Comment by nologic01 | Hacker News Reader

Comment by nologic01 | Hacker News Reader

Ok, naive question but I would dearly love to have an informed answer as another phone of mine is dying, it is still TINA and I can't for the life of me project how this will play out:

The first iphone was released in 2007. That's like 15 years ago. People were probably toying with "smart" phones already few years before that. Its not that the world doesn't know how to manufacture smartphones and/or write software for them.

How come in 2023 there isn't an open source mobile phone that is reliable as a daily driver phone? With a financially healthy entity behind it. NB: It doesn't matter if its android or linux or firefox os or anything else.

Surely there is a market for such a device. 1% of a billion is a big number.

Is it because (pick all that apply or please add the true reason):

* Are manufacturers a cartel and won't allow breaking the duopoly? (Doesn't seem plausible, Samsung toyed with Tizen)

* Are governments happy to strike deals with the duopoly? (Doesn't seem plausible, at least in the EU)

* Is it the telecoms that somehow don't want this to happen? Why would that be?

* Is it the FOSS community that can't get its act together? (Doesn't seem plausible, both F-Droid and e.g., KDE mobile communities seem more than on top of supporting this)

So where is the bottleneck? A free mobile device is maybe the most important piece of technology one could have at this moment. I can imagine various actors that would not want it to happen but at the same time all sorts of other actors that would very much want it to happen.

Not an expert, but have worked on phones before. My guess: much more aspects of mobile phones are patented or secret, parts aren’t available to small purchasers, weak OSS culture around PCB design at the level of complexity a mobile phone achieves, lack of support from carriers, general complexity and variety of disciplines required to successfully construct a phone.

Remember that there’s also not really that many open source computers either. The difference is mostly that computers are designed to be general purpose OS booters, whereas phones are locked down consumption slabs.

> Surely there is a market for such a device. 1% of a billion is a big number.

>* Are manufacturers a cartel and won't allow breaking the duopoly? (Doesn't seem plausible, Samsung toyed with Tizen)

I don't think it's a simple duopoly, but an application of Conway's law across the whole globalized tech industry [1]. Keep in mind that originally Android was the open source option, and it's been gradually become less and less amenable to that.

In my mind, I think of the "progress" of the consumer electronics industry (especially in the last 20 years) as undergoing constant and aggressive "distillation" process, converting people away from the FOSS side of the spectrum which offers freedom and control towards the locked down centralization side which offers dependence, convenience, and safety. The 1% market is basically the left over Azeotrope that can't quite be gotten rid of.

The experiments to harness that 1% market are the exception that proves the rule. Locked down centralized solutions are optimal for the tech market as it exists today. Samsung thought they could squeeze some money out of the 1% by being the "master" of Tizen, but realized that the game theory works out such that they would be better off as a servant of Android.

On top of all that, I think as the smartphone capture gets deeper and deeper, being in 1% is not even necessarily a 1-to-1 for FOSS values anymore. I used to fantasize about having a cool handheld Linux computer I could control freely. Now I'm more obsessed than ever about avoiding globalized tech, but I just don't care about smartphones anymore. I carry around an iPhone as the mark of the beast, but at least (I tell myself) I don't let it into my soul.

[1] https://en.wikipedia.org/wiki/Conway%27s_law

Because manufacturing mobile phones is hard and expensive. Companies have large teams of hardware and software people will incentivized to develop phones and lots of them fail.

The best we’ve got so far are exactly custom android builds like grapheneOS or lineage that replace software on phones other companies build. Apparently google doesn’t like that, though, so they’re going to continue to make AOSP worse and worse to protect their spyware.

The best case currently probably varies between something like grapheneOS on a pixel phone or lineageOS which is less privacy focused but has a much wider range of supported devices.

> Because manufacturing mobile phones is hard and expensive

I appreciate that, but in 2021 there were like 1.8 billion phones manufactured and sold.

Are you saying that none of those existing manufacturers can afford or is interested to offer a FOSS mobile?

What advantage does offering a FOSS based phone offer them?

A small but loyal customer base to start with, and a platform that can develop in directions that a duopoly would never have an interest in.

Purism is targeting exactly that, and they had to increase the price of their phone from $600 to $1300 due to the problems I mentioned in another comment.

Check out the Volla phone. Then look at the price and imagine a retail scenario where a person is going to purchase a "no name" FOSS phone or a Samsung. The scale of the marketing is unfathomable.

I agree with you. But you are assuming everyone in the world thinks logically, let alone understands why FOSS is a better option.

> But you are assuming everyone in the world thinks logically, let alone understands why FOSS is a better option.

I know and I have accepted this argument for a long time. But as things get ever more commoditised and the unease with the existing arrangements keeps monotonically growing the silence is kinda deafening.

If we could somehow get a basic white-label device going the magic of FOSS would work its miracle eventually. Think of modern linux desktops and servers. They are not for everyone but they are indispensable in their niche.

Thanks for the Volla tip. Will check it out.

1% of a billion in RETAIL sales is a big number.

Manufacturers do not care about orders less than 10,000 units. They make their money off of volume.

The only government that seems to care about these topics is China, although the rest of the world seems to be catching up on a strong desire to tie the individual to a device.

Telecoms require $500,000 per device for certification through GSMA.

FOSS can only do so much with closed modems. Porting to new devices comes with a heap of issues that require time, which means scaling to market is not viable without $$$.

Here is why Librem 5 phone was very hard and expensive to build: https://puri.sm/posts/breaking-ground/

> This development comes at a high price. We have a team of about 15 developers full time working on this for almost two years.

This is not peanuts but it also doesn't strike me as a serious bottleneck. It is a rounding error in the budget of any serious entity that might want to get involved in this.

Because just like with the iPhone mini, a small but vocal minority is not worth serving if development costs are too high.

Most people don't care about hacking their phone. They just want one that works.

Building a high quality device is a very different beast from the "small" software-projects or gadgets we get from the FOSS-sphere. And even in software there are not that many big projects who carry their whole own ecosphere. There were many projects building FOSS-Smartphpones over the years, most of them failed because of lack of ability I would say.

At the end of the day, while there are many who want FOSS-products, not many are willing to cope with bad experience for they beloved tools. So the actual market is very small if you can't deliver the daily drivers.

Reply on news.ycombinator.com