Josh On Design

Yesterday I ran across a post on Hacker News about a new phone. From the website it appears to be a brand new OS with features to help you keep focused and clear away the distractions. While the site was pretty, the lack of good UI screenshots was a red flag. What did this OS look like? How can a new OS possibly be built by a small company and be any good?

In the comment threads it turns out this is a stock Android 8 device with a custom launcher and skin, no different than what Samsung or Motorola will sell you. The disappointment in the comments was palpable. Clearly many people want a new take on the phone experience; one that focuses on productivity and efficiency, and doesn't sell your data to the highest bidder.

As many comments pointed out, such an idealized device is impossible to build. Even assuming you could get low quantity hardware at a decent price and quality, you would still need Android because without it you won't have any apps.

However, upon further reflection I think it might be possible to create a small but profitable phone company, if we are willing to think differently.

Think Different

The first problem is apps. Without a popular platform you can't get the apps people need. And your platform won't be popular without apps. This is a difficult bootstrapping problem that kills most platforms. Alternatively, integrating support for existing app platforms into anything new will kill you with engineering support costs. Blackberry tried this. Windows Phone tried this. Even with the resources of Google it took years to support Android apps on Chrome OS, and the results are mixed.

You could try to get people to build new apps for your new platform, but that won't help you with the big guys like Youtube and Twitter who both disallow full 3rd party clients and won't make their own port to your custom platform.

The only viable solution is the web. Make PWAs (progressive web apps) run really well on your platform. So good that using Twitter, Youtube and Facebook through the web feels as nice or better than using native apps.

Of course several companies have tried exactly this. Mozilla's own Firefox OS went down the same route and failed. There were similar efforts inside of Nokia when I was there that also never went anywhere. How can anyone else succeed with an approach that big companies have already tried and failed?

Ongoing Revenue

I think the problem is that we assume a phone is a product sold once, and that the software is merely an expense in building the phone. Thus the only rational strategy is to spend as little on software as possible, which is exactly what has happened to so many Android phone vendors.

No, I think we must accept that software incurs an ongoing cost, even without new features, and must therefore have ongoing revenue. How could this work?

A company could try to build a thriving app store and take a 30% cut of all sales, but we've seen how this not that profitable and leads to anti-user behavior. No, I think the answer is both retrospectively obvious and yet something no one has tried: charge the end user a monthly fee

By charging a monthly fee the OS vendor has consistent revenue to support keeping the phone's software secure and updated. This is the only way we will see OS vendors invest in keeping older phones working rather than encouraging the constant cycle of new hardware.

Yes, I know what you are thinking. You don't want to lease a device, you want to own it. But realistically you are already leasing your phone today. You can't use a single phone for a decade. Phones will never become heirlooms. If you get an iPhone and never buy any 3rd party apps you are still paying your carrier for service, you will still have to buy a new phone in a few years when Apple stops shipping you OS updates, and you still paid a premium to Apple for the hardware. If you buy an Android phone it's the same, just on a shorter timescale.

There is a monthly total cost of ownership (TOC) for a phone, which is really a monthly lease. It just doesn't feel like a lease because we pay for it two or three years at a time.

If we change our viewpoint so that leasing a phone is perfectly reasonable and in fact preferable, then it means the OS vendor is working for the user, not for app devs or data collection agencies or carriers.

A Different Kind of App Store

Now we can move on to the problem of how we get apps. 3rd party developers won't make anything for your platform without money, and without an existing user base there is no money. Let's do something crazy and flip the app store upside down. Instead of the developer selling apps and giving a cut to the store, let's have the store pay the app developers!

If the OS vendor gets monthly revenue, then a percentage can be shared with app developers based on how much value they provide to users. The vendor could also sponsor the developers who maintain open source libraries, like openSSL.

There are still challenges of course. If you aren't using Android, or buying a million chips at a time, the hardware vendors aren't going to give you device drivers for your OS. The only reasonable way to deal with this is to go with an Android stack (kernel, filesystem, device drivers, baseband) and replace the top half with something new. And building this new top stack isn't trivial either.

Building new hardware and software stacks are hard problems, but they are also known solvable technical problems. As long as we have a viable business model to start with I think it is possible for a small company to build a new phone this if they manage expenses wisely. They won't be a huge mega company, but they can be profitable and successful.

Would you pay a monthly fee for quality and privacy?

So my question to you is, would you pay a monthly fee to use a phone that works only in your interest? One where the phone will be supported for five or ten years. One where app developers are paid to make good software instead of trying to trick you with slot machine game mechanics. If a thousand people say yes, then I think it's possible to build a new kind of smartphone.