Good point except the reference was not on movability but the way they are connected. If these devices can connect to WIFI, they may be immobile but yet will have all the typical units of wireless units to be connected.
IoT designs are the sort of embedded systems that would greatly benefit from ESL (most engineers are now familiar with C/C++ while RTL design in Verilog and/or VHDL requires additional training) and hardware/software co-design to optimize performance, power and silicon resources.
Virtually IoT is also a M2M connectivity if you look broadly, but IoT will have a larger reach and distance covered as compared to M2M connectivity. It is also quite sure that these both technological threads will grow a lot. And this will also open many new commercial and technology developments as well.
>> Cars that talk to other cars and sense ways to avoid accidents.
Possible only after all the legal systems have been worked out. The challenge will not be the technology capability. Rather, the legal instruments that will guide the interface of the talking cars and the dumb ones on the road. Imagine in five years someone selling you anti-virus/firewall to protect your fridge and your lighting because someone can hack them and video all you do.
We will need to redesign the ways we build embedded technologies as the era of IoT comes upon us. A new way of thinking will be needed. As we expand this sector and include human systems as extensions of the nodes, we will experience immense supply chain issues. We may even need a new OS that can help unify these. Cloud will benefit exceedingly as it can help to make interoperability easier.