100 days of postmarketOS [LWN.net]

The postmarketOS distribution looks back at its first 100 days. "One of our previously stated goals is using the mainline Linux kernel on as many mobile devices as possible. This is not as easy as it might sound, since many Linux-based smartphones (Android) require binary drivers which depend on very specific kernel versions. It's a tremendous task to rewrite these drivers to work with the current kernel APIs. Nevertheless, some people have been doing that since long before postmarketOS existed. In the case of the Nokia N900 this has been going on for some number of years and almost all components are now supported in the mainline kernel. This has allowed us to use the mainline kernel as the default kernel for the N900, jumping from Maemo's 2.6.x to mainline 4.12!"

100 days of postmarketOS

Posted Sep 5, 2017 18:10 UTC (Tue) by mb (subscriber, #50428) [Link] (9 responses)

> In the case of the Nokia N900 this has been going on for some number of years and almost all components are now supported in the mainline kernel.

Too bad the device is completely obsolete by now.

100 days of postmarketOS

Posted Sep 5, 2017 22:20 UTC (Tue) by m4rtink (guest, #95458) [Link]

Well, not really - a surprising number of people are still using the N900 as there is still no real alternative in the form of a real Linux smartphone with a hardware keyboard.

Also the original OS (Maemo) and it's infrastructure is still being maintained by the community after taking over from Nokia back in ~2013.

you are probably not the market for this project

Posted Sep 5, 2017 23:09 UTC (Tue) by h2 (guest, #27965) [Link] (6 responses)

I have to commend this project for trying to get people out of the habit of believing that something is obsolete that is still fully serviceable and operational. I guess that was Steve Jobs' real contribution to the world of consumerism, the concept of disposable 500-1000 dollar devices as a normal and worse, acceptable part of existence.

These guys really seem to have the right idea, a solid platform, Debian, 1 kernel, 1 single package per device, the potential to create a much more free mobile computing experience, hats off to them, I wish them luck.

I have some older phones, they work fine, hardware is solid, no issues at all, I can swap in my sim card and they work great, better than my new one actually, but of course, android is obsolete, by design, which means insecure, which means, dangerous.

I liked their blog and wiki information, they seem very clear headed about the reality of their project, particularly in the realm of the non free firmware security issues, but they also seem very realistic in terms of how they are approaching this. If I had any interest in phone hacking I'd jump on this project, sadly I don't, but this really seems like a good way forward, they also covered the problems with other projects that have tried, partially, to deal with some of these issues, and I have to say, it's a good thing most of those projects faded away as far as I can see.

you are probably not the market for this project

Posted Sep 6, 2017 1:05 UTC (Wed) by pabs (subscriber, #43278) [Link]

FYI, postmarketOS has nothing to do with Debian, they are an Alpine Linux derivative.

you are probably not the market for this project

Posted Sep 6, 2017 10:01 UTC (Wed) by pizza (subscriber, #46) [Link] (4 responses)

> I have to commend this project for trying to get people out of the habit of believing that something is obsolete that is still fully serviceable and operational. I guess that was Steve Jobs' real contribution to the world of consumerism, the concept of disposable 500-1000 dollar devices as a normal and worse, acceptable part of existence.

I'm sorry, the N900 is functionally obsolete for its original use cases. It lacks sufficient RAM and CPU resources to run modern applications (notably including a browser capable of handling today's WWW), and its radio/RF support is just barely 3G, which severely limits your connectivity options today.

3G is not obsolete

Posted Sep 6, 2017 11:53 UTC (Wed) by Alterego (guest, #55989) [Link] (3 responses)

For several Billion people, having a robust 3G phone is not obsolete.

3G is not obsolete

Posted Sep 6, 2017 12:20 UTC (Wed) by farnz (subscriber, #17727) [Link] (2 responses)

2G isn't yet obsolete - but for large swathes of the world, 3G is going to become a missed generation. 3G service is considerably more complex and expensive to provide than 2G service, while not offering as compelling a set of benefits as 4G service offers.

India's a good example here - Jio is building out a 4G only network, and there are signs of operators considering 3G switchoff in India in favour of running a mixed 2G/4G network (where 2G provides the dirt cheap handsets with service, and national coverage, while 4G provides a better option in areas where people can afford to pay a little bit more than the bare minimum for service).

3G is not obsolete

Posted Sep 6, 2017 14:01 UTC (Wed) by smckay (guest, #103253) [Link] (1 responses)

AT&T shut down 2G 8 months ago. Considering its age and replacements, I don't think it's a stretch to call GSM obsolete. Of course, in tech "obsolete" is another way to say "cheap, stable, and functional".

3G is not obsolete

Posted Sep 6, 2017 14:51 UTC (Wed) by farnz (subscriber, #17727) [Link]

The US is a special case, for three reasons:

It rolled out GSM as a replacement for D-AMPS, rather than as a network in its own right.
When it was rolling out GSM, it did so in the full knowledge that it was going to want to upgrade GSM cells to UMTS in the foreseeable future, and put in suitable backhaul links.
US consumers have generally been able to afford to upgrade from GSM-only devices to GSM + UMTS dual mode or GSM + UMTS + LTE triple mode devices.

This isn't true in most African countries, or in India - GSM was rolled out as "the" local network, and no forethought was put into building an ATM network suitable for UMTS speeds (as UMTS often didn't exist when these networks were planned). In contrast, LTE just needs an IP network (no circuit switching in LTE), and is thus easier to deploy if you've not built a fast circuit switched network based around ATM technology - Ethernet is cheap to add as an overlay now.

On top of that, Jio's deployment of an LTE-only network in India is putting market pressure on cellphone companies to sell cheap LTE phones - there's no equivalent pressure for UMTS devices, as everywhere with UMTS either has GSM or LTE, so if you're going cheap, you simply remove UMTS capabilities. You can now buy (wholesale, for rebranding) LTE-only devices for comparable prices to GSM-only devices.

Finally, LTE is designed to be much easier for the operator to scale quickly - unlike UMTS and GSM. Thus, if you're deploying today, you can deploy an LTE network that's not that great, knowing that you can quickly scale the bottlenecks up as and when you have customers to pay for it, instead of having to correctly design the network for future demand.

100 days of postmarketOS

Posted Sep 6, 2017 1:08 UTC (Wed) by pabs (subscriber, #43278) [Link]

> Too bad the device is completely obsolete by now.

The name "postmarketOS" indicates that this project is targeted at obsolete devices; ones that you can no longer buy on the market, but maybe still buy second-hand.

100 days of postmarketOS

Posted Sep 7, 2017 18:44 UTC (Thu) by joib (subscriber, #8541) [Link]

Once it matures a little bit, this could be cool in conjunction with something like https://www.usenix.org/conference/hotcloud17/program/pres...

Also, hopefully the "project Treble" stuff Google is launching with Android 8.0 will make it easier for pmOS and other phone OS projects to get up and running on hardware. Though of course it will be a while until android 8 devices will be old enough to be thrown away to secondary users.