Re: librem-5-smartphone-final-specs-announced

I think this is the right framing. A lot of people would like privacy, but they don’t have the time, energy or knowledge to try and keep their privacy online, and the current hoops that you have to jump through are so high to maintain your privacy, that even people who care about these issues rarely do them. If privacy means giving up social media (Facebook, Twitter, etc.) and all the convenience and easy communication from online services, then most people decide to throw out privacy. If companies start offering convenient ways to keep people’s privacy, then the market size is potentially enormous. The market for people who value software freedom is smaller, but it is probably also in the millions if it were a competitive option.

We are the early adopters, who are willing to pay a lot for a bulky phone with lower specs and we are willing to deal with all the hiccups of a developing platform that will have very few apps and probably some bugs along the way, because we really value privacy and freedom.

However, as hundreds and then thousands of apps are added, and Purism improves mobile Linux as a platform and improves its online services and the hardware specs improve in subsequent versions, it will become easier and more convenient for people to start switching. The good thing about Purism is that it is trying to create a new ecosystem of mobile Linux and privacy web services based on free software, that will grow over time and attract more users. If Purism can prove the market, then other companies will jump in and drive down prices and offer more apps and more online services based on privacy and freedom. The more people who are using mobile Linux, developing GTK and HTML5 apps, and running Mastadon, XMPP and Matrox servers and using encrypted email, the more investment and development will be poured into the ecosystem, and the momentum will grow until we have a viable alternative to Google, Facebook, Twitter and surveillance capitalism in general.

I am very appreciative of the engineering that is going into the Librem 5, but I didn’t put down $599 in the crowdfunding campaign because I thought that the hardware in the Librem 5 was worth $599 or now $699. The most I ever paid for a phone before the Librem 5 was $210, but I want to live in a world where people can have digital rights, software freedom, and online privacy, so I’m willing to pay 3 times as much for a phone if the company making it is working toward those goals.

I want the hardware to be good, because it will attract more buyers, earn better reviews and generate more buzz for mobile Linux and grow the ecosystem, but I would have pre-ordered the Librem 5 even if it had worse hardware specs. I think the long game to create a new mobile ecosystem is far more important than worrying about how the Librem 5 compares to the OnePlus 7.

9 Likes

assuming that one is also able

For some people they may be 100% privacy lovers but if they don’t have the $500 then they don’t have the $500 no matter how much they love privacy.

Noone has to give up Faceborg (etc.). You just wouldn’t be doing it from the app on the phone at Day 1.

Having a Librem 5 and still using Faceborg on e.g. a laptop isn’t necessarily a bad thing to do because you are still isolating phone content and phone location etc. from Faceborg.

In any case the essence of freedom is to be able to choose what you do and don’t use or share - even if that choice is not fully consistent with some principle or fully logical or fully wise.

If someone wants to (and is able to) write a Faceborg app for the Librem 5, that’s fine with me - but I won’t be using it.

WhatsApp maybe?

2 Likes

Ok, time for a little bit of deflating realism here. And keep in mind I truly hope there are enough of you happy people willing to buy this phone so it will be the first in a line of many…

This phone is not cheap, and I don’t think the right sacrifices were made in its construction. For example, why would I need a removable battery for “future-proofing” if the phone is going to already (assumption here) going to be cruising by with low-end hardware? And why would you give a 5.7 in screen just 720pixels to work with? I could understand if it was a small 720 screen, and I wouldn’t mind…but at this size you could probably count the pixels easily. 3Gigs is acceptable, and I don’t care about the 32Gigs as the phone does have expandable storage. I kind of wish they hadn’t bothered putting in a camera, or at least not two. With these specs, I highly doubt the picture quality will be more than almost OK.

Being a privacy-conscious linux user myself, I wish these specs were at least current-proof, much less future proof. I buy a mobile device once every few years, and since the prices of mobile devices began to skyrocket, I stuck to the highly competent low-to-mid tier stuff. But honestly, this makes me want to look for some slightly used oneplus X instead (which was half the price with better specs when it came out a few years ago). But with these results, I think I’d rather continue to limit my mobile usage and wait for something else.

I truly wish purism the best, and hope you guys are enthused enough to support them, because devs should get paid. However, I really hope they release a higher-spec version for not much higher price within a year, because otherwise I don’t see this becoming more than a verrry niche product.

TL;DR: I think Purism dwindled a niche-audience into a significantly tinier niche.

1 Like

Because the imx.8m chip is not very power efficient and to my knowledge was not meant to be a mobile device and 720p would be less taxing on the battery life .

Second why does everyone keep harping on the processor ?? I mean are you going to have a few MS slower touch response time ? sure probably . Are apps going to open a few ms slower than a more capable chip ? Sure .

But linux is linux . This is going to be the same OS in 5 years it is now just with a bit more functionality and security .

And anyone using this device as a workstation probably isnt going to be doing heavy video editing or any enterprise level graphical work .

I think youre going to get this device in hand and be pleasantly surprised when you get to using it .

Hell I remember getting my mytouch 4gslide back in the day that I remembered being fast AF with 800mhz dual or single core chipset . Add that to the fact linux is in no way as resource intensive as android and … I really think this device is going to be good out of the box .

All that snapdragon 855 with 12gb ram you just dont need all that on a phone . 90 percent of people surf the web watch some youtube and text and call and the L5 should do all of that flawlessly on a properly rendered device.

6 Likes

If you wait until the phone is actually released, your comments could be informed by facts, rather than be speculation.

I have no idea what the picture quality will be, or how it will compare with my current phone. If I want better photos than my current phone, I use my digital camera but, a lot of the time, I don’t have my digital camera with me and I do have my phone with me. So, therefore, I am glad that they did put in a camera. I expect that if they had no camera at all then people would criticise the phone for not even having a camera.

I really hope that demand is enough for the upcoming phone such that Purism is encouraged to develop subsequent models. The future will tell.

4 Likes

That’s not how resolution works. 720p is 1280x720 pixels, so 921,600 total. For a 5.7" screen, each pixel would be just under 0.0039" square. The angular resolution of a circular aperture (e.g. your iris) is given by the Raleigh criterion. The angular resolution required to resolve an individual pixel is 0.0039/d, where d is the distance from your eye to the phone.

The page I referenced gives the resolution limit for most people as 5e-4 rad and for completely optimal conditions as 2e-4 rad. So most people would have to hold the phone 7.8" in front of their face to see a pixel, and Carrots McGee, who has record-breaking vision and is sitting in a well-lit room with the phone brightness on maximum, would be able to see individual pixels at 19.5". These numbers are slightly smaller without rounding, but let’s be pessimistic and go with these.

Now, let’s see what Purism gave up by forgoing a 1080p screen. On a 5.7" 1080p screen, pixels are 0.0026" square. The minimum viewing distances for John Doe and Carrots McGee on the flashy battery-hungry 1080p screen are 5.2" and 12.9", respectively.

So, if you have world-class vision and really need that 13" - 19" viewing range, then yes, you’re missing out by having the lower resolution screen. For everyone else, there’s no practical difference.

Edit: as @kieran points out, the actual screen resolution is better than 720p. Mininmum viewing distances for 720x1440 resolution are 7.1" and 17.7" for normal and optimal conditions, respectively; the viewing distance window between “HD+” and 1080p resolution is less than 5" for even the most sharp-eyed people in the world in optimal conditions.

7 Likes

In fact, the final specs say 720x1440 i.e. 1,036,800 pixels.

Which by my calculations makes the size of a pixel 0.0035" x 0.0035" i.e. slightly smaller still.

Let’s wait until we can hold the phone in our hand before making bold claims about whether we can “count the pixels easily”.

6 Likes

Government and big companies need surveillance and thus will defend it. We do not have a free market. The government can easily give companies or industries an unfair advantage by giving them free money, making children use their products in school, changing laws, making certain products necessary for jobs, etc. Thanks to this privacy will not get convenient any time soon.

If you want more people to go for privacy, then that can succeed. But if you want most people to go for privacy, then you might need multiple lifetimes.

I did not say that being a big privacy lover was a sufficient condition, just that it was a necessary one.

Personally I was doing fine with the N900 not too long ago. And even now it is the abandoned software and lack of freedom that bothers me about it, not at all it’s hardware. Phones do not expire just because higher-spec phones become available. The L5 might be useful to me 10 years from now just like the N900 was useful for me around 10 years after it’s release.

I agree with this. Privacy-lovers with a lot of cash and who do not care about specs is a pretty small niche. Though it is also a niche that is pretty much dominated by Purism right now.

as Captain America said “then we will do so together” and we will learn to fail better.

Please fix FAQ.
In https://puri.sm/faq/what-are-the-phone-specs/ mixed up Front and Back Cameras

1 Like

The front facing flash makes it a good devices for pranks though.

1 Like

Your preferences are also a tiny subset of the market and would probably result in fewer sales of the Librem 5.

We can debate endlessly what niche of the market Purism should have prioritized in its design decisions. For example, people who value security are delighted by the Smartcard reader for an OpenPGP card, but I suspect that many buyers would have rather have 64 GB of Flash memory in place of that Smartcard reader. Some people clearly would have been willing to pay $150 more for the Librem 5, if it had a 1080p screen, 8GB RAM and 128GB of Flash, but would that generate more sales than the current specs?

In looking at Purism’s design decisions, it appears that Purism deliberately chose to include features that no other phones provide, in the hopes that people who really care about those features would be willing to pay a premium for lower spec hardware in order to defray the high development costs of the Librem 5. Look at its unique features:

  1. First phone in the world with a Smartcard reader for an OpenPGP card for greater security,
  2. First phone in the world with hardware kill switches and the ability to turn off all sensors at the circuit level.
  3. First phone in the world with a replaceable cellular modem,
  4. First phone to not require any binary blobs in the kernel and get the FSF’s Respects Your Freedom endorsement.
  5. First Linux phone since the Nokia N900 released in 2009 that can run standard GTK/GNOME software (UBports has problems).
  6. Only 6.6% of new mobile phone models in 2019 (16 out of 244 models) have a replaceable battery according to the gsmarena database.
  7. Only phone currently on the market that can provide convergence as a desktop PC without special hardware like Samsung’s DeX docking station or Asus’s PadFone tablet shell (Nexus 5 and Fairphone 2 with UBports and Windows Phone were capable of convergence without special hardware, but they are no longer for sale.)

The strategy appears to be serving niches where people are willing to pay a lot for a lower spec phone, so that Purism can pay for several years of software development and specialty manufacturing.

Unfortunately, this strategy doesn’t appear to have worked for Nero, but it will work for a lot of niche buyers. I’m a good example, since I don’t care that much about processing power, but I do care a lot about the environmental impact of planned obsolescence, so I really want a phone with a replaceable battery and cellular modem that can be supported for a decade by the Linux community.

A lot of what Purism can do in the future is constrained by the SoC. Purism can’t offer a stellar camera, because the i.MX 8M Quad doesn’t have a DSP or ISP and it can only handle 5 MP at 15 fps, so it will have to add another chip just to process the camera’s output.

The Vivante GC7000Lite GPU can probably handle a 1080p screen, but a 1440p or higher screen will probably be too much of an energy hog. Also, a 1080x2160 or higher screen means that the DCSS interface will have to be used for the Librem 5’s screen, which means that the eLCDIF interface which is limited to 1920x1080p60 will have to be used to run the external monitor. The last kernel submission from Purism that I can find shows that the Librem 5 screen is using DCSS, but presumably Purism would like to run the L5 screen with eLCDIF and the external monitor with DCSS in order to do convergence at higher resolutions, since the kernel submission mentions that Purism is still working on eLCDIF.

Purism could switch to the i.MX 8M mini in the future which will be more energy efficient with better performance at 14nm, but its GPU and VPU is too weak to be able to handle convergence as a desktop PC.

8 Likes

I think, we should open a new thread about privacy lovers. Or would it better be a thread about surveillance lovers?

Replace “lovers” with “clerics” and create both proposed threads in round table?

how about software freedom “clerics” ? i think it was RMS that started the church-of-emacs… isn’t that what RYF is ? :wink:

1 Like

@kieran Speculation is endemic to the tech sector at large. How many people out to buy a brand new phone (not a model from last year) do not compare it to the model from some other brand that is soon to be released? It has already been admitted that for this focus on security/privacy (which I don’t take for granted), you’ll be paying somewhere in the neighborhood of $500 to $600 more than you would for a comparable device without that focus. So if you had brought up the GNOME factor exclusively I may have agreed with you, but if it turns out Purism singled out the camera as the secondary selling point(which I highly doubt)…well that would be doubly disappointing for me personally.

Jt0, I’ll readily admit I didn’t phrase that bit correctly, and appreciate your thoroughness, but I did already realize it wasn’t 720p we were talking about. That said, I stand by what I said because if the focus won’t be on performance, and won’t be on social media for most of us (I assume), then what would be the need for such a large screen?

I realize that making this point here is like criticizing an iPhone on the Apple forums, but one of the huge upsides to Linux on desktop is how much faster it makes old and new machines run alike. This, on the other hand, will be a significant downgrade from all of its competitors. It’s not that privacy and security are not worth sacrifice…it’s that the sacrifice being asked here is going back 5 years in time in terms of hardware performance AND paying the 5 to 6 hundred dollar premium.

@amosbatto I would contest that. My thought process was taking into account the market as a whole and what sacrifices would be tolerated to make this a reasonable economic proposition to someone who’s wanting a bit more privacy. That customer would more likely want fewer more refined hardware features as opposed to all of them but old and worse. This would especially be more effective in an era where the market is saturated with ridiculous things like a phone the size of a tablet with a 4k screen and 6 cameras. The customer that knows just how stupid overkill that is would be willing to let go of most of that stuff they’ll never use.

Do I think this is objectively the best and totally worth it option for all of maybe 4 thousand paranoid hacker types across the world who will change out a modem? Yes, absolutely and some other enthusiasts will join in as well. On some level I come pretty close to that description (it’s not an insult), but if we’re talking about the future of privacy here…this product will not expand privacy consciousness or the free philosophy, nor make it any more accessible.

And that’s why I’m making these comments, not to be a contrarian or to show off how disappointed I am, but to encourage Purism to take those considerations more seriously next time if they get the chance. Because otherwise pretty soon they’ll turn all your concerns into marketing points as Apple and Facebook are already in the process of doing, and thus the rest of the world will care less and less if you have the option to be private/free or not.

The question is how is Purism going to convince people to pay a lot for a lower spec phone that is capable of covering their development costs and be the guinea pigs for a new mobile ecosystem (some features won’t work, there will be bugs, and the number of apps will be very limited at first). Purism needs to attract a certain group of people who are willing to pay extra and endure all the tribulations of a new system. The best way to attract that group of hardcore users is to offer them features that they can’t get anywhere else.

At this stage, Purism can’t market to buyers who aren’t willing to pay high prices for lower spec hardware, because its development costs are simply too high. It would be great if the company could market to people who want security a little bit, but aren’t willing to give up much to get it, but Purism doesn’t have millions in venture capital to produce a phone like that and its software isn’t going to be ready for users like that in the next two months.

In the short term, Purism has to focus on the niche users who are passionate about certain features of the phone and are willing to pay a lot for those features. People who care deeply about software freedom, user digital rights, security/privacy or planned obsolescence are willing to pay a lot for the Librem 5 and settle for a phone with a lower resolution screen, less RAM and less Flash memory and weak CPU cores. They will endure all the problems with a new mobile OS and its limited app store, plus recommend it to their friends, post on social media, report bugs, create apps and translate them, install frequent buggy updates, (and waste time arguing about it on the forum :slight_smile: ) to help the mobile Linux ecosystem grow.

In other words, Purism is marketing to the right set of people. After it has recovered its development costs and it has a couple hundred apps in the PureOS store, it can start producing more competitive phones that will appeal to larger and larger sets of mainstream users. Todd Weaver is very clear in his public interviews that he wants to eventually reach the mainstream (so that his daughters don’t have to give up their digital rights to participate in the modern world) and to do that Purism has to make a phone and online services which are convenient and easy for ordinary people to use.

Purism will eventually get to place where it can offer a phone with better specs at a more competitive price. I don’t think that you will be disappointed with version 2 or 3 of the phone (as long as you don’t need powerful CPU cores and a high quality camera that the i.MX 8M can’t provide).

What we are really arguing about is what should be version 1 of the phone that will allow it to be a commercial success to repay its development costs. If we examine what you want in version 1:

  • No camera
  • Higher resolution screen
  • Battery sealed in case
  • Don’t worry about longevity of phone

vs what Purism is offering in version 1:

  • 13 & 8 MP camera
  • 720p screen
  • Replaceable battery
  • Design against planned obsolescence

Most of the early adopters buying the Librem 5 are willing to live with a lower resolution screen in their phone, but they can’t live with a phone without a camera. Trading a higher-resolution screen for no camera won’t reach a wider group of early adopters for the Librem 5.

If you get rid of the replaceable battery and the openable case, you will lose the tinkers and environmentalists who want a phone that is fixable, but you don’t gain that much. You will save a couple dollars on the phone’s bill of materials, but you still need an openable case to be able to change the cellular modem. If you decide to get rid of the M.2 card and solder down the modem, then you either have to use a global modem with proprietary blobs in the kernel (which would cause thousands of people to cancel their orders) or you have to produce 4 different models (PLS8-US, PLS-E, BM818-E1 and BM818-A1) which would increase the manufacturing costs. Since you need to have an openable case to change the cellular modem, you might as well provide a replaceable battery as well.

Getting rid of the goal of fighting planned obsolescence and providing lifetime software support may save Purism some money over the long term, but it also dramatically lowers the long-term value of the phone and people’s willingness to invest in the phone, if they think it will only be supported for 2-3 years like a normal Android phone. Also, providing a lifetime of software support doesn’t cost Purism that much, because NXP will be providing kernel updates for the next 9 years for the SoC, and Purism can rely on the Linux kernel, GTK/GNOME, Qt/KDE/Plasma Mobile, Debian, and most of the other software it uses to keep providing updates. Providing updates will be less work than for an ordinary phone since the Librem 5 doesn’t have binary blobs in U-Boot and the kernel. Purism certainly won’t spend half a million euros upgrading PureOS to Linux 5.X like Fairphone spent on the Android 7 upgrade.

The camera probably costs around $15 - 25, but once you add in $10 - $20 for a higher resolution screen, you aren’t saving much. You are only going to save a couple bucks by gluing down the battery vs making it replaceable, since you need to have an openable case to change the modem. In summary, your design recommendations will lose a lot of passionate Librem 5 buyers and you don’t save very much on the total bill of materials.

4 Likes

I understand that what is being implicitly compared here is … Librem 5 v. what the same money could buy today from a surveillance capitalist. However some of us are also comparing Librem 5 v. our current phone.

In my situation the Librem 5 has more RAM, a bigger screen, more pixels on the screen, more pixels on the rear camera, more pixels on the front camera. While none of those are necessarily unambiguously better aspects, if we are just looking at basic hardware specification numbers (smartphone epeen), those numbers are moving forward.

5 Likes

This is how I’m looking at it also. The specs for the Librem 5 is definitely a step up from my current 2015 phone in every way but the storage. With the SD card, though, that is also taken care of :slight_smile:

There won’t be the massive App Store, but I really don’t view that as a problem. I value freedom above installing a ton of apps on my phone. And as desktop Linux apps are ported to the phone, that gap will start to close.

5 Likes

Same for me. In comparison to my current Galaxy S3 Neo, it’s almost only getting better. Double the RAM, some more pixels. Two (possibly) better cameras.
And back in the day, the S3 was a flagship, originally released with 1GB RAM.

So, I just noticed that my S3 has NFC. Never used it :wink: It also has FM radio, which I used once, I guess.
The only drawback might be the weight of the L5, and maybe the battery standby, but that remains to be seen. L5 has almost double the battery capacity.

3 Likes