[XorNot]

The problem with this idea is that it assumes we have to make trade offs: case in point, remember pre-smartphone? I had a PDA which didn't have a camera, didn't have wifi - just a hardwired connection.

Now I have sitting next to me a smartphone which has 5 cameras, GPS, wifi, bluetooth, NFC, 128GB of storage and 4G.

Basically on a mass production scale it is always going to be cheaper just to put every feature in one device and build a billion of them.

EDIT: Which is not to say the system doesn't have some possibilities - at the right price point I'd replace every light switch in my house with a plate of this, and have them remotely control relays so I could remap everything. For task-specific physical applications you might want to remap/customize to taste, there's a lot of potential.

[h0l0cube]

> Basically on a mass production scale it is always going to be cheaper just to put every feature in one device and build a billion of them.

When it comes to physical widgets, there's a limit to how many you could practically have on one device. I think the real value here is that it could be open to market to allow a real diversity of physical augments allowing people to improvise devices that are unlikely to be manufactured into a single form factor, but might be perfect for their niche use case.

[snek_case]

Why can't the widgets connect to a Raspberry Pi using USB ports? Using plain old USB ports as a connector is a downgrade in terms of aesthetics, but it's a massive upgrade in terms of reusability and versatility. All of a sudden, you can connect your sensors to laptops, PCs and Raspberry Pis alike. You can already get a webcam on eBay for $6. You could design any sensor and actuator to have a USB interface.

Don't get me wrong, like I said, USB devices would look less cute and tidy than the Pockit, and the Pockit is a great achievement, but USB devices truly are more simple and versatile.

[h0l0cube]

Low friction. Lowering the entry point at which someone can fashion a device to their needs means you have a much bigger market. Approachability is exactly why iPhone succeeded where stylus driven PDAs failed. Something like Pockit makes custom hardware more approachable to a wider audience. Maybe with Pockit, R-Pi might even have its iPhone-like killer app to take it from a niche/techinical crowd into regular consumers hands?

[vidarh]

The aesthetics is an important factor. I'd pay far more for something like Pockit than for a bunch of components hanging off a Pi with no coherent design.

[snek_case]

You can use a 3D printer to print a case that fits all the components and makes it look however you want. You're also not restricted to having your USB devices be right next to the Pi's board. It's just more flexible.

Another thing to consider is that the connection to the blocks on the Pockit might be fragile. If you push that button/joystick too hard or bump something, it could easily disconnect.

[vidarh]

I could. That requires me to have a 3D printer and be willing to invest the time in figuring out how to make a case. I don't, and I'm not.

Of course it's going to be more flexible, just like it's more flexible to design your own device from scratch. But I'd trade that flexibility away for more time to do things I actually want to be doing any time, and I'd be willing to pay quite a bit of money to get that kind of flexibility without having to spend as much time on it.

[snek_case]

I just don't think the tradeoff is worth it. The Pockit will have a relatively small community, and you may be forced to use their software packages to develop for it. It may not be very robust or very versatile. Those are all real downsides, in addition to the higher cost.

[vidarh]

I'd be inclined to be looking for excuses to put these all over the house.... I don't need them, but I want them just from this video.

[RajT88]

What I like about platforms like this is the possibilities they open up for major players to experiment with weird form factors again. Basically the cost of software development, mostly.

I loved some of the insane designs prior to manufacturers deciding that the iPhone form factor was the way forward.

That said, Moto/Google did try something like this once, "Project Ara" which never really went much of anywhere. It led to a few modules being released for the Moto G and that was it. (IIRC)

[RajT88]

Looked it up. It was the Moto Z line.

[scoot]

> have them remotely control relays so I could remap everything

No need for relays – you can replace the bulbs with smart lightbulbs that can be controlled via Zigbee. Ikea smart bulbs are high quality and good value.

[vidarh]

ZigBee is good, but can grt annoying when you have a lot of them, and you have to think about how to make the mesh resilient (my house is a nightmare for signal propagation...). It takes frustratingly long for all of mine to switch off. A Lightwave bridge would be a way to solve that without relays, though. They provide what you need to turn the lights on and off, and you could use Pockit for larger control panels.