4 ms·

Now for a bit of hubris: anyone else reading this feel that building a robot that could do pest/weed control and harvesting is almost a triviality compared to s

Now for a bit of hubris: anyone else reading this feel that building a robot that could do pest/weed control and harvesting is almost a triviality compared to say, fixing enterprise software to maximize customer purchases and ad revenue? What are we all doing? Seriously, why can't each of us take a year of our lives and automate all the most awful jobs in the world? Asking for a friend :-P

wccrawford 8y ago
See the XKCD comic about pictures in parks and kinds of birds.
- zackmorris 8y ago
  Hah thanks for that! https://xkcd.com/1425/ https://xkcd.com/1425/ Still, image recognition is effectively a solved problem at this point for AI. Maybe there are still problems with notation and terminology that prevent widespread adoption?
  - stevenwoo 8y ago
    Is there an update to this story or is it not true or some other possibility? https://www.wired.com/story/when-it-comes-to-gorillas-google-photos-remains-blind/ https://www.wired.com/story/when-it-comes-to-gorillas-google... If you prefer a story not on wired (sometimes it shows a subscription nag for me) - https://www.theverge.com/2018/1/12/16882408/google-racist-gorillas-photo-recognition-algorithm-ai https://www.theverge.com/2018/1/12/16882408/google-racist-go...
    - PhasmaFelis 8y ago
      I don't think that's changed, no. My hypothesis, incidentally, is that it's really hard for an AI to tell the difference between human and ape facial features at all; they're too similar. It seems easy to us because a big chunk of the human brain is devoted to face processing. The AI pretty much just knew "primate face," and knew that pale primate faces were usually human, but with a dark primate face it could only guess. (Which doesn't excuse Google for releasing it in that state, of course.)
  - bluGill 8y ago
    There are many different weeds, and some look a lot like non-weeds. Weeds of the same species often look different in different area. Companies are working on the problem, but you can't train on weeds in one area and go to the next.
Max_aaa 8y ago
As someone who dabbles with gardening and know some farmers. It is not as trivial as you think. There is already a lot of technology and automation on big farms.
gamblor956 8y ago
You have it exactly backgrounds. Building yet another database/analytics platform is trivial compared to automating even the simplest real-world tasks. Because in the real world, you don't have full control of everything, and you can't just hit the reset button or restore from backup.
rripken 8y ago
I do enterprise software, I have a small farm AND I just completed a masters in robotics. Probably wasn't intentional on your part but I feel like you just left a Baby Ruth in my pool. Farming isn't horrible. Programming Enterprise Software isn't that hard. Safely doing robots in the real-world is.
chicob 8y ago
When I read articles such as this one I still see a lot of hype. I'm not a Luddite or anything, bu I think the future of sustainable, productive agriculture lies in agronomic science, not laser-robot overlords or the next OGM fad. If driving a car is less demanding than driving a tractor, let alone a tractor coupled with equipment, since we're still some years away from generalized self-driven cars, what can we say about farming in general? There's a lot of work to be done that would make agriculture sky-rocket in terms of efficiency and environment protection. The last great changes in agriculture have been to lock away the farmer from the equipment, managing software, and crops they use, and to make it easier for large companies to guarantee sales and futures markets to gamble on predicted yields. In the past, the food industry was in between farmers and consumers. Now it seems independent farmers are the middle man corporations want to remove: an annoying part of the chain of production. So this has nothing to with technological advances, but in feeding the hype that everything can be automated with ease. This is a publicity stunt for the zeitgeist of the AI dream. And, by the way, pest and weed control is not a triviality. Not if you want to do it right. Automation is not the mere substitution of humans: it's also the adaptation of the chain of value so that it can be automated. It always comes with a cost that is sometimes underestimated.

wccrawford 8y ago

See the XKCD comic about pictures in parks and kinds of birds.

zackmorris 8y ago
Hah thanks for that! https://xkcd.com/1425/ https://xkcd.com/1425/ Still, image recognition is effectively a solved problem at this point for AI. Maybe there are still problems with notation and terminology that prevent widespread adoption?
- stevenwoo 8y ago
  Is there an update to this story or is it not true or some other possibility? https://www.wired.com/story/when-it-comes-to-gorillas-google-photos-remains-blind/ https://www.wired.com/story/when-it-comes-to-gorillas-google... If you prefer a story not on wired (sometimes it shows a subscription nag for me) - https://www.theverge.com/2018/1/12/16882408/google-racist-gorillas-photo-recognition-algorithm-ai https://www.theverge.com/2018/1/12/16882408/google-racist-go...
  - PhasmaFelis 8y ago
    I don't think that's changed, no. My hypothesis, incidentally, is that it's really hard for an AI to tell the difference between human and ape facial features at all; they're too similar. It seems easy to us because a big chunk of the human brain is devoted to face processing. The AI pretty much just knew "primate face," and knew that pale primate faces were usually human, but with a dark primate face it could only guess. (Which doesn't excuse Google for releasing it in that state, of course.)
- bluGill 8y ago
  There are many different weeds, and some look a lot like non-weeds. Weeds of the same species often look different in different area. Companies are working on the problem, but you can't train on weeds in one area and go to the next.

Max_aaa 8y ago

As someone who dabbles with gardening and know some farmers. It is not as trivial as you think. There is already a lot of technology and automation on big farms.

gamblor956 8y ago

You have it exactly backgrounds. Building yet another database/analytics platform is trivial compared to automating even the simplest real-world tasks. Because in the real world, you don't have full control of everything, and you can't just hit the reset button or restore from backup.

rripken 8y ago

I do enterprise software, I have a small farm AND I just completed a masters in robotics. Probably wasn't intentional on your part but I feel like you just left a Baby Ruth in my pool. Farming isn't horrible. Programming Enterprise Software isn't that hard. Safely doing robots in the real-world is.

chicob 8y ago

When I read articles such as this one I still see a lot of hype. I'm not a Luddite or anything, bu I think the future of sustainable, productive agriculture lies in agronomic science, not laser-robot overlords or the next OGM fad. If driving a car is less demanding than driving a tractor, let alone a tractor coupled with equipment, since we're still some years away from generalized self-driven cars, what can we say about farming in general? There's a lot of work to be done that would make agriculture sky-rocket in terms of efficiency and environment protection. The last great changes in agriculture have been to lock away the farmer from the equipment, managing software, and crops they use, and to make it easier for large companies to guarantee sales and futures markets to gamble on predicted yields. In the past, the food industry was in between farmers and consumers. Now it seems independent farmers are the middle man corporations want to remove: an annoying part of the chain of production. So this has nothing to with technological advances, but in feeding the hype that everything can be automated with ease. This is a publicity stunt for the zeitgeist of the AI dream. And, by the way, pest and weed control is not a triviality. Not if you want to do it right. Automation is not the mere substitution of humans: it's also the adaptation of the chain of value so that it can be automated. It always comes with a cost that is sometimes underestimated.