Re: Re: Hey judge... Is it a derivative work, though?
Yeah, the judge seems to be saying that if it was a derivative work, Genius would be guilty of copyright violations themselves, because they don't have a license to create derivative works. It's not even that it doesn't matter -- Genius themselves shouldn't want their own argument to be true, or they'd be liable to be sued by every artist or label whose lyrics they watermarked.
But when the judge starts with "Even accepting the argument..." that's a pretty good hint that they don't accept that argument...
Yeah, I wonder why Genius didn't go that route to begin with...they do have a lot of minor labels and independent artists that they seem to work pretty closely with, and several artists who do post their own lyrics directly to Genius and seem have some kind of official partnerships (I'll occasionally see artists posting photos of themselves working at the Genius offices...or at least I did pre-COVID)...I very much doubt that these generic major lyrics licenses cover everything that Genius hosts already. But Google might be smart enough to make sure they only copy lyrics for tracks that they have a license for?
So how exactly does one copyright/license cover potentially dozens of different sets of lyrics? Wouldn't the lyrics license only cover fully accurate transcriptions? Can I just submit the text of a book, say it's the lyrics to some Beyonce track, and it's magically covered? Because apparently Google can just lift anything from Genius and THAT'S magically covered...?
I have seen a few instances where people submitted lyrics to completely different tracks on Genius, in addition to the usual arguments about the correct interpretation of a particular line...I know that wasn't part of this suit exactly but how is that handled? If it's not the correct lyrics, it can't be copyright by the musician/label because it isn't their work, so wouldn't that copyright still be held by the transcriber (who, by the ToS, licenses it to Genius alone)?
Re: 'If people know I'm a liar they might not hire me!'
"You also got to love how his excuses shoot each other in the back, as the dupes he hired to represent him try to both argue that he keeps making 'mistakes' because he lacks experience and that those 'mistakes' should be overlooked because he's just filing so many cases, something which I guess requires no legal expertise nor adds any?"
Settlements don't add court experience, nor do cases that are dropped before going to court when a settlement offer is refused. How many of those cases he files actually go anywhere? That might explain how he can still be lacking experience...
As you correctly point out, the problem here is the GOP. They don't control the whole country though, and in the parts they do control they're literally killing off their strongest block of supporters. The problem will solve itself if it has to...
But blue states are generally doing much better. Here in the northeast (RI in my case) we're seeing lower transmission rates than most of Europe as far as I can tell. Those idiots in Florida and Arizona are getting exactly what they voted for -- rule by the Church of the Almighty Dollar.
Here's another fun fact -- the tests that are giving the most massive amounts of false positives are the same ones they're using at the White House! At least there's a silver lining there... ;)
So glad I was born early enough that I had the right make this decision for myself. I do know some people who are quite cautious about staying off of social media, to the point of wearing masks to parties and events so they don't have to worry about being in the background of some photo. I think that's a bit excessive, but I also think it's pretty obscene to take that right away from someone, particularly while they're still a minor.
And while grandmas displaying pictures is pretty normal, kids not wanting to have their photos displayed is ALSO a completely normal thing. We were lucky that those photos were usually physical, so you could sneak them out of the frame and destroy them. My brother and I used to do that after our mom insisted there was nothing wrong with having nude photos of us hanging in the front hallway. You can't do that if they're posted to Facebook, and the whole damn world can see it. So it's nice to see that at least some kids today still do have options.
Sure you could do geofencing or time boundaries or other methods to cut down that data...but it's not mentioned exactly what that would be and how it would work. You could also just upload the GPS coordinates of everywhere you've been. Those are all different systems than what is described above, which all sacrifice privacy and security for convenience. Can you get away with a bit of that without any real harm? Probably. But that's a different system, and we have no data on how that would be implemented.
And I do need to see each number more than once to make a comparison, unless I'm telling the server exactly which numbers I've seen and when. If I don't store the data, then someone I met today might have been in last week's list, so I need the full two weeks of data every night. If I do store the data, then that's potentially a gig or two per day for two weeks that I've gotta store. Also, two weeks is an average. I've seen some doctors stating that the incubation period can in some cases be as long as a month. So we probably don't want to limit tracing efforts to only two weeks.
Storing hashes might help a little, you can maybe reduce the memory requirements by half...but if you go much further than that I think you're going to start having collisions, so you'll have to start checking in with the server to see if those matches are actually valid, and once again you've started sending a bunch of data back to the server beyond the specifications given above.
So the given plan is potentially infeasible for a lot of users, and there are no plans to address that, so nobody can say what kind of system we might end up with if they have to start hacking in solutions to these issues...but it won't be what they've described so far.
We can't manufacture tests fast enough for that. And the tests that we CAN manufacture are currently showing false negative rates as high as 50%. We do NOT want people going back outside right now just because one test said they were OK. Even if this system was working perfectly right now, if you get an alert, you need to quarantine, even after you test negative.
So...the phone doesn't need to store the data, and it can reduce the volume transmitted by...storing the data? You need either a few gigs transmitted or a few gigs stored, you can do one or the other and not both, but you need at least one. Probably you want to have ways of doing both, because neither option is going to work for everyone.
I don't think it really matters if you're going clubbing every night. You need to transmit the IDs of everyone infected, not the IDs of people they were in contact with. Of course going clubbing every night might increase the transmission rate, but it's not directly increasing the number of IDs to be transmitted. Now, if you manage to isolate yourself pretty well, there's a possibility you can reduce the volume by not transmitting keys that you were using when nobody was around -- this assumes that the key exchange is an exchange rather than a broadcast though, if it's a broadcast you have no way of knowing who received it or when. And broadcast would seem to be more reliable. But if it is an exchange, you can distribute only the keys which were actually exchanged with someone. However...people who live with family or a spouse or roommates, or even some people who live alone in apartments and things, are going to be recording contacts all the time. And it's a random number, so your phone shouldn't have any way of knowing that this is the same contact over and over again. So for some reasonably large percentage of the people, I think you will have to broadcast nearly every single one of those codes, for every day that they might be contagious.
I really need to stop talking to myself and get to sleep...but one more thought... :)
But...the ID number is a 128 bit value that changes every 10 minutes? With 1.5 million active cases right now in the US, that certainly would be a long broadcast...
(1,500,000(624)*128)/8 = 3456000000 bytes per day. Three and a half gigs if I'm understanding this right....
Sure, we aren't gonna get 100% market saturation, but we want as much as possible, right? And this might not be only for the USA? And we can't just broadcast today's numbers...you might have been infected two weeks ago and just now installed the update...so that count is going to be a bit larger than just the current active cases too. And the number of cases is still rising. So what, everyone downloads a couple gigs on their cellphone every night? I feel like that could be a problem for a lot of people...and sure, you can save the list and only download updates, but a couple gigs of storage space could also be a problem for a lot of people.
But I guess it'll be alright...I'm sure if that's a problem then someone can figure out a way to do all of that processing in the cloud instead... :)
Here is one other more practical problem...this is part of the operating system. How many Android manufacturers basically never release updated roms after the first couple months? Best case you get two years, worst case you get nothing. And often they're significantly delayed, although I'd hope that at least the updates that do ship will rush to include this. Still, I wonder how long it's going to take before a significant portion of Android users even have this feature...?
Google has said before that they want to pull more control away from the manufacturers...soon they might be able to claim that doing so is a critical public health issue...
But here's the question -- who has verified that this is exactly how it works in practice? Because the last reports I saw (in Wired) indicated that Google wasn't even willing to state that on the record, let alone any kind of independent verification. Keep in mind that this is the same company that said they weren't and didn't intend to be snooping on a bunch of peoples' wifi...and then three years later we found out that they actually were when then they lost a lawsuit and were ordered to stop...and then six years after that they lost another lawsuit over the same issue and had to be ordered to stop AGAIN. And that was only a couple months ago so who knows if they even bothered to comply this time, since they apparently didn't before. So yeah, I'm not exactly going to take their word for it when they swear that THIS product is different and THIS time they're really truly honestly not spying.
I'll consider believing it when someone like the EFF analyzes some packet captures over at least a couple days...but even that seems pretty difficult to do in a realistic scenario (the average Android user sends so much data to Goog, it'd be a needle in a haystack...)
Not that it matters to me...given that my newest phone is an LG V20 with no play services, and a Librem 5 is on my wishlist, there's a good chance I'll never actually own a device capable of running this stuff... :)
Re: Would the rest of the system get it right like this guy.
That is probably all true, but I think it's more than just IP. It's blame in general. The professional organizations also don't want to get blamed for spreading information that turns out to be false. They won't publish anything until it is sufficiently proven.
The real question, IMO, is do we really want doctors and nurses in times like these to be saying "screw it, we can't waste time, this sounds plausible and I'm willing to take the risk/responsibility myself"? It lets good doctors move quickly to save lives, while bad doctors risk doing more harm due to missing or incorrect information. And there's also the third group that will sit back and wait and follow the "official" procedure even if they know it sucks just so they don't get sued later...
I feel like wearing that is likely to cause my landlord to revoke the license to my apartment... they evict people for negative online reviews or submitting maintenance requests for backed up storm drains or anything less than freaking worshiping them basically. They're just as spineless and thin skinned as the copyright ones.
(Meanwhile, I suspect any copyright maximalists I run into probably wouldn't get it...)
Ironic as it is, I think they may have a point here. Twitter regularly gives Trump a pass when he violates their terms; why shouldn't other governments get the same treatment?
Can you imagine what they'd be saying if Twitter banned Trump, even by accident? Surely they have a way of preventing that, and if they're gonna do that it's only fair they apply it to other government officials as well...
That's not the only issue though. My mother is a nurse, and while everyone is panicking over how few ventilators their hospital has (12), she keeps trying to point out that they have even fewer nurses capable of operating them! They could have five hundred ventilators and it wouldn't make a difference...
I work for one of the largest pharmacy chains in the US. There have been times where we've found major privacy issues -- HIPAA violations of the kind that we have already been fined millions per day for. Generally the response from management is "Skip it for now, the priority is that we meet our artificial, self-imposed deadline and we can worry about privacy later."
It's quite refreshing to see some people actually trying to put privacy first. And frankly, it delayed deliveries by ONE DAY. If you can't make it one extra day without someone bringing you food you've got much bigger problems.
Fuck, even the "removing trashcans" thing sounds like a great idea to me. Dumpsters are already a great source of identity theft materials, you think there should be zero attempt to mitigate those friggin' gold mine dumpsters?
Privacy is worth putting a bit more thought into, Tim.
On the post: Judge Tosses Out Genius' Laughable Lawsuit Against Google Over Licensed Lyric Copying
Re: Re: Hey judge... Is it a derivative work, though?
Yeah, the judge seems to be saying that if it was a derivative work, Genius would be guilty of copyright violations themselves, because they don't have a license to create derivative works. It's not even that it doesn't matter -- Genius themselves shouldn't want their own argument to be true, or they'd be liable to be sued by every artist or label whose lyrics they watermarked.
But when the judge starts with "Even accepting the argument..." that's a pretty good hint that they don't accept that argument...
On the post: Judge Tosses Out Genius' Laughable Lawsuit Against Google Over Licensed Lyric Copying
Re: Re: One copyright, many documents
Yeah, I wonder why Genius didn't go that route to begin with...they do have a lot of minor labels and independent artists that they seem to work pretty closely with, and several artists who do post their own lyrics directly to Genius and seem have some kind of official partnerships (I'll occasionally see artists posting photos of themselves working at the Genius offices...or at least I did pre-COVID)...I very much doubt that these generic major lyrics licenses cover everything that Genius hosts already. But Google might be smart enough to make sure they only copy lyrics for tracks that they have a license for?
On the post: Judge Tosses Out Genius' Laughable Lawsuit Against Google Over Licensed Lyric Copying
One copyright, many documents
So how exactly does one copyright/license cover potentially dozens of different sets of lyrics? Wouldn't the lyrics license only cover fully accurate transcriptions? Can I just submit the text of a book, say it's the lyrics to some Beyonce track, and it's magically covered? Because apparently Google can just lift anything from Genius and THAT'S magically covered...?
I have seen a few instances where people submitted lyrics to completely different tracks on Genius, in addition to the usual arguments about the correct interpretation of a particular line...I know that wasn't part of this suit exactly but how is that handled? If it's not the correct lyrics, it can't be copyright by the musician/label because it isn't their work, so wouldn't that copyright still be held by the transcriber (who, by the ToS, licenses it to Genius alone)?
On the post: Copyright Troll Richard Liebowitz Says It's Really Unfair That He Should Have To Tell Clients And Courts How Frequently He's Been Caught Lying In Court
Re: 'If people know I'm a liar they might not hire me!'
"You also got to love how his excuses shoot each other in the back, as the dupes he hired to represent him try to both argue that he keeps making 'mistakes' because he lacks experience and that those 'mistakes' should be overlooked because he's just filing so many cases, something which I guess requires no legal expertise nor adds any?"
Settlements don't add court experience, nor do cases that are dropped before going to court when a settlement offer is refused. How many of those cases he files actually go anywhere? That might explain how he can still be lacking experience...
On the post: Facing Multiple Lawsuits, ICE Decides Not To Punish Foreign Students For Furthering Their Education During A Pandemic
Re:
As you correctly point out, the problem here is the GOP. They don't control the whole country though, and in the parts they do control they're literally killing off their strongest block of supporters. The problem will solve itself if it has to...
But blue states are generally doing much better. Here in the northeast (RI in my case) we're seeing lower transmission rates than most of Europe as far as I can tell. Those idiots in Florida and Arizona are getting exactly what they voted for -- rule by the Church of the Almighty Dollar.
On the post: Appeals Court Judge: Supreme Court Needs To Unfuck The Public By Rolling Back The Qualified Immunity Doctrine
Re:
Yeah...they routinely ignore the "qualified" part of "qualified immunity"...
On the post: The Case For Contact Tracing Apps Built On Apple And Google's Exposure Notification System
Re: Re: Re: Re: Don't minimize the problems
https://www.msn.com/en-us/health/other/abbott-test-still-misses-many-covid-cases-nyu-study-says/ar-B B142fy9
Here's another fun fact -- the tests that are giving the most massive amounts of false positives are the same ones they're using at the White House! At least there's a silver lining there... ;)
On the post: Court Tells Grandma To Delete Photos Of Grandkids On Facebook For Violating The GDPR
Glad I'm a 90s kid...
So glad I was born early enough that I had the right make this decision for myself. I do know some people who are quite cautious about staying off of social media, to the point of wearing masks to parties and events so they don't have to worry about being in the background of some photo. I think that's a bit excessive, but I also think it's pretty obscene to take that right away from someone, particularly while they're still a minor.
And while grandmas displaying pictures is pretty normal, kids not wanting to have their photos displayed is ALSO a completely normal thing. We were lucky that those photos were usually physical, so you could sneak them out of the frame and destroy them. My brother and I used to do that after our mom insisted there was nothing wrong with having nude photos of us hanging in the front hallway. You can't do that if they're posted to Facebook, and the whole damn world can see it. So it's nice to see that at least some kids today still do have options.
On the post: The Case For Contact Tracing Apps Built On Apple And Google's Exposure Notification System
Re: Re: Re: Re: Re: Re: Verification?
Sure you could do geofencing or time boundaries or other methods to cut down that data...but it's not mentioned exactly what that would be and how it would work. You could also just upload the GPS coordinates of everywhere you've been. Those are all different systems than what is described above, which all sacrifice privacy and security for convenience. Can you get away with a bit of that without any real harm? Probably. But that's a different system, and we have no data on how that would be implemented.
And I do need to see each number more than once to make a comparison, unless I'm telling the server exactly which numbers I've seen and when. If I don't store the data, then someone I met today might have been in last week's list, so I need the full two weeks of data every night. If I do store the data, then that's potentially a gig or two per day for two weeks that I've gotta store. Also, two weeks is an average. I've seen some doctors stating that the incubation period can in some cases be as long as a month. So we probably don't want to limit tracing efforts to only two weeks.
Storing hashes might help a little, you can maybe reduce the memory requirements by half...but if you go much further than that I think you're going to start having collisions, so you'll have to start checking in with the server to see if those matches are actually valid, and once again you've started sending a bunch of data back to the server beyond the specifications given above.
So the given plan is potentially infeasible for a lot of users, and there are no plans to address that, so nobody can say what kind of system we might end up with if they have to start hacking in solutions to these issues...but it won't be what they've described so far.
On the post: The Case For Contact Tracing Apps Built On Apple And Google's Exposure Notification System
Re: Re: Don't minimize the problems
We can't manufacture tests fast enough for that. And the tests that we CAN manufacture are currently showing false negative rates as high as 50%. We do NOT want people going back outside right now just because one test said they were OK. Even if this system was working perfectly right now, if you get an alert, you need to quarantine, even after you test negative.
On the post: The Case For Contact Tracing Apps Built On Apple And Google's Exposure Notification System
Re: Re: Re: Re: Verification?
So...the phone doesn't need to store the data, and it can reduce the volume transmitted by...storing the data? You need either a few gigs transmitted or a few gigs stored, you can do one or the other and not both, but you need at least one. Probably you want to have ways of doing both, because neither option is going to work for everyone.
I don't think it really matters if you're going clubbing every night. You need to transmit the IDs of everyone infected, not the IDs of people they were in contact with. Of course going clubbing every night might increase the transmission rate, but it's not directly increasing the number of IDs to be transmitted. Now, if you manage to isolate yourself pretty well, there's a possibility you can reduce the volume by not transmitting keys that you were using when nobody was around -- this assumes that the key exchange is an exchange rather than a broadcast though, if it's a broadcast you have no way of knowing who received it or when. And broadcast would seem to be more reliable. But if it is an exchange, you can distribute only the keys which were actually exchanged with someone. However...people who live with family or a spouse or roommates, or even some people who live alone in apartments and things, are going to be recording contacts all the time. And it's a random number, so your phone shouldn't have any way of knowing that this is the same contact over and over again. So for some reasonably large percentage of the people, I think you will have to broadcast nearly every single one of those codes, for every day that they might be contagious.
On the post: The Case For Contact Tracing Apps Built On Apple And Google's Exposure Notification System
Re: Re: Verification?
I really need to stop talking to myself and get to sleep...but one more thought... :)
But...the ID number is a 128 bit value that changes every 10 minutes? With 1.5 million active cases right now in the US, that certainly would be a long broadcast...
(1,500,000(624)*128)/8 = 3456000000 bytes per day. Three and a half gigs if I'm understanding this right....
Sure, we aren't gonna get 100% market saturation, but we want as much as possible, right? And this might not be only for the USA? And we can't just broadcast today's numbers...you might have been infected two weeks ago and just now installed the update...so that count is going to be a bit larger than just the current active cases too. And the number of cases is still rising. So what, everyone downloads a couple gigs on their cellphone every night? I feel like that could be a problem for a lot of people...and sure, you can save the list and only download updates, but a couple gigs of storage space could also be a problem for a lot of people.
But I guess it'll be alright...I'm sure if that's a problem then someone can figure out a way to do all of that processing in the cloud instead... :)
On the post: The Case For Contact Tracing Apps Built On Apple And Google's Exposure Notification System
Re: Verification?
Here is one other more practical problem...this is part of the operating system. How many Android manufacturers basically never release updated roms after the first couple months? Best case you get two years, worst case you get nothing. And often they're significantly delayed, although I'd hope that at least the updates that do ship will rush to include this. Still, I wonder how long it's going to take before a significant portion of Android users even have this feature...?
Google has said before that they want to pull more control away from the manufacturers...soon they might be able to claim that doing so is a critical public health issue...
On the post: The Case For Contact Tracing Apps Built On Apple And Google's Exposure Notification System
Verification?
That all sounds...actually pretty decent.
But here's the question -- who has verified that this is exactly how it works in practice? Because the last reports I saw (in Wired) indicated that Google wasn't even willing to state that on the record, let alone any kind of independent verification. Keep in mind that this is the same company that said they weren't and didn't intend to be snooping on a bunch of peoples' wifi...and then three years later we found out that they actually were when then they lost a lawsuit and were ordered to stop...and then six years after that they lost another lawsuit over the same issue and had to be ordered to stop AGAIN. And that was only a couple months ago so who knows if they even bothered to comply this time, since they apparently didn't before. So yeah, I'm not exactly going to take their word for it when they swear that THIS product is different and THIS time they're really truly honestly not spying.
I'll consider believing it when someone like the EFF analyzes some packet captures over at least a couple days...but even that seems pretty difficult to do in a realistic scenario (the average Android user sends so much data to Goog, it'd be a needle in a haystack...)
Not that it matters to me...given that my newest phone is an LG V20 with no play services, and a Librem 5 is on my wishlist, there's a good chance I'll never actually own a device capable of running this stuff... :)
On the post: Emergency Room Doctor: Getting Best COVID-19 Treatment Ideas Via WhatsApp
Re: Would the rest of the system get it right like this guy.
That is probably all true, but I think it's more than just IP. It's blame in general. The professional organizations also don't want to get blamed for spreading information that turns out to be false. They won't publish anything until it is sufficiently proven.
The real question, IMO, is do we really want doctors and nurses in times like these to be saying "screw it, we can't waste time, this sounds plausible and I'm willing to take the risk/responsibility myself"? It lets good doctors move quickly to save lives, while bad doctors risk doing more harm due to missing or incorrect information. And there's also the third group that will sit back and wait and follow the "official" procedure even if they know it sucks just so they don't get sued later...
On the post: OK, Landlord: If Copyright Supporters Are Going To Insist Copyright Is Property, Why Are They So Mad About Being Called Landlords?
Actual landlords will hate it more
I feel like wearing that is likely to cause my landlord to revoke the license to my apartment... they evict people for negative online reviews or submitting maintenance requests for backed up storm drains or anything less than freaking worshiping them basically. They're just as spineless and thin skinned as the copyright ones.
(Meanwhile, I suspect any copyright maximalists I run into probably wouldn't get it...)
On the post: Chinese Embassy Gets Briefly Suspended From Twitter; Insists 'Free Speech Must Be Honored' On Platform Banned Across China
They may have a point...
Ironic as it is, I think they may have a point here. Twitter regularly gives Trump a pass when he violates their terms; why shouldn't other governments get the same treatment?
Can you imagine what they'd be saying if Twitter banned Trump, even by accident? Surely they have a way of preventing that, and if they're gonna do that it's only fair they apply it to other government officials as well...
On the post: Another Federal Court Says Chalking Tires Is A Violation Of The Fourth Amendment
Re: Re: Re: Re:
Chalking the ground or a wall is often ruled as vandalism, why wouldn't the same apply to a car?
On the post: Over-The-Air Updates Could Turn Millions Of Inexpensive Devices Into Much-Needed Ventilators To Treat Seriously-Ill COVID-19 Patients -- If Manufacturer Helps
Re: Re: Re: but what if ResMed is right?
That's not the only issue though. My mother is a nurse, and while everyone is panicking over how few ventilators their hospital has (12), she keeps trying to point out that they have even fewer nurses capable of operating them! They could have five hundred ventilators and it wouldn't make a difference...
On the post: GDPR Concerns Temporarily Result In The Removal Of Trash Cans From Ireland Post Office
Sounds good to me
I work for one of the largest pharmacy chains in the US. There have been times where we've found major privacy issues -- HIPAA violations of the kind that we have already been fined millions per day for. Generally the response from management is "Skip it for now, the priority is that we meet our artificial, self-imposed deadline and we can worry about privacy later."
It's quite refreshing to see some people actually trying to put privacy first. And frankly, it delayed deliveries by ONE DAY. If you can't make it one extra day without someone bringing you food you've got much bigger problems.
Fuck, even the "removing trashcans" thing sounds like a great idea to me. Dumpsters are already a great source of identity theft materials, you think there should be zero attempt to mitigate those friggin' gold mine dumpsters?
Privacy is worth putting a bit more thought into, Tim.
Next >>