6 ms·

The Crazy-Tiny Next Generation of Computers (2015)

27 points by apsec112 6y ago

rkagerer 6y ago
I'd love to buy compute power by the millilitre. 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)?
- ohnoNotAgain321 6y ago
  E-dust see Iain M. Banks collection of Culture novels.
- II2II 6y ago
  > What do you think are today's major roadblocks to dust-scale computing? 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.
segfaultbuserr 6y ago
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? [0] https://news.ycombinator.com/item?id=23186842 https://news.ycombinator.com/item?id=23186842 [1] https://en.wikipedia.org/wiki/Health_and_safety_hazards_of_nanomaterials https://en.wikipedia.org/wiki/Health_and_safety_hazards_of_n...
- anitil 6y ago
  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.
  - segfaultbuserr 6y ago
    A related article on the health effects of these TCE leaks in the Silicon Valley. * In Search of the Cookie Dough Tree https://web.archive.org/web/20130523092702/https://www.aarongreenspan.com/writing/essay.html?id=88 https://web.archive.org/web/20130523092702/https://www.aaron...
supernova87a 6y ago
What do we do when every dust mote is sensing and giving off wifi signals in our houses?

27 points by apsec112 6y ago

I'd love to buy compute power by the millilitre. 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)?

ohnoNotAgain321 6y ago
E-dust see Iain M. Banks collection of Culture novels.
II2II 6y ago
> What do you think are today's major roadblocks to dust-scale computing? 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.

segfaultbuserr 6y ago

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? [0] https://news.ycombinator.com/item?id=23186842 https://news.ycombinator.com/item?id=23186842 [1] https://en.wikipedia.org/wiki/Health_and_safety_hazards_of_nanomaterials https://en.wikipedia.org/wiki/Health_and_safety_hazards_of_n...

anitil 6y ago
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.
- segfaultbuserr 6y ago
  A related article on the health effects of these TCE leaks in the Silicon Valley. * In Search of the Cookie Dough Tree https://web.archive.org/web/20130523092702/https://www.aarongreenspan.com/writing/essay.html?id=88 https://web.archive.org/web/20130523092702/https://www.aaron...

supernova87a 6y ago

What do we do when every dust mote is sensing and giving off wifi signals in our houses?