Here, depending on where you place the thing you see a different thing than the sensor does, the optical path is completely uncontrolled, unless they're doing eye tracking and post processing?
This seems somehow worse than a cell phone camera.
As a comparison point, many people shoot street photography "from the hip", aiming the camera at the rough spot they want to get with a forgiving aperture setting (around 8 or more). This help either for speed, or engage with the scene without having to stick a camera on one's face.
I think this camera is basically aimed at the same crowd, those who will want some idea of what the camera is pointed at, but want to stay "in the moment" and no fully focus on the capturing process. This is pretty close to a phone camera perhaps, but I'd assume it's a lot more fun and carefree.
I'd totally love to play with one.
The from-the-hip cameras typically have a waist-level viewfinder, which still is a good approximation of the optical path or you don't and you shoot completely blind. In that case any camera does the trick and so does a phone.
Usually the hip part is so you get a better, more candid, portrait due to being inconspicuous. There's some composition advantage due to the perspective difference too.
> Vibe Over Pixels
This isn't about getting a good shot, it's about the vibes and the fun. The viewfinder is the gimmick that catches the eye but it technically doesn't do anything.
That is the case with twin-lens reflex cameras, for example, which are not commonly used for street photography. What the GP meant was holding an SLR or rangefinder camera at hip level, forgoing the viewfinder completely and shooting blind.
Interesting you say this as I have been on my photography journey these last 2 years and keep hearing the opposite opinion: People "cheating" by using a low aperture so they don't have to think about the entire composition as it gets blurred out.
This kind of advice was given before the advent of good enough autofocus where indeed you've rely on the machine to focus your imagine. But at the same time you delegate the choice of the subject to the machine instead of spontaneously capturing "everything".