I don't know if is this an ideal way to judge the authenticity of Thai food, but I think it's refreshing as an applications of artificial senses beyond images and sound.
I've been playing with the idea of doing some machine learning on artificial smell or taste. Does anyone here have experiences with such things? In particular I'd love to learn something about what the options are for the sensors and peculiarities of the sampling process compared to light/sound.
You get what you optimize for. I can imagine food being served, if there's enough incentive to make this machine happy, that conducts electricity in all the right ways, but doesn't actually taste any better.
>The machine evaluates food by measuring its conductivity at different voltages.
They should really have kept the functioning principle behind more smoke & mirrors, because this suddenly makes the whole thing seem very childish and ridiculous. Conductivity may slightly cross over with the tongue's means of perception, but the main determinant of a food item’s flavor is the sense of smell, which is incredibly complex. We don’t even know how it works, but it seems to perform an in-depth analysis of every volatile molecule’s shape and features.
Thai food in NJ/NY is in a sad state. The best Thai food in NYC doesn't hold a candle to an old lady making some gai gap tiam on the street of Chiang Mai.
Whatever the other merits of San Francisco, it does have the best Asian food in the continental U.S. (Hawaii has better).
Nowhere else (and I've looked), comes close.
"Better" here means: the largest number of good options, the highest mean among the better-than-average, and (somewhat aggregate of the previous) the most likely that any not-obviously-bad choice will be good.
11 comments
[ 3.2 ms ] story [ 31.8 ms ] threadI've been playing with the idea of doing some machine learning on artificial smell or taste. Does anyone here have experiences with such things? In particular I'd love to learn something about what the options are for the sensors and peculiarities of the sampling process compared to light/sound.
They should really have kept the functioning principle behind more smoke & mirrors, because this suddenly makes the whole thing seem very childish and ridiculous. Conductivity may slightly cross over with the tongue's means of perception, but the main determinant of a food item’s flavor is the sense of smell, which is incredibly complex. We don’t even know how it works, but it seems to perform an in-depth analysis of every volatile molecule’s shape and features.
Nowhere else (and I've looked), comes close.
"Better" here means: the largest number of good options, the highest mean among the better-than-average, and (somewhat aggregate of the previous) the most likely that any not-obviously-bad choice will be good.
The Japanese food in San Francisco proper is resoundingly mediocre - I can't think of one good ramen place.
If you want good ramen, go to San Mateo or San Jose. Same thing with sushi.