Today, I did a presentation about our SmolPhone project trying to design a smartphone-like device with one week or more of battery life. The trick is to use a microcontroller instead of a full processor, a memory LCD display instead of an OLED screen and network connexion providing bandwidths in kilobytes to IoT instead, and so on. Once you settle on such setup, you have some issues to build the hardware but it seems doable (even if I'm personnaly unable to even help someone here and even if we don't have the prototype yet), but the real fun arises on the software side. Linux is simply not an option, and even simple applications must be rearchitectured to fit on such a limited hardware.

In fact, you have to wisely chose the feature that you will port. For example, I guess that video will be near to impossible to implement in our setup, and I think that it is where the project really becomes interesting. This is the occasion to explore the features you really need on a smartphone and the futile ones that we could easily drop. To make the long story short, I think that the feature that we need to remove from our smartphone is the complexity. I'm not speaking of the pseudo-simplicity sold by advertisement where everybody can use the device, but more of a real simplicity that makes it easy to hack the device, so that everybody can add what they need without the rest of us having to pay for the complexity of features we don't want.

It's the first time I tried to explain the motivations behind the SmolPhone in a public presentation, while the previous ones were mostly on the technical project. It's not completely clear for me either. We need more thinking, more prototyping and more work, but at least I think there is something interesting here.