Live data from Hacker News

PinePhone Manjaro Community Edition

pine64.org

141–150 of 169 posts

Re: PinePhone Manjaro Community Edition

#141

Earlier quoted context omitted.

> In the past when I see people using Ubuntu, I always wondered why? Debian is great. I particularly respect Debian as an organization. On my home laptop, I've used Debian, Arch, Ubuntu, Fedora, etc etc. But on my work computer, I've always used Ubuntu (or sometimes MacOS), because: - I prefer Ubuntu LTS to Debian's testing/stable approach. - Lots of other devs use Ubuntu, so instructions are often focused on the dis…

> I prefer Ubuntu LTS to Debian's testing/stable approach. May I ask why?

Because I like having a stable system with a known "good for" date, rather than "when it's ready". I particularly like using LTS for systems that I don't want to have to touch very often...just set unattended security updates on and leave it be.

I realize I could do something similar with Debian, but testing isn't generally stable enough for the use cases I have in mind, whereas stable gets stuck with some very old packages. If I need a new package toward the end of the lifespan of a LTS, I can often find PPA. If not, I'll just compile it myself.

It's just my preference. To be honest, probably a big part is that Ubuntu is mostly what I've used the past decade.

Re: PinePhone Manjaro Community Edition

#142
post #97

Earlier quoted context omitted.

Your work is impressive! What's amazing is the mere seconds you see a productive screen in. Android takes full 30 seconds for me.

That requires a customized PID 1 init program and a lightweight GUI (that map app is just ~10MiB of code + some code for postgresql that runs in the background) - most of the delay there is waiting for the postgresql to start up, actually. Arch Linux to some lightweight GUI based on Xorg/i3wm takes about 8s. Similar for my Electron/X11 fullscreen based GUI for the phone (also in the video). Both are much larger. Sadl…

Is this one of the cases where PostgreSQL is overkill and something like SQLite would be a better fit?

Re: PinePhone Manjaro Community Edition

#143
post #90

Anyway, with all those distros piling up for PinePhone, I finally decided to release p-boot with display support and GUI multi-boot menu. https://xnux.eu/p-boot/ I'll make some multi-boot SD card demo image next, perhaps with postmarket, arch linux, mobian and ubuntu touch. :)

A bit late but thanks so much for this work -- I consider myself a late early adopter and your work is going to make it much easier for me to take these phones for a spin.

Re: PinePhone Manjaro Community Edition

#144

Can someone help me understand the relationship between phone hardware and carriers in the US? I recently bought a new, unlocked Samsung phone on Amazon. When I put a Verizon SIM card into it, a Verizon-specific firmware module got added. I don't understand why that would be necessary. Asking because I'm attracted to the idea of a phone I truly control, but not sure what that means for carrier compatibility.

> Asking because I'm attracted to the idea of a phone I truly control, but not sure what that means for carrier compatibility.

I haven't looked in a while but IMO if you want this, the only feasible way is likely to go with carrying an internet hotspot and off-shoring 2/3/4G functionality. Ideally, a phone without the second processor is the only way to actually achieve this -- then you have "only" the chip hardware and wifi chips to deal with/secure.

Re: PinePhone Manjaro Community Edition

#145
post #110

I wonder are there similar devices with 3-4” displays instead of 6”? I don’t even need purposely-built Linux phone, I am OK getting one with Android and installing Linux myself. I want modern hardware (LTE, GLES 3.1, ideally optical camera stabilization), reasonably new Linux kernel, and good software support for basic functions (touch screen, phone, SMS, camera, web browser).

It is effectively impossible to install an upstream kernel on random phone hardware. It is very hard to field a phone that can take a kernel without proprietary blobs tied to a particular ancient kernel. So, there are two live efforts in that direction, PinePhone and Librem 5.

There're devices out there which started as proprietary hardware, then community implemented the support. RTL-SDR was initially a TV tuner, Raspberry Pi SoC was designed for a set top box, people made XBMC (now Kodi) re-purposing game console into a Linux media player. Why that never happens to cell phones, what's so special about them?

Re: PinePhone Manjaro Community Edition

#146
post #73

Earlier quoted context omitted.

I believe postmarketOS team closely collaborates with Alpine. It cannot be said about Manjaro, neither in relation to Arch or Arch ARM.

Manjaro is downstream from Arch.

That's not my point. Being a downstream does not imply collaboration.

Re: PinePhone Manjaro Community Edition

#147
post #110

Earlier quoted context omitted.

It is effectively impossible to install an upstream kernel on random phone hardware. It is very hard to field a phone that can take a kernel without proprietary blobs tied to a particular ancient kernel. So, there are two live efforts in that direction, PinePhone and Librem 5.

There're devices out there which started as proprietary hardware, then community implemented the support. RTL-SDR was initially a TV tuner, Raspberry Pi SoC was designed for a set top box, people made XBMC (now Kodi) re-purposing game console into a Linux media player. Why that never happens to cell phones, what's so special about them?

Cellphones have a lot of peculiar, absolutely undocumented and very buggy peripherals, and tricky system-level power management demands.

Pi's SOC was meant to be designed in to a variety of products, so needed to be documented and understood. Cell phone chips are dumpster fires, as a rule, and are forgotten by their manufacturer within months of first delivery.

Re: PinePhone Manjaro Community Edition

#148
post #136
post #127

Earlier quoted context omitted.

I meant code contributions, but thanks for the links ;)

Ah, you can contribute just by doing something for the project you're interested in. None of the development is centrally organized. There's plenty of stuff to do. I can give you a bunch of hints if you're a kernel dev. :)

I'm not (yet?) a kernel dev; I haven't ever contributed to any OS project, but I'd like to start with the pinephone.

Re: PinePhone Manjaro Community Edition

#149
post #47

Earlier quoted context omitted.

use one of the (unofficial) images containing non-free firmware: https://cdimage.debian.org/images/unofficial/non-free/images...

> unofficial And there lies the problem.

But not any sort of actual one--do you imagine that "unofficial" has a technical meaning of greater significance than that Debian people don't care for proprietary binary blobs?

Re: PinePhone Manjaro Community Edition

#150

Earlier quoted context omitted.

...and it's not even just the controversies. Manjaro feels like they play fast and loose with everything . Arch doesn't include AUR helpers by default specifically because of the potential for user-generated packages to install malware or do something damaging (with the intent being that if you want to use the AUR you really ought to understand the implications). Manjaro not only includes AUR helpers--it seems to act…

Unfortunately, the only other easy distro is Ubuntu, and not having the AUR or the Arch wiki or the quite good Manjaro forums and having to use potentially-dead PPAs for things because Ubuntu packages are out of date is less easy.

The OpenSUSE distros--Leap and Tumbleweed--don't get a lot of attention but they're worthy, relatively easy to use, alternatives to Ubuntu and Arch respectively.

Leap is an LTS distro that is more up to date than Ubuntu LTS and Tumbleweed is a rolling release that has enough QA that it doesn't blow up as frequently as Arch.

Post reply on HN