of the company doesn't pan out. There's numberous examples of CEOs that attempted to pivot when it wasn't the most profitable course of events and then receiving a nice severance more than most employees make in a lifetime of work for failing.
It's almost as if the more and more you compensate your c-suite execs with your own shares, the more you would expect to see them chasing short-term profits regardless of the degradation of firm longevity...
Knowing what to measure against is harder the higher you go up the corporate ladder.
Incentives get skewed towards short term employees and short term rewards. Companies treating their employees as resources has only hurt loyalty. No one starts at a company in the US anymore unless they're average or below workers. The whole idea of companies not being willing to pay existing employees what they would pay a new hire is stupid too. You're willing to pay more for someone you're going to have to train and is a literal risk, versus a known commodity with known output. Generally you pay more for stability on the market not less.
The smartphone market is filled with phones subsidized by information harvesting money and where the economies of scale side with entities in a race to the bottom.
I’m personally happy to pay a premium for a product with different priorities.
Your phone hardware might work for many years, but you will probably stop getting security patches for the software soon, so it won't be in great condition security-wise. Newer applications might start using features in newer version of the OS. The software, frozen in time, will make it gradually less and less usable even though the hardware is fine.
By contrast, my PC is over 10 years old and runs the latest up-to-date versions of the kernel, drivers and other software. I'm hoping the Librem 5 will be similar.
Hard to do when Apple intentionally slows down the phone with new firmware updates and Android OEMs abandon replaceable batteries for a soldered on unit whose ability to hold a charge drops off precipitously after 2 years.
I've never seen a cellphone with a soldered battery, and most of the slowdown in iPhones is caused by the OS slowing down to make up for reduced battery capacity. With apple and other manufacturers having to replace batteries in their own phones, I seriously doubt we'll ever see soldered batteries.
Same dilemma for me.
But I think environment priority is more important because freedom can't exist without environment.
(now, it just could happen that my current phone breaks for a very unfortunate, totally unplanned, "accident" :-))
Look at where the modern PC ecosystem is right now with windows: huge chunks of unusable plastic,fiberglass, and metal with serious software flaws that turn them into useless trash shortly after people buy them. When the community (much less, the individual consumer) can’t deal with broken software that’s tied to their (unupgradable) hardware it all just accumulates in landfills. People accept that you have to buy new phones and laptops every couple years because they have no other practical choice.
I’m not sure I could say one should have a higher priority than the other, in fact I’m almost certain I could say they can’t.
Assuming it's not an iphone, the phone you bought will likely be getting few/no more security updates and is obsolete, that's why it's so cheap now and in part why it was so cheap then. As you said, the value add here is in running mainline linux and that means your phone really will be usable for several years. Some of the other big pluses over android are ease of development, no java bloatware, more FOSS and no spyware. Those things are just eating your specs anyway.
Yes Java has many issues, but writing complex GUI apps in plain C in this day and age is the wrong thing to do. The early Gnome founders learned this the hard way which led to Mono, which at least had automatic memory management.
I fear this OS is destined to become like Samsung Tizen, which is riddled with security holes. It's not as if we don't have excellent libre languages, e.g. Rust and OCaml.
I applaud the open nature of this phone, but using the Gnome stack looks to me like a mistake.
Couldn't disagree more, writing apps in C is a breath of fresh air compared anything in java but especially android. I can start with a simple C file and write hello world in a few lines of C (dated examples: http://zetcode.com/gui/gtk2/firstprograms/) and run it on my desktop, IME the complexity curve remains quite flat. AFAIK mono never gained much acceptance (and was highly controversial at the time) and most of the gnome ecosystem is written in C, I'd be surprised if there was more C# code than JS and python in gnome today.
> I fear this OS is destined to become like Samsung Tizen, which is riddled with security holes.
Is the gnome desktop riddled with security holes? I don't know much about tizen but they seem to have written a lot more of their own stuff, the vast majority librem is battle hardened.
> It's not as if we don't have excellent libre languages, e.g. Rust and OCaml.
Rust is not mature enough, especially in it's UI libs and is virtual unheard of outside the HN bubble. The most mature UI available appears to be the GTK bindings though so it's mostly the same stack that you think is a mistake. OCaml despite being older is even more unknown and from a quick google it's most mature UI library seems to also be an out of date GTK (https://github.com/garrigue/lablgtk). Betting on either of those would be the death of the project.
> I applaud the open nature of this phone, but using the Gnome stack looks to me like a mistake.
Gtk and Qt are the only serious options as of today without having to sink in years of dev work. If anyone really wants to avoid C then there's the GObject system that lets you write code in js/python/xml/others, but I'm hoping this turns out to be the phone ecosystem that respects users and doesn't sacrifice their time and battery for the developers convenience.
> it's most mature UI library seems to also be an out of date GTK
What makes you think it's out of date? There is a gtk3 branch, gtk2 was used mostly because it was stable and cross platform. It's used by Coq, Unison, Frama-C, zeroinstall and others, and it's rock solid.
> Betting on either of those would be the death of the project.
Lolwat? What makes you think so? Rust Gtk bindings are official, and suggested to be used by Gnome team. OCaml is rock solid and used by the industry, for GUI apps as well.
For me, I need authy, anki, and a couple of other apps or else I need a second phone. Luckily, anki is open source too and hopefully somebody ports the linux version for the smaller screen.
I agree that the price is very painful especially when compared to Chinese Android phones that offer phenomenal value (just not in terms of privacy). I hope the open source phones succeeds regardless.
im an in debt, highschool dropout, unemployed, and ill biy this on credit just to take a stand. i cant wait to set my google phone on fire. will be like finally ridding myself of a disease.
get it together man. priorities. the value of an old phone stands up to forcing cash in meaningful directions? forget my wallet/welfare/bills, badd things have to stop at any cost.
What's wrong with installing Lineage OS on your google phone? If you stick to F-Droid you're pretty good in terms of open source. There are still a few binary blobs that you can't avoid of course. Those may contain some backdoors or bugs that will make you vulnerable, however you will no longer send your data to Google etc. all the time. Which is a huge difference.
Lineage dropped support for a LOT of phones in the last 6 months.
When you show up, looking to do the legwork for a specific model, the community directs you to the donate button instead...
If Librem can keep their repos current? Im down to switch from a Lineage-deprecated phone.
Besides, a phone running a kernel that isn't 3 years old and terribly out of date? Vulnerable to Spectre, Rowhammer, and a host of other script-kiddie-hackable holes? So what if its "mine" for $200 less when it isnt mine and patched?
Apple did one thing well with iPhone that Android didnt - they kept themselves as a single point of accountability for how the phone runs, how it updates, how security patches get distributed.
I hope that Librem is still around to sell me a hat/shield to turn a decent & specific single-board computer into a phone.
I hope their around long enough for that sbc to be RISC-V based.
I am happy that they made the LTE connection an M.2 card that can be easily upgraded. But without a discernible upgrade path for the rest, this feels like it might not be enough for long enough.
With some rough finances, Im gonna have to save up. This phone will be nice enough at$700, and a Category ~15-ish M.2 LTE module should run around $100-$200.
I do want to see some reviews and tutorials though. A configurable web-app sandbox is absolutely necessary.
Yeah larger screens gives them some more leeway internally but does kind of restrain them a bit too because the larger screen wants more battery which requires a certain amount of space.
If you've ever seen a tear down of the iPhone X for example the boards are actually stacked to maximize the space available for battery in the thin form.
I thought the same thing when I bought one 3 years ago, didnt use a case but used a screen protectoe and then I had to replace the screen three or four times in a year. Also each time the screen broke the digitizer stopped working as well leaving the phone unusable.
YMMV though. Im pretty hard on my stuff (I was also 22 and partying a lot when I had it)
Agreed. Repaired my screen, battery and cleaned my camera. Never used a case (but had to replace my screen after dropping it on tile/concrete a few times...really the only deadly materials to it). Had a screen protector but honestly all the drops happen on the corners so the protector is really just an overpriced sticker for me. Hoping future Librem 5/Pinephones can replicate the experience.
They've stated that they couldn't fit everything in 5" size (which was an optimistic target early on) a few months after the crowdfunding campain ended.
I'm a backer and I'd love to have 4.5" or less phone with the same features as Librem, but AFAIK there is no other options on the horizon.
Screen size is a hugely personal preference. I prefer having a screen as big as possible, and used a 7" fonepad in the past (that bezel though!). Now the 6.67" Oneplus 7 Pro is more than enough, and fits in more pockets.
I respect people that want a smaller computing device, and I'm sad the market barely has options for you. Nowadays a 5.6" Pixel 3a is considered "small".
One could ask if the market really big enough ? I haven't heard the 3.3" 2018 Palm Phone being a huge success, despite being virtually alone on this market.
Kill switches are just an addendum if you don't trust the software. Of course the OS can also prevent apps from accessing parts of the phones like the accelerometer if the user configures it that way.
I'm extremely excited for the Pinephone and intend to order one as soon as it becomes available. I may resurrect Hellaphone on it (edit: oops fix link https://bitbucket.org/floren/inferno/wiki/Home) or at the very least start building my own library of no-bullshit phone tools, free from modern phone crap.
I definitely intend to pick up a PinePhone. At 200$ I'm not going to be annoyed if the software isn't 100% or it can't replace all functions of my current Android device. For 200$ email, SMS, calling and VoIP are pretty much enough. The $650 Librem are asking, I'm sure is well justified, but is a much harder pill to swallow for something that I am well aware will not be able to replace my current device immediately.
Cool, I want to get one of those also, but they're not selling it and I don't see any links about when we can buy it, for how much, and when it will ship.
I can't find the info for the price, but I remember seeing it before and it was very low, like $200, which would make this a very exciting entry into the notAndroidOrIOS market. The Librem 5 was cool at first, until they started delaying it indefinitely and the price is too high for mass adoption. By having a very low price on the Pinephone they are courting a lot more tinkerers to get one and help develop the OS versus the Librem which people would kind of (rightly) expect to be fully-formed at launch.
As things stand today, we expect a fully functional prototype in August
The Allwinner A64 platform has numerous developers working on it compared to the iMX8, which has really fleshed out fully featured hardware support over the last few years. Running most Linux distros should be a breeze on the Pinephone!
That's a bit of an overstatement. It has some interested and active developers, but support is far from fleshed out, especially for the poratble device use.
There are lots of patches floating around, that you can apply to the mainline linux tree, to get CPU DVFS, thermal management, to improve I2S support, etc. Suspend to RAM is nowhere to be seen, leaving you with always ON SoC, which will drain the battery in half a day doing nothing.
It's all fixable, but let's not pretend A64 has fully fleshed out HW support in the mainline Linux kernel.
OTOH, there's a great potential, because Quectel EC25 broadband module also runs Linux and has some potential for being hackable:
The pinephone only has a wifi radio and lacks any cell radios, thus it is incapable of connecting to cellular networks. It's basically an 'internet tablet' and not a phone (at least in the 'traditional' sense) like the name suggests.
we also settled on the number, and the implementation of, privacy switches on the phone – there will be 4 switches in total: for the i) BT/Wifi module, ii) the modem, iii) cameras (front/back) and iv) lastly for the microphone.
and right on the pinephone page it says it has this chip:
Key Benefits
● LTE Cat 4 module optimized for M2M and IoT applications
● Worldwide LTE, UMTS/HSPA+ and GSM/GPRS/EDGE coverage
● Multi-constellation GNSS receiver available for applications requiring fast and accurate fixes in any environment
● Feature refinements: supports DFOTA*, eCall and DTMF*
● MIMO technology meets demands for data rate and link reliability in modem wireless communication systems
I think you are talking about the Necunos device. The pinephone is ... a phone. See https://www.pine64.org/pinephone/ where they list the modem being a "Quectel EG-25G with worldwide bands".
I've been looking for a "FOSS" phone that is based purely on AOSP. I don't want a custom ROM, Play Store, cloud, or to have to create any accounts or sign into anything. I just want a simple Android phone with decent specs that I can install APKs on. Does this exist at all? Or do I have to buy/pay for a "full" Android phone and reflash it?
Buy an unlocked bootloader phone (ie oneplus) and put lineage os on it. Lineage doesn't come with any of the google apps or the play store, just a few basic apps and the ability to put whatever else you want on it.
This is what I've been doing for a few years now. Some apps require Google Play Services (Revolut?), some show a warning that can be ignored and run fine afterwards (Authy, Signal), some have limited functionality like no notifications (Slack), but many work just fine (Firefox, Spotify, and even Messenger Lite)
I looked at the page, it is interesting, but I have never heard anything about this. They page didn't also say anything about who they were/what they do?
Do you have anywhere with more information about this/the folks behind it?
Their website does a pretty horrible job of selling the OS. I was interested until i saw 0 screenshots of what the OS actually looked like. None on their website or github.
I did just that with a Google Pixel 3a, but you should be able to do that with any Google Android Device. I actually compiled AOSP and LineageOS from source.
But to keep in mind, those all have proprietary Drivers.
I don't understand how they're still committing to shipping in Q3. From the last monthly update it doesn't look like they're all that close to a final release.
Ever since I saw the project come out, I've been thinking about what would be required for me to switch from Android. Right now I think that my hard requirements are:
- Firefox (pretty sure they have it)
- Signal (I use it all the time)
I don't know if some kind of TOTP store (think Google Authenticator or andOTP) is a hard requirement or a soft one, but it would surely be useful. I can live without my banking app (there's always the website); I can live without Uber or the Google apps; I can live without Kindle and all my streaming (could always keep a SIMless phone around to use as a media device).
As long as it has a browser and Signal, I really think it could be competitive.
Without a browser, it kind of ceases to be a smartphone. It's just a flipphone. So I don't see how that is really something you need to note.
For me it's Google Maps. There is just no comparison. I don't want to imagine trying to drive around finding places without it. It's not just the navigation and traffic, it's the business hours lookup feature and all that.
I hadn't thought about TOTP stores (Authy), but that's certainly way up there. Unless they can read a NFC yubikey.
See, the problem is, as long as Google is allowed to own services like Maps and YouTube, as well as Android, you've ensured you can't ever switch to a new platform: Google is extremely active in not only refusing to support competing mobile platforms, but actively blocking them from accessing their services, as they've done with Microsoft and Amazon's competing platforms over the years.
I've found for most map queries, Google, Bing, MapQuest, and HERE are all pretty much comparable/a wash.
It's weird that someone as tied to Google Maps would be looking into a privacy-oriented phone like Librem 5.
AFAIK, Google Maps will record your location/speed even while you are not using it (to get their traffic information), it will record any nearby wifi APs to help populate their aGPS database (so your private APs will get recorded along with their locations on Google's servers), and who knows what else.
The phone would be shipped with WebKit-based browser, Plasma Mobile does the same thing. When I'd have mine, first thing I'd do (or should start already!) is to help making a port of Fennec (current Firefox for Android) to Linux happen.
Right, Signal won't be on it at release, maybe ever.
They are recommending Matrix as a substitute, so I am getting used to that. Riot.im is the client, and Matrix is a little more capable than Signal (does interest groups), and doesn't demand your phone number or rely on Google Play services, so there's that. As usual, the problem is getting other people to switch, too.
A Matrix <-> Signal gateway server somewhere would help a lot of people make the jump.
While I really appreciate an enthusiast product (and is a proud owner of similar devices), I think Purism bit off a bit more than was necessary. Something a little bit closer to mainstream hardware would have gotten a device out earlier, and in the hands of more potential developers faster, with more time to implement whatever bits and pieces were unsatisfying.
The Pinephone has a fraction of the cost and have the potential to bigger developer mindshare. The specs are worse but that's not why you buy any of these things anyway.
Phones hardware have peaked and that makes it all the more interesting for someone interested in software and what it can do. Next year will be interesting.
I’m not convinced. None of the mainstream hardware I’ve seen could run wayland or X acceptably due to screwy graphics drivers and documentation hidden behind NDAs (yay Qualcomm.) If you’re going to bother putting work into something at least pick something that
1) isn’t actively fighting you
2) actually does what you want.
(Unless you meant SBC hardware, there’s plenty of IMX stuff out there not the least of which were the kindles.)
EDIT
It looks like Librem5 has a serial port [1] so you can build PureOS and flash your phone with a custom build. In that case you can configure the kernel and turn off SMT if desired. Nice :-)
I was referring to Symmetric Multi-Processing kernel support. Since the Librem 5 has a serial port via usb, the kernel can be easily customized to suit an individual's needs. If you mess up you can just flash the last known good. It is hacker friendly :-)
These specs are really impressive. I'm actually in the market for a phone upgrade, and the $700 price is not an objection, given what I'd be supporting.
What does concern me is the apparent lack of a verified boot option for the phone. I'm not sure, but it appears there's no trusted boot option to use signed firmware and OS. Android and iPhone have had verified boot for years now. Given Librem's focus on privacy and security, and assuming my understanding of the phone's current capabilities is correct, I'm surprised that Librem wasn't able to come up
with some solution, particularly given their innovative approaches to firmware security in laptops.
And I jumping to the wrong conclusion? Did they end up implementing a verified boot feature? Last I checked, it wasn't looking very likely. They had pretty much ruled out TPM.
Regardless, a big congratulation to librem. They've accomplished a very difficult goal. I'm guessing that if they don't have a verified boot currently, they're working on it for future versions of the phone.
Sad that this didn't end up using a low-power OLED screen. I can live with 720p resolution, but I'm surprised that for the cost an OLED screen wasn't possible.
I was hoping that such a screen on an open device would help drive the adoption of dark themes in applications for power-saving reasons.
Still looking forward to trying the device, though.
I prefer LCD. AFAIK, all OLED screens on the market use PWM brightness control in the 100s of Hz range, which for many people causes phantom array effect during eye movement. LCDs can have DC backlights, avoiding this annoyance.
What I have noticed is OLED's true blacks. It looks great when reading at night[0] -- it's just the letters floating in darkness. With a backlit LCD, it's a moderately-bright gray rectangle with letters in it.
100% duty cycle PWM is the same as DC, so you can mitigate it by setting the hardware brightness to maximum and scaling image brightness down in software to compensate. This isn't a good solution because it sacrifices color bit depth, causing banding.
My comment might be controversial, but for 649$, the hardware specs are felt like a smartphone of 2-3 years ago (or even more).
They could have invested in a octa-core phone and more RAM or a better rear camera. 13Mpixels cameras are there since the Samsung Galaxy S4, for instance.
When the phone will come out, I will be already old.
Sure, the final user isn't the typical iOS/Android one, but the lack of fundamental applications and a real app ecosystem should led them thinking of reducing their tag price.
I love the idea of a phone that isn't built by Apple or a spyware company, but I'd be worried about software stability and support. GNU/Linux on the desktop has always been flaky for me, and flakiness is not an acceptable characteristic for a phone.
It's great to see this project coming to fruition. My main reservation about
owning a smartphone has always been the idea of a device that knows everything
about me under the control of God knows who. This phone partially addresses
that issue, but can anyone here who's knowledgeable about encryption comment
on that aspect? Without some form of encrypted storage it would
seem like game-over when a well resourced adversary gains physical possession.
That is correct, in that case you need to encrypt either the files you want to be safe, or use full-disk encryption. I have no idea whether Purism will support selective file encryption or FDE out of the box, but it'd be pretty cool if they did.
That said, under the hood your phone is basically a Linux device, so doing this yourself might need some basic know-how of Linux administration, but shouldn't be too hard (regarding file encryption, FDE may be trickier).
Not everyone agrees with the privacy marketing of the Librem 5
> The Librem 5 is a phone that is falsely marketed as a secure and private device. This is not true.
> The Librem 5 does nothing for privacy. Installing a Linux distro is no more private than just using an ungooglified ROM. The mic kill switch is useless as they don't block access to the sensors so audio can still be gotten. The camera kill switch is also useless as you could just cover the camera with tape.
> They also weaken security. They use old hardware which doesn't have many hardware security features such as a hardware backed keystore.
> They aren't shipping firmware updates. Now your firmware will be outdated and insecure.
> Instead of using Android, they use PureOS which is a massive security decrease as you won't have kernel hardening, SELinux, seccomp filters, verified boot etc. like Android already has. PureOS doesn't even have a proper app sandbox.
One thing that isn't clear to me. The phone clearly supports GPS in terms of hardware, but does it actually do anything with it? It won't have access to Google or Apple navigation, and that is certainly something I use my current phone's GPS for.
PureMaps, Navit, GNOME Maps, Marble, TangoGPS, various OSM web UIs...
There's plenty of GNU/Linux apps, some of them providing online and offline turn-by-turn navigation, and some of them even already working well on mobile form factors.
The software will surely need some time in order to get polished enough to be used by a total newbie out-of-the-box, but it's not like there's no software for that at all.
I use OSMAnd on my phone, and am quite fond of it, for the record.
There are numerous other applications, as pointed out by other posters, that would be a better fit for the Librem. Most are using OSM as a map source, which I consider a feature rather than a bug (though I admit the map can sometimes be incomplete, the opposite is also true sometimes).
Can someone tell me why they opted for the imx8m with 4 A53 cores rather than the imx8 which adds 2 A72 cores (3-6x faster) and 2x the GPU performance?
A $35 raspberry pi 4 has way more computing power than this phone while being 20x cheaper.
I'm sorry, this device profile doesn't really sell me on why I need to upgrade from my Android TracFone; it doesn't even give me confidence that it will work with the networks I need it to. This is part of the problem I've found over the years with consumer Unix-derivatives - too many "figure it out yourself" answers, not enough "just works, or we'll show you how to figure out whether it works or not first". That's the big reason why Microsoft & Apple still are the default desktop OSes, I think.
They already have a list of LTE bands that will work depending on if you buy the Europe or US variant. That list is only for the Gemalto PLS8-US and PLS8-E. There is another Broadmobi BM818 module as well.
For the US, AT&T, Verizon, and T-Mobile support the 2, 4, 5, and 17 LTE bands that work with the PLS8-US.
All of Europe should be supported on the PLS8-E.
The BM818 mention doesn't specify which version. On the FCC's website, there is a manual that lists bands 2, 4, 5, and 17 so at least it will work in the US.
When buying any sort of unlocked phone, you usually have to check all of the bands with your carrier just in case. When you buying a phone from a carrier, they have already done all of the work for you in terms of checking compatibility.
I sure hope that page is meant to be a technical spec sheet and not intended for consumers because it's mostly gibberish terms I've never heard of to me even as a technical person. The frequency check site they link to also has a confusing layout, is US centric and I'm not sure what I'm supposed to use it for.
It would be much better if they could simplify the chart to display country, network and the variants of their phone. This is a very common problem with phone retailers and why I never order internationally. This is the final checklist item before I pre-order.
That doesn't help me since I don't know jack shit about bands. I went to their page and to the page they link to, but I still cannot figure out if the phone will even work where I live?
And if just some bands are supported, may that affect network speeds? I don't think this phone is going to very successful because at the very least they should have made some interactive guide where you simply can select where you live and it will simply display if it will work for you.
Unfortunately, as long as the standard response is dismissive like yours, that's what the majority will do. When someone expresses confusion, the goal shouldn't be to send them to the mainstream product, it should be to go back and explain your product more clearly. Stop being an elitist and start recognizing that helping people is a basic requirement of supporting a product or service.
When you have a piece of consumer-focused hardware which could help bridge the gap for mainstream users into a world of Free & Private, what you should be doing is actively seeking ways to bring them in. Besides that, it's an expensive commercial product. It needs to sell very well, or else it will make no impact where it needs to - in the mind of the average user.
Focus on the high-level person if you want. But have a simple version so you extract as much money as you can.
Thanks, that link was helpful. The US version will work on two LTE bands on my carrier in Canada. I was waiting to order until the radio bands were finalized.
Any mention of "just works" is going to be a lie. It's good to be honest about what you're selling, yet sellers rarely are. Just the fact they're being honest is a major selling point for me.
Oh, I agree about honesty as a selling point. That's the only thing that makes me engaged in this device. However, there's a difference between being honest in a technical sense, and seeking to solve problems from a consumer perspective without lying. And honestly, when it comes to phones, "just works" had better not be a lie, it should be requirement #1.
If Midwestern farmers could get a functional phone exchange working with a few handsets and fencing, we in the modern era shouldn't have more personal setup work than they did.
I have wanted this phone for months but right now I kind of wish somebody would do a crowdfunding campaign to get one of the postmarketOS phones to great support.
Pick a newer phone and a desktop , either plasma or phosh and do a crowdfunding campaign and just spend all the money on getting things working as perfectly as possible. Buy a bunch of refurbished phones of that model and sell them with the software pre installed.
Like purism you could bundle it with a subscription or partner with a company and take a cut.
Looks like the best option is the EU radio if you're in Australia. However, that won't support Telstra's 3G, which might cut you out of voice calls. If you have the option of provider, go Optus, that will give you most of the 3G and 4G bands they have on the EU Radio. US radio only supports Telstra 3G and Vodafone 4G, and nothing else.
It's rather unfortunate for those who travel as well. Most mid-tier or high-end smartphones on the market today seem to work fine for LTE and voice calls on nearly every major carrier worldwide.
I like they use main processing unit from a Dutch company; is its ROM open source or/and can be accessed?
All ARM-based Sytem-on-Chips do have a proprietary ROM, which is used mainly for boot-up and multicore management. But there's no access to check if it may do something else.
>> The mobile baseband will most likely use ROM loaded firmware
The immutable ROM that finds firmware and jumps to it is indeed usually proprietary, but since it's tiny and immutable, it's considered "part of hardware" and is okay even for the FSF "Respects your Freedom" certification.
The one mutable/loadable blob the i.MX8 requires is firmware for the DDR4 PHY (not training code you run on the CPU, but something that runs on the PHY apparently??) and they're using an embedded tiny cortex-M4 core to just load it: https://puri.sm/posts/librem5-solving-the-first-fsf-ryf-hurd...
(IMO, that's a silly requirement from the FSF, I think keeping the main CPU free of blobs that it doesn't run i.e. blobs for other things does not have any meaningful benefits)
> (IMO, that's a silly requirement from the FSF, I think keeping the main CPU free of blobs that it doesn't run i.e. blobs for other things does not have any meaningful benefits)
Depends on how privileged the peripheral or subsystem is. And what data it is trusted with.
If it can't harm the main CPU and is properly confined then it's fine in my book.
I wouldn't pre-order something like this because I don't yet know if it will be reliable. I see the value in what they are offering but it isn't battle tested yet.
I don't want to relive the early smartphone days where the phone crashes when I'm trying to answer a phone call.
I think this first model is for enthusiasts who can afford it and want to fund the development. It won't be perfect yet. If you think about it as a donation it makes a lot more sense.
So we knew about the Gemalto PLS8, and I noted as someone on Verizon that the PLS8 wasn't approved for Verizon but other variants/similar Gemalto chips were. Still no word on an option for those, but the Broadmobi BM818 is new, and I haven't found much about it yet.
Most results on search link back to Purism, so it may not be a widely used chip. It got FCC approval last December, and the documents there say it supports LTE on bands 2, 4, and 13, which would technically make it capable of supporting Verizon? But it isn't on Verizon's approved modules list, so it's likely it couldn't be activated at this time.
I'd be curious what Purism's thoughts were on the chipset choices and what networks they're supposed to work with.
It will be curious to see if these will be vulnerable to electronic device interceptions with the NSA TAO[0]. I'd imagine they are already working on building a replacement PCB to drop in and image.
There's alot people including journalists, whistleblowers, high level execs, people being targeted(T.I.'s) etc. would easily pay triple MSRP for a secure phone from government entities. It would be interesting to see if they can do it on a software level anyways.
The key question here is : can the phone be booted from the removable storage card ? Without such final control of your data, you can be pressed to decrypt your internal disk by any repressive acting party - and those are rising steadily. Hence the only save way to give users REAL control of their data, privacy - and safety from torture, repression etc, is a removable storage which holds the OS and user data. Anything less looks like cosmetic pseudo privacy and security.Can Librem boot the OS from the sd card, then ?
I saw a demo of the test board on video and the touch interface latency was very high. is it the hardware or the software? has it improved? It was quite frustratingly slow by the looks of it.
This phone has a higher screen resolution than my laptop. It also has almost as much memory, but I guess that's a prerequisite for running Gnome. Why didn't they go for Xmonad?
I have put my root folder on a 32GB M.2 in my X220 and that has been more than enough storage for all of the different distros I have run the last 5 years.
I expect to replace my current laptop with the Librem 5 as my daily driver. Having a real linux-distro on my phone is a dream come true, especially when I can dock it with my monitor at home and use it as a regular laptop.
Same here. I've been occasionally hitting 32GB with my / when I let my system grow really messy, but a quick look at big and unused packages was enough to clean it up. For a GNU/Linux system, 32GB is a lot.
It would still be nice to have more for added comfort, but it's not like it's a big deal.
It isn't about the hardware, and it also isn't about PureOS itself. It's about people caring enough to build an entirely new mobile operating system which actually respects your privacy and gives you tools to ensure it.
I'm really amazed at what they got running in the few years this project existed. Of course the phone won't be some spectacular mobile gaming machine, but that's not what I paid for. I paid for a privacy-respecting, secure phone. I paid for getting Linux to run on mobile OSes. That's what is important to me and many others.
Same here. I'll gladly downgrade some specs if it means I can run all free software on my phone. Trying to get google off of an unlocked android phone is still an uphill battle. Root access out of the box means I can install the software I want and remove software I don't (a novel idea).
The Purism team has accomplished an impressive feat with the Librem 5. Porting GNU/Linux onto a smart phone is a huge win for free software. I hope the transparency and upstream patches continue long in the future.
When I pre-ordered the phone, I also saw it as donating to (or at least contributing to the development of) open-source software in a way -- Purism is a for-profit company, but they've been doing a lot of software work on this phone and they've been pushing the changes upstream, too. I highly recommend looking through their blog (https://puri.sm/posts/), especially the Librem 5 [month] Software Update posts for the work they're doing.
It's worth mentioning that they have a strong commitment to their social purpose, not maximizing profits, which I really like. Read more about it on their webpage: https://puri.sm/about/social-purpose/
It's better because it uses mainline open source Linux drivers.
Your Android phone has drivers written for a specific kernel version. The userspace software can be kept up to date for a little while by making it work with the old kernel, but when applications eventually start to depend on newer kernel features it's time to chuck the phone in trash and buy a new one.
It's a small (2.5k to 3k) batch and unusual tech specs (i.e. last time completely detached from SoC cellular/wi-fi modems were made in 2012 according to Purism). If there was a demand for 100k batch, it would be cost closer to $350 I believe.
That makes sense. I guess there is always a community of Early adopters with disposable income willing to pay high dollar to play with the phone and contribute to it's development. Guess my idealistic mind wants to shortcut to the part where the phone is in an iPhone hardware/UI polish level with a linux OS solution, at which point I'd be happy to pay top dollar for it!
Whats the software selection going to be like for this phone?
I bought an Ubuntu phone, and while the phone was nice enough, the lack of apps made it a bit crap. No Whatsapp (I was happy enough using telegram, but my friends were not). The maps application was web based, so didn't launch unless you had internet connection.
The only real obstacle for me is that a large fraction of what I do with my phone is listen to Spotify. It's not clear to me whether or how well it will be supported. (More research: looks like I would have to install Ubuntu Touch and a third party app?)
You probably won't have the option to store your playlists locally though, and streaming music with mobile net could get expensive. The other big apps without good mobile web versions are Whatsapp, Signal, and Telegram. Those could be show-stoppers for some people.
Telegram Desktop would run fine on Librem 5 on Plasma Mobile (magic of QT) and clunky on Gnome (but I might be wrong), as well as APKs on Android emulator like Anbox.
But it is pretty expensive. There are Chinese phones with 4 cores, similar amount of memory and same screen size that are 4 times cheaper. Would not it be better to take any of them and replace the firmware instead of making their own platform?
I mean sure? But on the other hand it still costs 650$ so if they'd charge me double for the privacy, it should be comparable to at least 300$ which is not
The response to that has always been, people will pay the Apple premium because to them it's worth it, for the camera, for iMessage, for ease of use, etc.
Why can't the same principle apply here, but with respect to user-toggle-able hardware switches and replaceable batteries?
What is the main thing you take issue with? The filename? The fact that the picture is not directly related to the project? The fact that the woman in the picture is attractive?
I just found out they didn't use AOSP. Ok so it's based on Linux but isn't anyone worried that they're re-inventing some wheels by not using AOSP? Or is there a licensing issue I'm not aware of?
Edit: Sorry but my cynicism stems from backing a lot of open source attempts, including N900 with Maemo, ZTE Blade and Jolla. I ended up giving away two of those and the last is rotting in storage.
No, AOSP reinvented the wheel and gave us a broken mess as an excuse to push us into using google software. There’s nothing the phone really needs that hasn’t already been written, the only reinventing is them writting yet another palm pilot home screen knockoff.
For the people complaining about the price/specs ratio: this phone is no standard off-the-shelve system, it had to be designed with user privacy and openness in mind while having a ridiculously small market. I'm amazed they did it so cheaply.
Personally I hope that they will be able to sell enough to continue development. Maybe this will lead to a second generation with multiple models for different market segments. I'd love a librem phone with a good old blackberry keyboard in the <1500$ price range with a little more RAM/CPU power and screws holding the back plate.