3 ms·

I feel like fitness/sleep trackers show up how superficial our consumer electronics market is - products were basically sold based on the concept, hardware capa

I feel like fitness/sleep trackers show up how superficial our consumer electronics market is - products were basically sold based on the concept, hardware capabilities or pretty UIs, and there was an assumption that the 'clever stuff in the background' would just fall into place. But, in fact, the algorithms necessary to track sleep, walking, cycling, running, etc, based only on how one part of your body moves are really not at all trivial, and indeed some of that may be impossible just from such a limited source of information. And if I'm buying a device which claims to be able to do some or all of those things, we should see proper empirical, independent trials in advance that demonstrate that it works.

jzwinck 13y ago
Remember Intel's Itanium? It was sold based on the concept and hardware capabilities, and there was an assumption that a "Sufficiently Smart Compiler" [1] would just fall into place. Knuth mentioned it later as "the Itanium approach that was supposed to be so terrific—until it turned out that the wished-for compilers were basically impossible to write." [2] This is quite similar when you put it the way you did: the hardware exists, it is intentionally stripped down to provide the theoretical minimum required to produce the results people want, and we hope that software people will figure out the rest. [1] http://c2.com/cgi/wiki?SufficientlySmartCompiler http://c2.com/cgi/wiki?SufficientlySmartCompiler [2] http://www.informit.com/articles/article.aspx?p=1193856 http://www.informit.com/articles/article.aspx?p=1193856