Hacker News Reader

Comment by georgeoliver | Hacker News Reader

My Pinephone doesn't work for me as a primary phone (yet), but I think it's still a very capable device at home with good integration (via KDE connect).

Without millions of dollars in corporate support, hardware vendors like Pine64 and Linux on mobile developers need user support to make progress. If you want a performant Linux phone, put some skin in the game.

View on HN

> If you want a performant Linux phone, put some skin in the game.

The reality is that this ain't going to happen. There will be never enough volunteer-based work to enable the level of effort that it requires to develop consumer devices like smartphones. Commercial actors have so much money and power, and people have bills to pay. This is something where full open source model just doesn't work.

Why is Linux Desktop viable (I'm certainly very happy with it at least) but phones are too much work?

Because state-of-the-art desktop hardware tends to work with Linux, due to the historical origins of the PC as a relatively non-proprietary technology. The same cannot be said of modern mobile phone hardware.

Non-proprietary only because IBM failed to sink Compaq reverse engineering efforts, and weren't able to drive the PC industry into MCA, PS/2 architecture.

I think the idea is the array of phones is so comparatively vast, and their release schedule is so frequent, that it's just much harder to reach any really reasonable market coverage for things like good drivers.

(I'm not sure I agree, I just think that's the argument)

Developing open source hardware is much more difficult.

You mean it's difficult to open the specs? Open-source community will develop the drivers themselves (it's happening with Pinephone).

Even Linux desktop suffers from a lack of contributors. Even many core-infrastructure projects are shockingly dependent on 1–2 devs who have been unsuccessful in attracting more contributions.

One example of where the manpower just isn’t there for both Linux desktop and the PinePhone is a solid maps app. All solutions are little more than tech demos compared to OSMAnd on Android. Yes, OSMAnd itself has grown through contributions from the community, but it basically soaked up already what little manpower there is. There are other examples where running Android apps on the PinePhone under a compatibility layer is seen as a necessity to get around the PinePhone's lack of manpower.

> Even Linux desktop suffers from a lack of contributors.

After IBM intervention and their contribution to GNOME 40 - I'm not sure about suffering anymore.

So yeah, we need more players on mobile Linux field.

I don’t think Anbox is not so much about man power. It’s for proprietary services or services that don’t have a decent Linux app. I recall the early days of Android very well [0], it had a similar lack of apps (then compared to Symbian and Windows Mobile). I don’t think it’s wise to say “this is not going to happen” one year after the first Community Edition PinePhones were delivered.

For an overview of the current PinePhone app landscape I suggest a look at https://LINMOBapps.frama.io – contributions welcome!

[0] I recently brought back old posts to my blog https://linmob.net that I wrote 1.2 years after the G1’s initial US release.

Purism puts a lot of effort in FLOSS development for Librem 5 (compatible with Pinephone). They are also working on the maps.

Sure, Purism puts in a lot of effort. But after a year of the PinePhone drawing on Purism's effort, the community can plainly see that it a drop in the bucket of what needs to be done.

Everyone I know working on something PinePhone-related is concerned about the small size of the dev community.

Great question. This probably isn't the answer, but it feels to me that I've had the same desktop hardware since the earth cooled, but a new phone comes out every 2 weeks.

It certainly isn't viable to make a living selling software to those customers, hence why most app vendors never care and rather target Android.

It took decades to get a usable stack that I am confident on running on most of today's consumer computers.

And things like Wayland and Mir and some desktops reinventing their own wheels all the time are/were seriously putting that into jeopardy.

it wasn't really viable for most people when the desktop hardware/firmware environment moved fast - for better or worse it's remarkably stagnant now

Disagree completely. webOS and Maemo were pretty great and open source. It took half of a decade for Android and iOS to adopt some of the principle features webOS shipped with. Both had large communities developing, often open source, software for them.

Reply on news.ycombinator.com