5 ms·
Google seems to be all over the map on this one. I would have thought that minimalism would be a good concept when building to reduce consumption; but rather, G
by cryptoz 8y ago
Google seems to be all over the map on this one. I would have thought that minimalism would be a good concept when building to reduce consumption; but rather, Google is going all-out with an enormously large and complex proposal it seems. And it is full of contradictions. It's also tone-deaf and reeks of a 'we know better than you' attitude, which is something Google needs to not fan the flames of.
> The area will have plenty of sensors collecting data - from traffic, noise and air quality - and monitoring the performance of the electric grid and waste collection.
...
> Sidewalk Labs told the BBC that the sensors will not be used to monitor and collect information on citizens, rather it will be used to allow governments to be flexible about how neighbourhoods are used.
Those two sentences are not compatible. The sensors will be used to observe - surveil - citizens in cars, citizens talking and at the same time, won't be used to monitor or collect information about citizens? How?
Does Google have any plans to describe how they will eliminate human-spoken conversations from their noise detection? Such work can be done. Will they do it? That seems like Problem #1 to address: Google claims they will shove microphones along the streets everywhere but "promise us they won't listen to you"........okay? I don't buy it for a second. Talk to us about it, at least, and don't just say "we promise not to monitor human speech" and leave it at that. This is a hugely important conversation and Google is ignoring it. You can't just plaster public space with corporate surveillance tools and then walk around whistling and ignoring the huge social movement against your never-ending privacy invasion.
> What is clear is that green will be top of the agenda
None of the proposal sounds green to me. The amount of effort that will go into building it seems likely to offset any gains it would save over time through pollution reduction, etc.
The entire thing sounds like it would need new factories, new dedicated infrastructure, and none of it is remotely green-sounding. Where will all of this stuff come from? Is it truly better and less pollutant than an old-world, more simple solution might be?
I'm hopeful for Sidewalk Labs. But it seems like the Reddit Redesign to me: making every single obvious mistake that you could make, while the outside world yells at you "please reddit don't be digg 2.0! please Google don't make a massive dystopian total-surveillance society! please!)
- studio_monolith 8y agoYou honestly can't comprehend how to collect data without collecting individual citizen information?
- ggggtez 8y agoDo you propose a city should not measure traffic or pollution? There is a difference between gathering aggregate data about a neighborhood, and individual surveillance. Though I understand many people with engineering minds don't like to deal with nuance. I think you are reading more into it than you should.
- cryptoz 8y ago> Do you propose a city should not measure traffic or pollution? No, I certainly propose a city should measure it. I propose it be done with taxpayer dollars, open source software, and in an otherwise fully open and democratic way by the city. I do not think corporations should build themselves into cities with a specific plan for surveillance in mind. > There is a difference between gathering aggregate data about a neighborhood, and individual surveillance. Not much of one, depending on the neighborhood and how the data is collected. We're all told that the employees at Google, Facebook, etc have very limited access to customer data, but that oversight is also very unknown and opaque to the outside world, and not trusted. I'm certainly not ready to trust Google with an idea of "no individual surveillance" in their city, when obviously as soon as I go in their with my Android phone they know who I am anyway. > I think you are reading more into it than you should. No way. Google is showing their ambitions at running human society in a very closed, scary, surveillance-enabled way. If anything I'm reading too little into their plans.
- jwalton 8y ago> No, I certainly propose a city should measure it. I propose it be done with taxpayer dollars, open source software, and in an otherwise fully open and democratic way by the city. This, I agree with. Google is not in this for the "public good". I know people who are ranked very highly in the Google maps "guide" program. I wonder what kind of witchcraft Google uses to get people to spend hundreds of hours donating information about their cities in order to improve a proprietary map program for free. -_-
- jwalton 8y ago> Does Google have any plans to describe how they will eliminate human-spoken conversations from their noise detection? I am very much not a fan of this project, but I will point out that the problem you're describing has an easy solution; a microphone can record conversations, but the output of a microphone can be fed to an algorithm that never keeps that recording and only records the noise level in dB. If you've ever used a toilet that flushes itself automatically when you leave, you've been in a bathroom with a tiny camera watching you. No one ever worries about the privacy implications of this, because that tiny camera does nothing with this information except decide if it's light or dark immediately in front of it. The image data isn't stored, it's just thrown away. Is Google going to record thousands of hours of sidewalk microphone recordings and store them somewhere? Seems far more likely they're just going to stream a "Currently it's 60dB at this intersection"; meets their requirements with minimal bandwidth and infrastructure.
- morogop 8y agoThe sensor used to flush a toilet does not capture detailed photos of me and promise to throw it away. It captures essentially no detail in the first place.
- cryptoz 8y agoThere's a big difference in the attitude of the society that builds these things vs. not, which is what I'm trying to describe. I'm aware that Google can create technical solutions that help protect privacy, that's not the point. That's what everyone says about OK Google and Alexa, but it is clearly not really true. These devices malfunction, frequently, with scary results. I'm not ready to trust Google, or anyone, especially when they are so vague on all the topics, with such an incredible opportunity at city-owned-level surveillance, regardless of what technical safeguards they plan for it. > Is Google going to record thousands of hours of sidewalk microphone recordings and store them somewhere? Seems far more likely they're just going to stream a "Currently it's 60dB at this intersection"; meets their requirements with minimal bandwidth and infrastructure. It "seems" to you that Google will actively choose to not record data that would be considered invasive. That's not good enough, though. First, it doesn't seem that way to me at all, I completely disagree with how much data I would expect Google to collect in this situation, and I completely disagree that a dB reading at a lat,lng point will be sufficient for Google's algorithms to do their required work. Again, the main point I'm trying to bring is not the technical issues, but the societal ones. It is too much surveillance issued directly to a company with no regard for discussing the issues openly. It's just not right.
- jimmaswell 8y agoTraffic cameras are often not high-res enough to even read a license plate. Weight sensors can also be used. I don't see the big deal about microphones used for decibel measuremrnts potentially being used to keep audio. Companies record video and audio outside their premises all the time.
- cryptoz 8y ago> Companies record video and audio outside their premises all the time. It's this kind of tone-deafness I was trying to describe in how Google talks about their plans. Yes, I'm aware that companies do that kind of surveillance on their own property. But what Google is trying to here is test out how to build entire cities. Surely you see the important difference and user expectation when it comes to a very clearly designed and secured corporate building or campus vs. an entire section of a residential and commercial city.