Are you sure about that ;)?
2 Likes
I don’t think anyone is sure, as terms “Central Europe” and “Eastern Europe” overlap a lot and change their meaning depending on contexts. So let’s say “former Eastern Bloc” instead 
5 Likes
I could get behind that (iron curtain)!
1 Like
That’s why I bought the L5. But asking free software to stop being free in order to fight non-free software is, as it seems, pointless.
1 Like
Who is asking or saying that? Re-read what I’ve said here.
I might be wrong about Pine64 and their intentions. I think some good points have been made here, but I still think that Purism is doing the real heavy lifting here.
At the same time, I’m sure Purism would acknowledge that they themselves are taking as their starting point a zillion lines of existing open source code. Everyone is just building on top of what other people have already built.
If everyone put temporary licence restrictions on individual pieces of the Linux ecosystem source base, it would be chaotic, a legal minefield, much more difficult to manage, …
Perhaps apt (dpkg) could be extended to have a package-freedom-date attribute - presuming that the unit of granularity for the temporary licence is the package, which may or may not be the case.
I’m not saying that it would be impossible to do this (temporary licence restrictions) but it is not something that I would welcome.
2 Likes
This existing code is most probably under a license as GPL2.0 or equivalent and i.m.h. opinion not permissive to add new source with restrictions. In that case purism could not add to the Linux or Gnome stack. (I’m not a lawyer)
1 Like
This is plain wrong. Selling binaries would never work with the PinePhone community, because they are almost all developers compiling their own stuff and distributing their own compiled versions of Purism’s code.
I actually think that TL Lim, the CEO of PINE64, created the company for what I would call humanitarian reasons, since he wanted to provide the hardware at an economical price for community FOSS projects and hardware tinkerers and mod makers. PINE64 could easily charge 50% more for its hardware than it does and it also gives away some hardware (like the PineBook donations) to charity, and sends some hardware for free (like the Nutcracker Challenge) to volunteers to work on creating FOSS drivers/firmware for it, so it isn’t fair to characterize PINE64 as “just in it for the money.”
PINE64 believes in a community development model for software, and it makes sense from that point of view to promote KDE Plasma Mobile, since it is the best organized community-developed GUI for mobile devices. PINE64 can point to a number of successes under this model, such as:
- the development of the Megapixels camera app by a postmarketOS volunteer for the Pinephone,
- the community work on improving the FOSS driver for the EG25-G cellular modem and for freeing parts of its firmware,
- the development of the Mobian distro which started with the PinePhone,
- UBports added the ability to run on Linux drivers to Ubuntu Touch (instead of only supporting Android drivers though libhybris) to Ubuntu Touch, because of the PinePhone and PineTab,
- The PinePhone caused a number of volunteers from postmarketOS and Mobian to start contributing to the development of Phosh, including Arnaud Ferraris, Clayton Craft, Bart Ribbers and Evangelos Ribeiro Tzaras (whose work led to him being hired by Purism),
- The PinePhone has attracted new volunteers to UBports, postmarketOS, Plasma Mobile and Mobian and given them a focus to target their development. It is important that these projects have hardware like the PinePhone where EVERYTHING can work and there are no mystery chips. The PinePhone gave Plasma Mobile the freedom to abandon Halium+libhybris, which should speed up its development in my opinion.
- The work to get Phosh, Plasma Mobile and Lomiri packaged into the major distros like Debian/Ubuntu/Mint, Fedora, Manjaro, Arch and OpenSUSE.
On order for mobile Linux to succeed it needs both PINE64 and Purism, because PINE64 is attracting a large number of volunteers (developers, bug testers, packagers, documentation, help on forums) to the platform, but Purism is paying for software development of the mobile environment which I believe to have the best chance of reaching the mainstream in the long run.
Both volunteer and paid development is vital for mobile Linux to gain mind-share and attract enough people to the platform to make it a credible alternative to the Android+iOS duopoly. My fear is that the PinePhone Pro will take away the upper-end of the market from Purism, so the company can’t continue paying for software development and there will only be volunteer development in the future.
However, I understand why PINE64 released the PPP. One of the goals of PINE64 is to create common platforms (A64, RK3399 and RK3566) that work on many different form factors (SBC, phone, tablet and notebook), and mod makers and tinkers can make things that work across a whole range of devices that are supported for many years, which is why PINE64 promises to manufacture its devices for 5 years so tinkerers have a stable platform for their DIY projects. Since PINE64 already has SBCs and laptops based on the RK3399, it needed mobile devices as well, so it made sense to create the PPP to create a broader platform.
Another of the goals of the PPP is provide better hardware for the FOSS projects, so it is easier for them to attract volunteer developers and users. The PinePhone was holding back the development of mobile Linux, because the GPU in the A64 is too weak to run a modern interface and browser smoothly, USB 2.0 is too slow of an interface to have good convergence as a desktop PC, the A64’s RAM interface is too slow, and the A64 only supports a maximum of 5MP in the cameras.
In other words, arguments can be made that the PPP will be good for the mobile Linux community in general, but I still worry that new orders for the L5 will dry up and we won’t get paid dev work on mobile Linux. However, Purism is also partly at fault due to its price hikes and its delays in delivering the hardware.
A lot of this debate comes down to what you envision mobile Linux becoming in the future and whether you believe that mobile Linux needs paid developers to succeed or not. I think that the work that Purism has done on phoc+phosh, libhandy/libadwaita, squeekboard, calls and chatty, and the transformation of the GTK/GNOME ecosystem to be adaptive and touch-friendly is vitally important for the future of Linux, but I know that there are others who think that volunteer development is all we need and Plasma Mobile or Lomiri are the right platforms for mobile Linux to succeed.
The copyleft licenses that Purism uses has generated a lot of good will in the community and has practical benefits for Purism, such as:
- Attracting PinePhone users to contribute code, do translations, report bugs, and publicize the software. These users are bug testing the software (and in some cases also fixing bugs), so that the software is better for L5 users.
- Getting its software adopted as official GNOME projects (libhandy, libadwaita, calls, chats and fractal-next), which will make it easier and cheap for Purism to maintain in the future.
- Getting Phosh packaged in many distros (Debian->Mobian->Ubuntu, postmarketOS, Fedora, Arch, Manjaro, NixOS, openSUSE), which is work done by volunteers.
- Getting volunteers to work on making the GNOME software mobile-friendly, by adding libhandy and libadwaita classes.
If Purism decides to no longer use standard copyleft licenses or starts delaying the public release of the code, then Purism will probably lose most of this labor by volunteers. It is important that people who order the L5 can have a positive feeling about their contribution to the development of Phosh, which helps generate new orders. My survey found that roughly half of the people who ordered the L5, also ordered the PinePhone, so it may leave a bad taste in their mouth to have to pay extra for Phosh when they are already helping to pay for its development. If Purism started putting licensing restrictions on Phosh or delayed its release, then many of the Linux distros would probably drop it, so Purism would lose volunteers and mind-share.
13 Likes
I fully agree with @2disbetter what He feeling and think about the bad light of Pine64 to Purism.
It is sad to see people speaking well of Pine64 only because it suits them as users but without caring about the company that created the real/hard concept like Purism does on Hardware and Software.
Releasing the code we develop under Free Software licenses is written into Purism’s social purpose. We’re not going to close our stuff, even temporarily 
19 Likes
I do think it’s also worth noting that (in my opinion) the Librem 5 software works a lot better than mobian on the pinephone in general. Main exception being VoLTE support. I expect that won’t change for some time. Though it’s harder for buyers to see without having out.
I’m hoping that when the PPP stops being developer only that Pine64 adds $100 to the price and donates $50 to Purism and $50 to other open source phone software efforts. In my opinion it would be in their best interest. Yes, it’s a slim hope, but a hope nonetheless.
Regardless, I’m tossing my money at Purism, as they’re doing a lot of good work on the software side; and encourage you all to do the same. If they truely can build a device that’s just as capable as iOS or Android that’s open source an respects user privacy, I’d like to see it.
8 Likes
We all would.
2 Likes
This is not correct. The Panfrost driver already support Vulkan by Panvk.
i.MX 8M only support H.264 encode by Hantro driver but not supporting HEVC enc.
i.MX-8M-plus it supporting hard accel H.265 encode.
2 Likes
The RK3399 Pro supports Vulkan 1.0, but RK3399S used in the PinePhone Pro is a bin-selected RK3399, which does NOT have Vulkan 1.0, according to its datasheet (p. 16).
The i.MX 8M Plus (which will probably be used in the Fir batch) has hardware video encode/decode, but the i.MX 8M Quad used in the Aspen through Evergreen batches does NOT have hardware video encoding. See the VPU section of its datasheet (p. 3)
You are talking about bundling where consumers are locked into using non-standardized components so the company could charge monopolistic rents for the components, but that analogy doesn’t apply in this situation. The Phosh mobile environment and its applications that Purism is creating is not locked to a specific product and any Linux phone user is free to use them. In addition, any company can legally take the schematics that Purism has published under the GPL 3.0 and create a clone of the Librem 5, so they can take advantage of Purism’s kernel work as well. Yes, it will take some work without the original CAD files, but the last commercially sold phone which open sourced its CAD files was the Openmoko Neo Freerunner released in mid-2008, and Purism is promising to open source those file when it recovers it development costs.
The same was said about the Linux kernel. I’m not sure why you think that Purism is the
only company out there who will write FOSS phone software.
Actually you’re quoting 2disbetter, not me, but his point is basically right. After witnessing the failed attempts by Motorola, MonteVista, Sharp, ACCESS, Nokia, FiH/Openmoko, Samsung, Intel, Palm->HP, Mozilla, Canonical and Acadine to develop mobile Linux, it is extremely unlikely that any company other than Purism will be daring or foolhardy enough to invest in developing mobile Linux.
All the other companies selling Linux phones (PINE64, Hallo Welt Systeme, F(x)tec, Jolla-Devices.com and Planet Computers) made the rational business decision to avoid the high costs of software development and let the community do most of the work. The only companies besides Purism that pay for mobile Linux development are Jolla Oy and Blue Systems. The few developers from Blue Systems are not enough to make a noticeable difference to Plasma Mobile’s development. Jolla Oy’s work on Sailfish OS is unlikely to ever go anywhere, since every company that has tried to launch a mobile phone with Sailfish OS (Intex, TRI, Inoi, Blackview, Aquarius, MIG, Qtech and Jala) has given up after 1 or 2 models and the community will also never embrace its proprietary Silica interface. Unaffiliated Jolla-Devices.com licenses Sailfish OS from Jolla Oy to do aftermarket sales of Xperias, but it is otherwise impossible to buy a mobile device with Sailfish OS preinstalled.
Tizen is now only being developed for smart TVs, since Samsung announced in May 2021 that it was switching its watches to Android-derived Wear OS. LG is also only developing WebOS as an OS for smart TVs, and LuneOS’s efforts to continue it as an open source project have died, since it hasn’t had a release since Oct. 2019 and postmarketOS has dropped the LuneOS UI due to lack of maintenance. The only other company which is actually developing Linux for mobile phones is KaiOS, which privatized Firefox OS, but none of KaiOS’s code is available to the public.
For the last 2 decades the GTK/GNOME community has talked about supporting mobile devices, but it never happened because Red Hat, SUSE and Google had no business interest in paying for its development, and Canonical decided to go with Qt for Ubuntu Touch. If Purism hadn’t stepped up to the plate, it probably never would have happened. We can see the same story with ModemManager. Everyone knew that oFono was buggy and poorly maintained, but it was Purism that paid Aleksander Morgado to add GSM supplementary voice services to ModemManager, so it was possible to use ModemManager instead of oFono on Linux phones. PINE64’s recent poll of PinePhone users found that 69.8% who “daily drive” the PinePhone are using an interface now based on ModemManager (Phosh, Plasma Mobile and Sxmo/Swmo) vs 6.7% with oFono (Lomiri) (the rest are “no preference” or “other”).
In theory, any company can pay for the development of mobile Linux, but in the real world, you really only have Purism and Jolla Oy that are making meaningful contributions to mobile Linux. Jolla Oy deserves credit for helping to maintain Halium, libhybris, oFono and Maliit over the years, but with Plasma Mobile abandoning Halium, libhybris and oFono (and Sxmo never using them), I don’t see Jolla Oy really doing much for the mobile Linux ecosystem any more. LuneOS and Ubuntu Touch/Lomiri are still using Halium+libhybris+oFono, but LuneOS is effectively dead and UBports is having a lot of trouble maintaining its huge siloed codebase. Sailfish OS + Ubuntu Touch + LuneOS together represent roughly 10% of PinePhone users according to the PINE64 poll, so I don’t see much future in the code that Jolla helps to maintain.
I care about mobile Linux becoming a viable alternative to the Android/iOS duopoly for normal, non-technical users, and I can’t see that happening any time fast without a company paying for software development, and so far Purism has been one company that has been willing to step up to the plate to make that happen. It is easy to throw spitballs from the sidelines (and I have my own criticisms of Purism if you bother to read this forum), but we have to be conscious of the business realities of paying for software development or we have to accept the slow development with 100% volunteer labor. Plasma Mobile started its development in July 2015, whereas Phosh started in Jan. 2018, yet postmarketOS and Manjaro both selected Phosh over Plasma Mobile as the mobile interface for their PinePhone Community Editions, because they didn’t believe Plasma Mobile was ready, which is a good indication of how fast we will advance if we rely solely on volunteer labor.
7 Likes
I recheck, NXP website show that ixm 8m has h.264 enc, which seems fair to me, but on further investigation it seems that ixm 8m doesn’t have h.264 encoder which seems unfair to me, but still the L5 can use other accelerations to encode video for the camera, which will cost more battery and temperature.
2 Likes
Yes it finished recently like mesa 22 but worked on mesa 21.
1 Like
OK, then I stand corrected. It makes me happy when the FOSS driver from the community is better than the Linux driver with blobs from the manufacturer. The last that I checked, work on OpenGL ES 3.0 and Vulkan in the free Etnaviv driver had been put on hold, so the PinePhone Pro will have an advantage over the L5 in graphics.
The big question is the energy consumption of the PinePhone Pro, but that won’t be an issue for people who buy the external keyboard mod with a 6000 mAh battery.
My hope is that we will finally have a competitive processor for Linux phones with the RK3588, but it is an open question how long it will take to get a FOSS graphics driver for the RK3588 and whether it will be energy efficient enough for a phone. This is an area where the PINE64 business model really helps, because PINE64 is planning to be one of the first companies to produce a dev board with the RK3588 and that will kickstart the community working on its FOSS driver.
1 Like
I think purism needs some services to make profit, like cloud storage, expressvpn, tutanota. these are online services that do not affect the software open source.