3 ms·

I like your list, except for IOT, which is innovation with negative value.

by marssaxman 10y ago

petra 10y ago
Thanks. And sure, consumer IOT is bullshit. But industrial IOT is very valuable.
- marssaxman 10y ago
  I hadn't heard of such a thing as "industrial IOT", but I can see how there might be a difference - more of an evolutionary development out of ordinary embedded control systems, I imagine. Do you have any examples handy? I'm curious what's going on in that field.
  - petra 10y ago
    Industrial(i.e. commercial) IOT examples: - waste disposal: system knows trash levels in each bin, allows optimization of routes, saving 30%. - agriculture: for trees: a bug trap, that measures the bug density in an area - allowing optimization of pesticide. etc for agriculture. - Smart sensors inside pipes, allowing cities to detect leaks in advance. Same for clogged pipes in restaurants. - Putting a lot of sensors in manufacturing processes, collection and analysis to see at what conditions there are problems with the product. - in restaurant chains, putting a lot of sensors, allowing management to keep an eye. also, measuring and tracking everything allows better stock management, critical in high-perishability stuff,
nostrademons 10y ago
Consumer IoT is actually pretty appealing, but it's been incredibly poorly executed so far. I would love to be able to get breakfast started while I'm still in bed, or let guests DJ the party from their cell-phone, or have a robot put away my dishes as soon as the dishwasher is done, or automatically transfer the wash when it finishes. Or imagine a product that, when somebody leaves a package on your doorstep, alerts you via smartwatch, lets you view a security-camera view of the premises around your house, and then gives you the option to open the door, have a robot sweep the package inside, and then shuts & locks the door again. It's just that a.) nobody has produced a product that actually lets me do this (other than the DJ thing, which you can approximate with various streaming services) and b.) I want to make sure that nobody other than me can do these things. Until someone solves the security & usability aspects and the industry actually adopts those solutions, it's just a gimmick.
- marssaxman 10y ago
  I can understand, abstractly, why the scenarios you describe might be appealing, but my actual emotional reaction is horror. Complicated systems fail in complicated ways, and the last thing I want in my home is anything that requires a long, fiddly, debugging process after it abruptly stops working for no apparent reason, as all automated gadgets periodically do. Debugging is most of what I do for a living! I don't want to be doing that kind of work just to keep my home functioning. Nor can I stand the idea of having devices in my home that send telemetry back to their manufacturers, because we've seen time and time again that any such central control can and will be used for harm. The recent case of the garage-door-opener manufacturer who bricked their inexplicably network-connected controller in retaliation for a nasty review is a perfect example. I don't care how convenient the device might be, if it requires me to maintain an ongoing relationship with its manufacturer, I'm not touching it. So - much as it might be nice to wake up, yawn, ask the house for coffee, and know that the pot has started brewing over in the kitchen, I most emphatically do not want to actually live in such an environment. I fear that I may be forced to do so, someday, should product manufacturers decide that IOT-encrusted gadgets are so much more profitable that they are no longer willing to sell the ordinary kind.
  - nostrademons 10y ago
    I wonder if people reacted the same way to the last wave of home appliances - the dishwashers, laundry machines, blenders, food processors, vacuum cleaners, and other machines that must've seemed like horrifying abominations to people used to doing those chores by hand.
    - marssaxman 10y ago
      That's an interesting point. Of course it feels different, and I think I'm being totally reasonable, but that's not saying much as everyone always thinks they are being reasonable! One aspect which feels like a significant difference to me is the centralization of control into cloud services. It's great that my washing machine performs the various steps in its job without human supervision. It would be rather less great if the washing machine needed to phone home and get permission from its manufacturer before it started. An IoT washing machine would have to offer a hell of a lot more convenience before it would make up for that kind of a disadvantage... but IoT devices as they exist today are basically just normal devices with extra remote control bolted on. That doesn't buy much, especially not when the remote control remains under the control of the manufacturer. If there were an internet-connected washing machine where the novel automation remained entirely under the end-user's control, that would be a different story; but as things stand today, if someone is trying to sell me an internet-connected washing machine, it's because they are not content to merely sell me a piece of hardware: they are trying to leverage that sale into some novel ongoing revenue stream. Well, I don't want to sign up for them to have an ongoing revenue stream, and I especially don't want them to retain control over the device they have putatively sold me. So, unless this new automation is, let's say, doubling the convenience the device offers, it's a net loss and I don't want it. That brings out the other difference I see between the IoT and previous generations of household automation projects: we are already well out along the curve of diminishing returns when it comes to household chores. There are already lots of automatic devices I don't bother with because they don't offer enough value to justify the space they take up. Sure, I have a refrigerator, laundry machines, and vacuum cleaner; but I don't have a food processor or an automatic coffee maker (I use a kettle and a french press). I used to have a roomba, but it wasn't worth the bother, so I clean the kitchen floor with an ordinary broom. Nor do I have an electric can opener, as my grandparents did, nor an electric towel warming rack, nor a motorized shaver, nor any of dozens of other "labor saving devices" as they used to be called, because they don't actually save enough labor to make them worth the cost of storing and maintaining them. In the case of a hypothetical IoT laundry machine, for example, there's little that further automation and remote control features could accomplish to save more time, because I still have to load the machine by hand and deal with all the sorting and folding and putting away afterward. Perhaps an IoT washing machine could, let's say, send a ping to my phone when the laundry is done. Well... that optimizes a part of the process that is already cheap, so it isn't worth much. I still have to go downstairs and do all the work after I get the ping, and there's nothing that any amount of Internet connectivity can do to help with that. In order to actually be significantly more helpful than today's machines, a future laundry machine would need to have some kind of robot hands coupled with machine vision, so it could sort and fold. That would be a true innovation, something valuable enough to be worth putting up with the increased fragility and novel failure modes - but there's no reason such a device would need Internet connectivity or a subscription to a cloud service.

I like your list, except for IOT, which is innovation with negative value.

petra 10y ago

Thanks. And sure, consumer IOT is bullshit. But industrial IOT is very valuable.

marssaxman 10y ago
I hadn't heard of such a thing as "industrial IOT", but I can see how there might be a difference - more of an evolutionary development out of ordinary embedded control systems, I imagine. Do you have any examples handy? I'm curious what's going on in that field.
- petra 10y ago
  Industrial(i.e. commercial) IOT examples: - waste disposal: system knows trash levels in each bin, allows optimization of routes, saving 30%. - agriculture: for trees: a bug trap, that measures the bug density in an area - allowing optimization of pesticide. etc for agriculture. - Smart sensors inside pipes, allowing cities to detect leaks in advance. Same for clogged pipes in restaurants. - Putting a lot of sensors in manufacturing processes, collection and analysis to see at what conditions there are problems with the product. - in restaurant chains, putting a lot of sensors, allowing management to keep an eye. also, measuring and tracking everything allows better stock management, critical in high-perishability stuff,

nostrademons 10y ago

Consumer IoT is actually pretty appealing, but it's been incredibly poorly executed so far. I would love to be able to get breakfast started while I'm still in bed, or let guests DJ the party from their cell-phone, or have a robot put away my dishes as soon as the dishwasher is done, or automatically transfer the wash when it finishes. Or imagine a product that, when somebody leaves a package on your doorstep, alerts you via smartwatch, lets you view a security-camera view of the premises around your house, and then gives you the option to open the door, have a robot sweep the package inside, and then shuts & locks the door again. It's just that a.) nobody has produced a product that actually lets me do this (other than the DJ thing, which you can approximate with various streaming services) and b.) I want to make sure that nobody other than me can do these things. Until someone solves the security & usability aspects and the industry actually adopts those solutions, it's just a gimmick.

marssaxman 10y ago
I can understand, abstractly, why the scenarios you describe might be appealing, but my actual emotional reaction is horror. Complicated systems fail in complicated ways, and the last thing I want in my home is anything that requires a long, fiddly, debugging process after it abruptly stops working for no apparent reason, as all automated gadgets periodically do. Debugging is most of what I do for a living! I don't want to be doing that kind of work just to keep my home functioning. Nor can I stand the idea of having devices in my home that send telemetry back to their manufacturers, because we've seen time and time again that any such central control can and will be used for harm. The recent case of the garage-door-opener manufacturer who bricked their inexplicably network-connected controller in retaliation for a nasty review is a perfect example. I don't care how convenient the device might be, if it requires me to maintain an ongoing relationship with its manufacturer, I'm not touching it. So - much as it might be nice to wake up, yawn, ask the house for coffee, and know that the pot has started brewing over in the kitchen, I most emphatically do not want to actually live in such an environment. I fear that I may be forced to do so, someday, should product manufacturers decide that IOT-encrusted gadgets are so much more profitable that they are no longer willing to sell the ordinary kind.
- nostrademons 10y ago
  I wonder if people reacted the same way to the last wave of home appliances - the dishwashers, laundry machines, blenders, food processors, vacuum cleaners, and other machines that must've seemed like horrifying abominations to people used to doing those chores by hand.
  - marssaxman 10y ago
    That's an interesting point. Of course it feels different, and I think I'm being totally reasonable, but that's not saying much as everyone always thinks they are being reasonable! One aspect which feels like a significant difference to me is the centralization of control into cloud services. It's great that my washing machine performs the various steps in its job without human supervision. It would be rather less great if the washing machine needed to phone home and get permission from its manufacturer before it started. An IoT washing machine would have to offer a hell of a lot more convenience before it would make up for that kind of a disadvantage... but IoT devices as they exist today are basically just normal devices with extra remote control bolted on. That doesn't buy much, especially not when the remote control remains under the control of the manufacturer. If there were an internet-connected washing machine where the novel automation remained entirely under the end-user's control, that would be a different story; but as things stand today, if someone is trying to sell me an internet-connected washing machine, it's because they are not content to merely sell me a piece of hardware: they are trying to leverage that sale into some novel ongoing revenue stream. Well, I don't want to sign up for them to have an ongoing revenue stream, and I especially don't want them to retain control over the device they have putatively sold me. So, unless this new automation is, let's say, doubling the convenience the device offers, it's a net loss and I don't want it. That brings out the other difference I see between the IoT and previous generations of household automation projects: we are already well out along the curve of diminishing returns when it comes to household chores. There are already lots of automatic devices I don't bother with because they don't offer enough value to justify the space they take up. Sure, I have a refrigerator, laundry machines, and vacuum cleaner; but I don't have a food processor or an automatic coffee maker (I use a kettle and a french press). I used to have a roomba, but it wasn't worth the bother, so I clean the kitchen floor with an ordinary broom. Nor do I have an electric can opener, as my grandparents did, nor an electric towel warming rack, nor a motorized shaver, nor any of dozens of other "labor saving devices" as they used to be called, because they don't actually save enough labor to make them worth the cost of storing and maintaining them. In the case of a hypothetical IoT laundry machine, for example, there's little that further automation and remote control features could accomplish to save more time, because I still have to load the machine by hand and deal with all the sorting and folding and putting away afterward. Perhaps an IoT washing machine could, let's say, send a ping to my phone when the laundry is done. Well... that optimizes a part of the process that is already cheap, so it isn't worth much. I still have to go downstairs and do all the work after I get the ping, and there's nothing that any amount of Internet connectivity can do to help with that. In order to actually be significantly more helpful than today's machines, a future laundry machine would need to have some kind of robot hands coupled with machine vision, so it could sort and fold. That would be a true innovation, something valuable enough to be worth putting up with the increased fragility and novel failure modes - but there's no reason such a device would need Internet connectivity or a subscription to a cloud service.