What do you think are today's major roadblocks to dust-scale computing? e.g. Power harvesting and storage? Scaling down activators for motes to "get around" and "do things"? Tooling to coordinate thousands of nodes on a common task? Bugs (as in insects that might eat your node)?
Like all new technologies, environmental and health hazards are usually learned in a hard and painful way. For example, I love the history of semiconductor development, but simultaneously, I also recognize that a large number of semiconductor fabs in the 1970s operated by Intel, AMD, Fairchild, HP, National, Commodore, etc, are now superfund sites. At many sites, cleanup works are still ongoing today. And usually it was not a result of malice, but inexperience - many operational and regulatory safeguards were simply not established - in fact, many leaks were only discovered decades later. I summarized some of these incidents in a HN commment at [0] if you are interested.
Following the same line of thought, I wonder how dangerous dust-scale computing is, biologically and environmentally. In general, there are a lot of ongoing research projects on the safety of nanomaterials, and have already showed the potential hazards cannot be ignored [1]. In the particular case, it appears to be just a relatively conventional MEMS chip that we are familiar with, not some radically new and different nanomaterials, so [1] doesn't apply here, but the question of "what will happen if these chips enter the environment (air, soil, water) and biosphere" remains, especially if these are going to be used at an industrial scale, and will be intentionally introduced to the environment for remote sensing. The worse case scenario - as bad as plastic in the 20th century?
That comment of yours is great. It's amazing how much damage can be caused by these leaks. A bit frustrating given that the tank's only job is to not leak.
7 comments
[ 5.1 ms ] story [ 25.7 ms ] threadWhat do you think are today's major roadblocks to dust-scale computing? e.g. Power harvesting and storage? Scaling down activators for motes to "get around" and "do things"? Tooling to coordinate thousands of nodes on a common task? Bugs (as in insects that might eat your node)?
see Iain M. Banks collection of Culture novels.
Feather dusters and vacuum cleaners.
More seriously: this is the type of technology that likely has few applications on it's own, but will probably find its way into something bigger.
Following the same line of thought, I wonder how dangerous dust-scale computing is, biologically and environmentally. In general, there are a lot of ongoing research projects on the safety of nanomaterials, and have already showed the potential hazards cannot be ignored [1]. In the particular case, it appears to be just a relatively conventional MEMS chip that we are familiar with, not some radically new and different nanomaterials, so [1] doesn't apply here, but the question of "what will happen if these chips enter the environment (air, soil, water) and biosphere" remains, especially if these are going to be used at an industrial scale, and will be intentionally introduced to the environment for remote sensing. The worse case scenario - as bad as plastic in the 20th century?
[0] https://news.ycombinator.com/item?id=23186842
[1] https://en.wikipedia.org/wiki/Health_and_safety_hazards_of_n...
* In Search of the Cookie Dough Tree
https://web.archive.org/web/20130523092702/https://www.aaron...