1. For most people, Arduino would be easier to use (being that many Arduino guys are non-programmers to begin with), with a great ecosystem and a variety of hardware.
2. On the other hand, no one would ever use this hardware for an actual product. Cost is too high, power consumption - I assume - is also up there.
3. As opposed to Arduino, this is not a "hard" real time system. It's severely limits the development of applications featuring, for example, motor control or orientation sensing (gyros/IMUs). And from what I'm seeing, even Arduino didn't make great strides in the commercial market; product development often requires very precise level control over your hardware as well as real development tools (namely a JTAG debugger).
So we're left with an interesting experiment, the longevity of which mainly depends on the community acceptance. It basically targets web developers that want to turn LEDs on and off. This is as far away from the "internet of things" as one can imagine, unless these things are one-off hobby projects.
Yes, embedded development sucks, and product development can be an exercise in futility and despair - but this is not the answer.
I entirely agree. I'd much rather be tied into someone else's ecosystem with an electric imp or Xively, than have to work with hardware designed by people who think the software is the difficult bit.
1. For most people, Arduino would be easier to use (being that many Arduino guys are non-programmers to begin with), with a great ecosystem and a variety of hardware.
2. On the other hand, no one would ever use this hardware for an actual product. Cost is too high, power consumption - I assume - is also up there.
3. As opposed to Arduino, this is not a "hard" real time system. It's severely limits the development of applications featuring, for example, motor control or orientation sensing (gyros/IMUs). And from what I'm seeing, even Arduino didn't make great strides in the commercial market; product development often requires very precise level control over your hardware as well as real development tools (namely a JTAG debugger).
So we're left with an interesting experiment, the longevity of which mainly depends on the community acceptance. It basically targets web developers that want to turn LEDs on and off. This is as far away from the "internet of things" as one can imagine, unless these things are one-off hobby projects.
Yes, embedded development sucks, and product development can be an exercise in futility and despair - but this is not the answer.