Ring alerts already give a screenshot of a person approaching your door. Is it difficult to glance at a picture and recognize a person? I think not.
On the other hand, if Amazon can finagle terms of service to collect all the biometric data, that's gotta be worth real money. At the very least, that's a control in their own delivery infrastructure
This dynamic makes me think it might apply to most large gatherings when there's any controversy involved. If you've ever been involved in a precinct caucus when there's disagreement about something up for debate, things can get pretty toxic as the night goes on. Same for city council meetings. That may be why most legislatures strictly enforce codes of conduct and decorum so you hear deadly enemies refer to "my esteemed colleague" even while implying their opposite number may eat children.
I've listened to a lot of Marvin Gaye's work. I can perform a note perfect cover of one of his better known songs. Given some prompting, I can probably write a song that has a similar feel or that incorporates similar elements.
It's fucked that all of those things are, if not explicitly illegal in all cases, certainly off-limits without licensing and permission. But, what legal basis makes it ok for AI to do the same on demand? If these AI cases come out in favor of the AI companies, does that make it safe for songwriters to have inspiration again? Can Ed Sheeran make another album without fear? Will it take laundering song ideas through ChatGPT (with attached logs) to show there was no direct copying?
I wouldn't necessarily expect favorable treatment from the arbitration company when there's a good chance they'll have to pursue Twitter or Musk personally in court for payment.
The immediate flood of posts on Nextdoor about "was that a gunshot?" in the most innocuous suburban neighborhoods where it was almost certainly a backfire or firework seems like Shotspotter is missing a trick by not targeting HOAs for installations.
On the other hand, police departments might monitor social media and get the same information that Shotspotter provides with the added benefit of having a user to ask for followup.
One might wonder why PDs can't rely on 911 reporting, but the unwillingness of police to respond to unsexy calls that won't result in immediate arrests, or super willingness to show up guns blazing could factor into people being reluctant to call the cops
Considering that the original Google search engine patent expired in 2018, it's reasonable to think that at least some of these new entrants were waiting until after that to make their pitch to get funding and develop the infrastructure to launch. Before that, I seem to remember most engines like duckduckgo used either the Bing or Google databases to generate results.
Clearview seems to keep finding itself stuck in a bit of a forked stick. Countries want it to follow local privacy laws by removing citizens' data - which seems impossibly complex to do with the vaguely undifferentiated mass of scraped data they've collected.
But, their business proposition is that they can successfully identify people from images,with minimal false positives. So, an inability to trawl their dataset and remove entries with a really important identifier seems like a big problem with their product. They get to either comply, or admit they suck.
Since the big thing these days is for universities to establish international cooperatives with other schools or even full blown satellite campuses, geoblocking is going to play merry hell with library IT departments. As bad as it is for collaboration in general, the visiting professor whose VPN resolves to a foreign IP is gonna be pissed when they can't participate in a departmental discussion
One of the biggest reasons why insecure products get bought is because consumers have been assured that things sold commercially are expected to be safe. Speaking of which - why is the DHS and FCC taking the lead when we have a Consumer Product Safety Commission? Do IoT devices have to catch fire or smother children to be eligible for scrutiny?
Execs that advocate for automatic systems that can intelligently and consistently identify infringing content really aren't thinking it through. If things like ContentID aren't adequately doing the job (not that they care about the unacceptably high false positive rate) then extra smarts that add better judgement essentially puts them out of a job.
The ability to identify copied work is pretty much the same as being able to identify original work - the entire point of the music industry isn't it?
It would certainly be up a courts to interpret (potentially badly) but the one ultimately in control of real encryption is the person with the password. Since compelling people to divulge passwords has generally been found to be unconstitutional, I don't know if this section really accomplishes anything other than more theater and potential litigation ammunition
Moosehead Brewery will inevitably challenge the use of the pheromone-laden liquid that hunters sprinkle to attract prey. Moose piss is quite expensive, but as Moosehead will show in filings demonstrating potential confusion in the marketplace, their cheap brew is substantially similar!
It's scary that the justification for escalating to more intrusive methods is 'we didn't find out what we wanted to know'. In a sane world, there would be an element of proportionality that weighed the importance of the expected findings against the invasion of privacy.
I'm not sure how that would work. Maybe a neutral third party would _judge_ what sort of action the situation would_warrant_
If your device pings the access point to see if there's a network, that's enough to establish your location even without establishing a connection. By default most wifi enabled devices like to introduce themselves to every router within shouting distance.
Body cameras are far from being universally adopted and in many cases the cameras are imposed on problem departments as a PR move. Cameras adopted as accountability theater sure can't be expected to halt a worsening departmental culture.
Censorship progression: A user anonymously posts a copied article and down goes the site. A user maliciously posts that article in order to take down the site. User-derived content goes away because of the liability.
Someone claims copyright over part or all of a legitimate article, taking down the site while the claim is adjudicated.
Finally, news and commentary posted without attribution out of fear of reprisals is attacked as copyright infringement and nobody can safely lodge a counterclaim of authorship.
So, the possibility of an insecurity that the government can exploit means that a suspected individual has no defense, since they left themselves open - but an insecure chain of evidence doesn't mean the government's effort isn't suspect since the expectation is that their insecurities wouldn't have been exploited? There's some cognitive dissonance in that.
Sure some crimes are so heinous that they require more aggressive investigation, but aggressive used to mean greater manpower and resources, not breaking the rules.
What's the point?
Ring alerts already give a screenshot of a person approaching your door. Is it difficult to glance at a picture and recognize a person? I think not. On the other hand, if Amazon can finagle terms of service to collect all the biometric data, that's gotta be worth real money. At the very least, that's a control in their own delivery infrastructure
Offline analogies
This dynamic makes me think it might apply to most large gatherings when there's any controversy involved. If you've ever been involved in a precinct caucus when there's disagreement about something up for debate, things can get pretty toxic as the night goes on. Same for city council meetings. That may be why most legislatures strictly enforce codes of conduct and decorum so you hear deadly enemies refer to "my esteemed colleague" even while implying their opposite number may eat children.
Sexual Healing
I've listened to a lot of Marvin Gaye's work. I can perform a note perfect cover of one of his better known songs. Given some prompting, I can probably write a song that has a similar feel or that incorporates similar elements. It's fucked that all of those things are, if not explicitly illegal in all cases, certainly off-limits without licensing and permission. But, what legal basis makes it ok for AI to do the same on demand? If these AI cases come out in favor of the AI companies, does that make it safe for songwriters to have inspiration again? Can Ed Sheeran make another album without fear? Will it take laundering song ideas through ChatGPT (with attached logs) to show there was no direct copying?
When you don't pay your bills?
I wouldn't necessarily expect favorable treatment from the arbitration company when there's a good chance they'll have to pursue Twitter or Musk personally in court for payment.
Missed opportunity
The immediate flood of posts on Nextdoor about "was that a gunshot?" in the most innocuous suburban neighborhoods where it was almost certainly a backfire or firework seems like Shotspotter is missing a trick by not targeting HOAs for installations. On the other hand, police departments might monitor social media and get the same information that Shotspotter provides with the added benefit of having a user to ask for followup. One might wonder why PDs can't rely on 911 reporting, but the unwillingness of police to respond to unsexy calls that won't result in immediate arrests, or super willingness to show up guns blazing could factor into people being reluctant to call the cops
Timing
Considering that the original Google search engine patent expired in 2018, it's reasonable to think that at least some of these new entrants were waiting until after that to make their pitch to get funding and develop the infrastructure to launch. Before that, I seem to remember most engines like duckduckgo used either the Bing or Google databases to generate results.
Impossible business model
Clearview seems to keep finding itself stuck in a bit of a forked stick. Countries want it to follow local privacy laws by removing citizens' data - which seems impossibly complex to do with the vaguely undifferentiated mass of scraped data they've collected.
But, their business proposition is that they can successfully identify people from images,with minimal false positives. So, an inability to trawl their dataset and remove entries with a really important identifier seems like a big problem with their product. They get to either comply, or admit they suck.
"public records"
We can't disclose the documents requested due to an ongoing investigation.
*The investigation is now closed*
All relevant documents have been destroyed as they were no longer needed for an active investigation
Prestigious international campuses
Since the big thing these days is for universities to establish international cooperatives with other schools or even full blown satellite campuses, geoblocking is going to play merry hell with library IT departments. As bad as it is for collaboration in general, the visiting professor whose VPN resolves to a foreign IP is gonna be pissed when they can't participate in a departmental discussion
Memorable quote
I'm surprised you didn't use her memorable quote from an interview on The Daily Show -
"I would say that the FEC and men's nipples are probably comparable. There are things that are done that have some value, just like men's nipples"
Re: What could go wrong? Let me count the ways...
Does the declassification clock start ticking from the record creation date in 2012 or from the entry date presumably sometime this week?
What people are used to
One of the biggest reasons why insecure products get bought is because consumers have been assured that things sold commercially are expected to be safe. Speaking of which - why is the DHS and FCC taking the lead when we have a Consumer Product Safety Commission? Do IoT devices have to catch fire or smother children to be eligible for scrutiny?
Magic technology
Execs that advocate for automatic systems that can intelligently and consistently identify infringing content really aren't thinking it through. If things like ContentID aren't adequately doing the job (not that they care about the unacceptably high false positive rate) then extra smarts that add better judgement essentially puts them out of a job.
The ability to identify copied work is pretty much the same as being able to identify original work - the entire point of the music industry isn't it?
Encryption control
It would certainly be up a courts to interpret (potentially badly) but the one ultimately in control of real encryption is the person with the password. Since compelling people to divulge passwords has generally been found to be unconstitutional, I don't know if this section really accomplishes anything other than more theater and potential litigation ammunition
Big game hunters rejoice
Moosehead Brewery will inevitably challenge the use of the pheromone-laden liquid that hunters sprinkle to attract prey. Moose piss is quite expensive, but as Moosehead will show in filings demonstrating potential confusion in the marketplace, their cheap brew is substantially similar!
It's scary that the justification for escalating to more intrusive methods is 'we didn't find out what we wanted to know'. In a sane world, there would be an element of proportionality that weighed the importance of the expected findings against the invasion of privacy.
I'm not sure how that would work. Maybe a neutral third party would _judge_ what sort of action the situation would_warrant_
Re: Re: Re:
If your device pings the access point to see if there's a network, that's enough to establish your location even without establishing a connection. By default most wifi enabled devices like to introduce themselves to every router within shouting distance.
Ineffective window dressing
Body cameras are far from being universally adopted and in many cases the cameras are imposed on problem departments as a PR move. Cameras adopted as accountability theater sure can't be expected to halt a worsening departmental culture.
Takedown
Censorship progression:
A user anonymously posts a copied article and down goes the site. A user maliciously posts that article in order to take down the site. User-derived content goes away because of the liability.
Someone claims copyright over part or all of a legitimate article, taking down the site while the claim is adjudicated.
Finally, news and commentary posted without attribution out of fear of reprisals is attacked as copyright infringement and nobody can safely lodge a counterclaim of authorship.
Both ways
So, the possibility of an insecurity that the government can exploit means that a suspected individual has no defense, since they left themselves open - but an insecure chain of evidence doesn't mean the government's effort isn't suspect since the expectation is that their insecurities wouldn't have been exploited? There's some cognitive dissonance in that.
Sure some crimes are so heinous that they require more aggressive investigation, but aggressive used to mean greater manpower and resources, not breaking the rules.