Librem 5 phone shows the cost of being different
arstechnica.com
arstechnica.com
“Let's assume the project is over-promised and will be late and under-deliver. […] None of that is as bad as NO smartphone alternative to Google and Apple, let alone an open and privacy-focused alternative. I'm a Linux user and I very much want a Linux phone. And there has been so, so many false starts and near misses over almost a decade. My patience and goodwill for Purism and the Librem 5 is effectively infinite. It's not over until they shutter the doors or can the project. Until that happens I will continue to hope they limp over that finish line no matter what.”
1. https://www.reddit.com/r/Purism/comments/dm2smy/is_the_libre...
Neither of these are completely in the open but all three are probably good enough for making an enthusiast operating system run. They are all variations of relatively well known, sometimes even documented, hardware, with lots of necessary peripherals with non-free firmware blobs.
Some make efforts to only communicate with these non-free peripherals over standardized protcols, which is the "good enough" part. For community oriented mobile operating systems, 2020 is probably going to be the best year yet.
Price is also a factor. Many more people will tinker with a $200 phone than a $800 one. It's probably better to cut corners with your first model in order to hit a price point that gets enthusiasts on board. Raspberry Pi did that and have gotten incrementally more open with time.
It's fun to install something like XFCE and have a full blow desktop environment.
For $150, replaceable battery, SD card slot, privacy dip switches, and open source the value is more than there.
Disagree. Viral is more important for a new product like this. Which means beating naysayers over the head with things working undeniably well.
See: Tesla Model S, Surface tablets made by MS themselves, Google's Nexus line of phones.
All bad things to compare to. In all of your examples, these lines were all established long after the companies entered the market. This is a device trying to create a new ecosystem from a company that hasn't produced a phone before for an ecosystem that basically doesn't exist yet.
Right now it should be seen as a prototype (and it is). But whereas Google, Tesla, and Microsoft have large numbers of engineers these projects rely quite a bit on community support. Imagine if Tesla launched the Roadster but couldn't service it themselves, instead relying on independent garages to figure out the best ways to work out servicing and bugfixes?
For these projects it's better to keep the specs modest (less thermal issues/faults), the price affordable (more developers can afford to purchase one), and the perception of support good (so people aren't on the fence because the manufacturer calls it quits or goes out of business a year later). It's especially important to keep the price low enough that the first models could be seen as hobbyist/secondary phones because making a phone ecosystem is really friggin hard.
Viral works when you're trying to sell a videogame, not when you're trying to create an ecosystem from basically scratch (or even just carve out a niche). No matter how Librem tries to market itself, it's a niche product with limited appeal especially regarding their price point. In this sense, too much attention from the wrong kind of consumer at this time will be their undoing.
Yikes. In my mind, this isn't actually a phone in its current form. It's like a small desktop computer put into a plastic case with a battery.
I have never seen the case up close until this point. It truly looks like a crude hobby project, almost to the level of being 3D printed.
I have no idea why Purism didn't leave most of the hardware design to someone who already has a generic modern smartphone done and completed, then they could have added on to it by swapping out some proprietary bits. Instead we get this Frankenstein of a project, sold to the public long before it's done (along with marketing copy that makes the company look unethical - there should be big warnings on the store page about how far along this project truly is and what you're getting for your $750).
The worst part about this is that it makes all open source projects look bad. It makes them look like a bad compromise. And that's not true at all - a lot of open source projects are great alternatives, and many are even better than their proprietary competitors.
$750 gets you an iPhone with sophisticated cameras, the fastest mobile processor on the market, and software support for 5 years if not longer. Walk into stores across the country to get service and support. What Purism is telling us is, essentially, if you want to take a stand against the proprietary world, you have to give up everything. You have to use a chunky, ugly, crudely assembled phone. You have to give up acceptable power management. And you still have to pay $750, all of this on the cusp of Apple releasing a new $399 phone according to rumors, Google selling $399 phones, and budget brands like Motorola providing an absolutely acceptable experience for under $200 on some models.
Either you didn’t actually read the article or you’re being intentionally obtuse. They very clearly state the reason that they couldn’t use the regular supply chain. All the usual suppliers refused to work within the “confines” of OSS.
They were lucky to find a CPU that works at all within their definition and needs. If you read the article you’d know that the author spends the entire first page explaining the compromises that were made in favor of OSS. There is no supply chain for this sort of project. Purism is the first company to actually attempt it.
Look, I have no horse in this race, I’m a happy Win and iOS user, I have no plans on changing that. But I commend the effort and definitely believe in the mission.
OSS hardware is a compromise right now! That’s just a fact. People like you who can only see the flaws and refuse to understand how we got here are unhelpful.
If you read Purism's marketing on their website, read the blog posts, and the verbiage on the store page, and then you run out and buy this phone, you'll be under the impression that you're getting something much more close to functional than it actually is.
Essentially, this product should be labeled as an early development kit. But the marketing on the website is even trying to sell this to parents, who should absolutely not buy this phone for their children (because it's not done and it'll be dead anytime you try to call them because the power management is broken).
If this doesn't matter to you, fine, but ignoring it is disingenuous.
To be exact, Purism isn't really the first - however, earlier successful attempts were made long enough time ago to be hardly relevant in today's context (like Openmoko) and/or very limited in their scale (like GTA04).
You could get this for $750, and it wouldn't be open source. That's the whole point of this project. Nobody is buying a Librem who doesn't care about freedom.
Mainstream devices are cheaper for a reason: ecosystem/app store lock-in and (in the case of Android) capturing your data. Economy of scale is a factor too.
Choosing open tech shouldn't be a sacrifice, but it usually is. There's less money coming in, so the end experience will never be as polished.
That means $550 is what the Android collection of companies is willing to pay to acquire your personal data stream.
(And a price at which they think they can make a profit off you at)
To be fair, when it comes hardware manufacturing, that's pretty much the state of affairs. They had to give up almost everything, and they have no choice but passing that on to you.
The PinePhone is a competing project with slightly lower specs, similarly good FOSS purity, that is slimmer, ships today and costs 1/5 of the price.
They clearly market it as being for tinkerers at this point and all the software support Pine64 is working on is getting the kernel running, leaving the rest up to the 5 (!) involved distribution communities.
I have received a PinePhone (early "BraveHeart" batch - the regular one doesn't ship yet, mind you) a few days ago. Although the specs seem "slightly lower" on the paper, I think it's important to note that Librem 5 has twice as fast RAM bandwidth and its GPU easily goes rounds around the PinePhone's. The eMMC is also around 30% faster on the Librem 5, and its killswitches are actually usable in everyday scenarios. The difference is quite easily perceivable when using both devices with the same software. It is a pretty neat device, sure, but I think it's important to be honest about how those phones compare to each other.
Also, PinePhone requires firmware blobs being stored on your rootfs for WiFi, Bluetooth, and (optionally) camera autofocus. Librem 5 has them stored on separate flash, making the peripherals that need them self-contained as per FSF recommendations. Both devices are pretty FLOSS-friendly in my book, but some people do care about that difference.
Librem 5 also has a separate GPS module, while PinePhone uses the one integrated to the GSM modem.
(PinePhone BraveHeart edition also has some sensors connected via the modem, but AFAIK that's going to change in later revisions)
Also AFAIK in some cases FSF even recommends storing firmware blobs on a separate ROM, so that they can't be updated. This apparently somehow makes the proprietary fimware an unchanging component of the hardware of the device and thus perfectly fine from the FSF perspective.
How this prevents security updates for the firmware or the use of free software firmware if it is developed in the future seems to be less important.
At least Purism has had the kernel booting for some time -- the i.MX8QM has had mainline kernel support from NXP for years now.
I don't think that should be encouraged. If you follow their blog and product description page and then place an order you're just about hoodwinked into buying something that's claimed to be at least mostly working.
I don't think that's ethical, even if the hardware and software itself is.
At $800, I might buy one, but it has to be really good.
Librem 5's price, on the other hand, finances a whole team of software developers that develop things people can (and already do) also use on their PinePhones. It also helps to move the SoC support forward, as PinePhone mostly relies on already existing sunxi efforts.
If you can't afford a Librem 5, get a PinePhone, but if you're explicitly about getting "closer to your dream" then Librem 5 seems like a much better option when it comes to showing support.
I'm not sure where does that come from. PureOS is a very lightweight Debian derivative which is built around regular GNOME. Majority on the work happens directly on upstream and benefits everyone even on desktops. Even parts that were actually built from scratch, like phosh, are being built with interoperability and reuse in mind (you can run phosh on any layer-shell compatible Wayland compositor, like sway; and run phoc with other shells as well, such as MATE). Also, libhandy is meant to augment existing Gtk widgetset, as opposed to some other efforts like UBports or Plasma Mobile that require you to completely rewrite your applications to use new QtQuick frameworks, as there's no effort being made to make regular Qt Widgets usable on phones.
Vast majority of the work put into mobile PureOS is just improving Debian and GNOME to work well on mobile.
(PS. Just today I've seen a community member sharing a video of Maemo Leste running on the Librem 5 on Matrix... ;) postmarketOS with various shells, including Plasma Mobile, also works already, and UBports developers already received their devices, so we may see it running there soon as well)
Then in comparison, there are libraries that have been used on mobileLinux devices for a while and even open source native applications (mostly on Plasma Mobile & Ubuntu Touch, Sailfish OS apps are still sadly mostly closed source).
Not making use of and extending all the mobile friendly software almost seems like starting from scratch to me.
BTW, what telephony midleware is Pure OS using ? Ofono, FSO or something else ?
The telephony middleware is a great example of that - it uses ModemManager, which is a part of a regular desktop stack and already well-integrated in GNOME (and other DEs) for many years. Thanks to that, you can just plug a USB 3G/4G stick to your laptop, install Chatty and start SMSing - just like that, plug'n'play (you may even be able to make calls if your modem exports audio over USB, but it's rare).
I've spoken with multiple ex-purism employees who were laid off before the librem5 shipped. Their comments largely echoed what's in Zlatan's Phoronix interview. [0]
At the moment I'm inclined to support Pine64 over Purism. We're not so desperate for a linux phone that we should throw our resources behind poorly run, misleading companies.
[0] https://www.phoronix.com/scan.php?page=news_item&px=Zlatan-T...
Unfortunately, the case is just awful and there's no excuse for this. Low standards - as seen in many consumer-facing FOSS products - must be to blame here.
Did the author actually make a phone call? If so, what was the quality?
FTA : "There's no way to access the camera."
So no idea if the device actually will take a photo or video or know what the quality of those outputs will be.
Considering that an iPhone 11 Pro would cost $60 per month over 2 years, could some of us Patreon this project at like $30 a month and let them hire up enough to land this?
I have doubts that a sales model vs the incumbents is going to have enough runway. Given the quote from teddyh / whiteinge, maybe it's just better to do more upfront funding instead of expecting open projects to follow the same physics as venture capital.
You can also start a text via contacts too now. There's been a lot of progress in a month.
Source: I have a Burch phone, and made a phone call.
I don't know if you realized this, but that's exactly what they did. They made a kickstarter on their website to finance making the phone. They released some of the hardware iterations to earlier backers, and made sure they understood that its a beta product now.
I'm also a little curious why they didn't choose an open android fork.
This isnt true - a faster phone will reach idle quicker and thus use less energy to accomplish a given task, such as rendering a web page. This particular phone also uses a CPU manufactured at 28nm which is much less power efficient than the modern 7-10nm process current smartphone CPUs use.
I wonder why they're able to be sold more cheaply?
But if you live in a different country, your only choice might be through reshippers/resellers, who will root the phone and install their own malware onto the system partition for $1 in profit, seal it back in its box, without the factory's knowledge. This happens somewhat-frequently.
Some problems mentioned in the article have been already fixed, by the way.
I think when they get to mass production batch, the Librem 5 will be a usable replacement phone.
Assuming they upgrade to the 3500 man battery, it could reasonably go all day.
> ... It is hard to do something truly different in the smartphone industry. Companies, especially smaller companies, are all working from the same parts bin with the same manufacturing partners. You take your Qualcomm SoC, your Samsung display, and your Sony camera sensor—and you take a flight to China and visit Foxconn, which, in addition to manufacturing, will even do engineering for you if you want. Smartphones are so samey because they have an established, for-hire supply chain that has a certain way of doing things, and it's much cheaper, faster, and easier if you just "go with the flow" and do what everyone else is doing.
> Big companies like Samsung and Apple have enough money, control, and connections to move the supply chain in whatever direction they want ...
> ... If you haven't noticed, open source smartphone hardware is not a thing that existed before now. There have been phones that run open source builds of Android, but those are full of closed-source firmware from non-open components. The usual hardware companies cautiously guard their hardware designs and drivers, and Purism's hardline stance on open source has ruled out almost the entire established smartphone supply chain ...
> ... "When we first approached hardware manufacturers almost two years ago with this project most of them instantly said 'No, sorry, impossible, we can not help you'." Others warned us, that it could never work, that it was too complicated, 'the industry does not do that,' and so forth."
> The thing to keep in mind here is that Purism has taken on an absolutely gargantuan task. It somehow scraped together a new supply chain of mostly open source components, it came up with a smartphone design from scratch, and it is building its own smartphone distribution of Linux. Two years is not enough time to do this. The OS and app package is not nearly finished, and it lacks basic smartphone functionality. The hardware is nearly finished, but you'll have a hard time taking advantage of it right now since the power management isn't really implemented, and support for things like the cameras are non-existent. If you really want open source smartphones to be a thing, though, this is where you need to start. The Librem 5 is a proof of concept.
> "Open source at any cost" really feels like the mantra the Librem 5 was made with. That cost is going to be significant, since 1) the need to go out and build its own supply chain means Purism is doing this with zero economies of scale, and 2) the company needs funding to build the OS. Get ready for some sticker shock: the Librem 5 is currently $750. That's up from the $700 early-bird price it had in 2019, and when the "Evergreen" 1.0 version launches, the price will go up again to $800. This is high-end smartphone money—and remember, you're not really buying a finished product. The Librem 5 is only for true believers in the idea of an open source smartphone ...
See, this is something I really dislike about the Librem 5 - they are pretty much trying to rewrite history to pain themselves in a better light. "We are the first ones tryin to do this, no one ever tried before, so all those bad decissions we did that make our phone unusable are not our fault!"
This is of course incorrect, as they totally fail to mention the Neo FreeRunner project, a open source smartphone from ~2009:
http://wiki.openmoko.org/wiki/Neo_FreeRunner
I've had it and many issues it had pretty much mirror the issues Librem 5 is having at the moment - problematic hardware component, buggy calling and issues with power management. Mybe thats why Purism fails to mention it, to hide the fact they managed to repicate the mistakes 10 years later ?
Still even http://wiki.openmoko.org/wiki/Neo_FreeRunnerwith all its faults the Neo FreeRunner pretty much jumpstarted the open mobile community, with many of its users then bein involved in the Maemo and Sailfish OS communities and now with the Pinephone or Fxtec Pro1.
For that reason it makes me sad how Purism fails to mention the Neo FreeRunner & paint themselves as being first. It's really rude to all the hard work the open mobile community did since then & that they are heavilly building on.
I don't agree with your stance about "replicating the mistakes 10 years later". The issues GTA02 had managed to make it past four earlier hardware revisions and reached the mass production (and there was even GTA01 before that!). Librem 5 isn't even mass produced yet - you had to opt in for an early development batch to receive one already. It's currently at its third revision, with the fifth one being scheduled to be the mass produced one. To be frank, so far, having the Neo Freerunner in the back of my head, I'm pretty happy with the progress on the hardware.
Classifying things like power management into "repeated mistakes from the past" isn't really fair IMO. Those things just take a lot of time and effort, and taking the notes from past experiences can only smooth some edges, but won't make that work magically done for you. The hardware is capable of decent power management, as the NXP's kernel tree can show, but the mainline Linux has its power management still at its infancy. It's going to get better with time, but it's going to take that time - we can't magically skip that.
There are also business constraints to consider there - it's easy to say "you should have waited with release until everything is fleshed out", but it's harder to actually finance a team of developers when you don't deliver anything for extended period of time ;)
As for being hard and needing multiple revisions to catch critical bugs - definitely agrew with that as well, even major manufacturers are no immune to hardware issues - Sailfish OS hit on Xperia X and likely 10 as well an issue with the proximity sensor becomming unreliable if turned on for too long. While it could still be a driver issue, given that official Android firmware has the same issues it looks more like a hardware problem Sony failed to catch.
Still, even if all the current issues woth the Librem 5 can be fixed in the final revisions, we are still hitting on marketing issues in how Purism "sold" the initial batches - eq. I don't think the expectations have ever been set as getting an expensive hand warmer, rather having just a minor issues here and there and possibly a less nice chassis.
In my opinion, these hardware revisions simply should not have been released to paying customers even if it would mean another delay.
Do you get autocorrected from Fxtec Pro1 ? I've not heard of Foxtex, ducking for it appears to show speakers and monitors, but not phones.