The patent is for capturing the metadata and analyzing it. Guess what the Google van was supposed to do? That's right: capture the metadata, and analyze it. Nobody's disputing that, nobody ever has disputed that.
The accidental part is that it turns out they were capturing more than metadata. The patent doesn't talk about doing that, there's no evidence Google ever intended to do that, and it's difficult to determine what they could possibly gain from it anyway.
So, here, let's improve the headline.
"Google Has Pending Patent For Exactly The Process They Tried To Implement, But Slightly Screwed Up"
The Android widget interface uses a 4×4 grid to arrange widgets and icons. With a larger display, there needs to be better use of the screen real estate. Some of the early tablets we've seen are running a basic stock Android home launcher. This makes space feel wasted by having giant widgets and shortcuts taking up too much room.
No, really? With a larger screen you have more screen real estate? What genius! Google has certainly not thought of that.
I mean, call me nuts, but I suspect this is pretty far down on their list of Critical Things To Change Right Now, but pretty high up on the list of Things That Must Be Fixed Before Release. It's not done yet. They still have time.
And, in at least one case: she's under 18, but she took the video herself and didn't release it until she was over 18. I'd personally consider it hard to claim she was victimizing herself, but it's been successfully prosecuted.
I sort of disagree. Humans are really, really bad at this kind of management, and a smart computer algorithm can often do better. Just look at the people who disable swap space because "it makes the computer slower". You can't trust humans to manage this optimally, and computers can, in theory at least, generate extremely complicated structures and processes (i.e. "if the user runs this program, he's probably about to be reading this data, so let's get this onto the SSD ASAP.")
Dude, it's Google. When I worked there I used hundreds of gigabytes as scratch space and then forgot to delete them. How much bandwidth do you think is really used by an average Wifi link? It's not really all that much - I'd be surprised if they ended up with more than a few gigabytes from driving around every day.
And chances are very good that "looking over the data" was a completely automated process. It's not like they're going to have humans pore over gigs of data.
First, you write some code, and then you test it indoors. Then you debug it until you run out of bugs.
Then you bring it outdoors, with the hardware suite in your passenger seat, and drive it around for a bit. Then you debug that until you run out of bugs.
Then you get someone in a remote location to do the same thing. Then you debug that until you run out of bugs.
Then you hire a few cars to drive around for a day or two. Then you debug that until you run out of bugs.
Then you cover a neighborhood, or a medium-sized city. By this point, it works. Is it missing stuff? Maybe. Is it picking up more than it should be? Maybe. But you think it works, and, honestly, that's all you can ever be sure of.
And then you keep scaling up. You fix bugs as they arise, you carefully test dubious areas, but overall, if it's getting the data you expect, you don't worry about it too much.
I imagine what they did was "record everything to a file", and then, later, "scan that file for beacons". In reality, they should have been doing "record beacons to a file", and then "read that file", but I can certainly see how the mistake was made.
Videos provided by the PR firm, placed on Youtube? Invariably re-encoded into something smaller. Bam! Checksum completely obliterated!
Videos ripped straight off the DVD or Blu-Ray disc, byte for byte, then redistributed? Data not changed! Bam! Checksum . . . completely intact!
So as I understand it, detecting an unauthorized video with MPEG 7 means you have to download it, determine what it's actually a video of if the checksum is utterly missing, and then, even if the checksum isn't missing, determine if it was authorized. This differs from the current approach, where you have to download it, determine what it's actually a video of no matter what, and then, despite the fact that it never had a checksum which would probably be gone now anyway, determine if it was authorized.
Can anyone out there describe a form of copyright infringement that this actually helps detect?
One that isn't invented for the sole purpose of being detected by this technique?
It's behavior like this that has kept me from buying Space Hulk or any Warhammer 40k minis. I haven't even picked up any of the Dawn of War games since I found out what they've been doing.
You hear me, Games Workshop? I am your ideal customer. I have lots of money and I want to give you some of it. But I'm not going to until you stop being a goddamn asshole.
I'm defending "flash programmers" because I, someone who used C++ for many years but has since switched to Lua due to the spectacularly higher ease of writing code, am hit by the same policy changes.
Yes, Flash is what's making the big waves, but this isn't about Flash - this is about Apple's ability to control what tools you use to develop iPhone apps.
After all, gas mileage is the most critical thing when people choose their method of transportation. That's why nobody drives cars, and everyone takes public transportation.
C'mon. If I were able to halve my transport time, I would gladly pay an extra 40% for fuel. Those are the numbers your page shows. It's a no-brainer for most people. It would expand our range and save us time, and it turns out that is what people focus on.
Otherwise, we'd all be using public transportation already.
People don't get "hypnotized into believing that something is burning them" via a bright light. Your claim is that a trained hypnotist can cause it to occur, presumably via the power of suggestion because that's all hypnotism really is. If the teller wasn't telling her "oh, this is burning you" then nothing of that sort was happening.
She's not an epileptic and she didn't have a seizure. If that were the case, he'd be getting sued for that.
There is no scientific evidence to indicate that anything you're claiming might have occured. There aren't even *unreputable* stories hinting at it, to say nothing of reputable ones. You've taken multiple unrelated concepts and jammed them together into a complete hypothetical.
Here, I'll list them.
* A trained hypnotist can cause burns with an *ice cube*. This is true! However, it's believed that this is because "cold" and "hot" are interpreted very similarly by the human body. LED lights are neither cold nor hot, and therefore this confusion cannot occur. * A *trained hypnotist* can cause burns with an ice cube. The teller was not a trained hypnotist and did not suggest, in any way, that she should be burned by anything. * She has claimed to get Tourette's. To the best of my knowledge, Tourette's cannot be gotten via LED lights. * I have no idea why you brought epilepsy up, considering that epilepsy does not result in any of the symptoms described and was not mentioned by anyone involved.
Unless you can show some reasonable method via which the teller caused this maliciously, or should have somehow known it was a problem - a specific problem, note, which has, to the best of my knowledge, never occured once in the history of humanity - then there is no way this is even remotely a valid case, and waving your hands and talking about magic doesn't change any of that.
If I get someone to hypnotize me into believing that people's gazes burn me, that does not mean I get to sue people for looking at me because omg burns. Perhaps she should be suing whoever hypnotized her into believing that a harmless LED light could burn her.
On the other hand, like it or not, it turns out that money is a really amazing motivator for human behavior. Compare Wal-Mart to FEMA. Wal-Mart gets it right. FEMA doesn't. Yes, Wal-Mart is doing it for a profit, but unlike FEMA they're actually succeeding in it.
So, yes, all of your points are completely true, but I also find them rather irrelevant if the end effect is what we're going for.
Getting your botnet removed from the user's computer makes bad financial sense, especially when you can get 95% of the same computational power while being totally invisible. They'd be fools to do otherwise.
Also, if conficker really managed to 'rent out' the computing power, the botnet would likely quickly decrease in size, as more and more people would take their systems to repairs, because they are so slow all of a sudden...)
You'd be surprised how true that isn't. Even Windows is reasonably good at idletime priority processes - unless they intentionally used CPU that the user was trying to use, people would probably never notice.
A bigger problem is that there are very few useful problems that are practical to calculate on a distributed botnet of that fashion. It's been tried and failed before, more than once - most real-world problems involve large databases of confidential data, which is obviously inappropriate for a network like this on multiple levels.
I don't know about you, but I really see no reason I should have paid an extra $100 over the last few years for ports that I will never, ever use. I mean, what about PS2 keyboard and mouse ports? What about so-called "standard" keyboard ports? I bet you could add SCSI to a modern computer pretty cheaply.
I mean, basically, your cutoff point is "well I might use it at some point in the future". Apparently a serial port is useful to everyone (it's not) while an ISA port is useless to everyone (it's not). As mentioned, the vast majority of people will use, *at most*, one PCIe slot for a flashy graphics card, an onboard Ethernet port, onboard audio out, and USB. And most people will honestly just rely on the onboard video.
That's it. That's all you need for a modern computer. Anything more is a waste of money for most users.
(also I have no idea how you can use modern sound cards, they're all absolute junk and I haven't had one that worked properly for the last five years, my "sound card" is - natch - connected over USB:P)
Slashdot Mirror
The patent is for capturing the metadata and analyzing it. Guess what the Google van was supposed to do? That's right: capture the metadata, and analyze it. Nobody's disputing that, nobody ever has disputed that.
The accidental part is that it turns out they were capturing more than metadata. The patent doesn't talk about doing that, there's no evidence Google ever intended to do that, and it's difficult to determine what they could possibly gain from it anyway.
So, here, let's improve the headline.
"Google Has Pending Patent For Exactly The Process They Tried To Implement, But Slightly Screwed Up"
SHOCKING!
No, really? With a larger screen you have more screen real estate? What genius! Google has certainly not thought of that.
I mean, call me nuts, but I suspect this is pretty far down on their list of Critical Things To Change Right Now, but pretty high up on the list of Things That Must Be Fixed Before Release. It's not done yet. They still have time.
And, in at least one case: she's under 18, but she took the video herself and didn't release it until she was over 18. I'd personally consider it hard to claim she was victimizing herself, but it's been successfully prosecuted.
It's called "an external hard drive". Plug it in via USB. One terabyte of space for a hundred bucks (probably less, now.)
I sort of disagree. Humans are really, really bad at this kind of management, and a smart computer algorithm can often do better. Just look at the people who disable swap space because "it makes the computer slower". You can't trust humans to manage this optimally, and computers can, in theory at least, generate extremely complicated structures and processes (i.e. "if the user runs this program, he's probably about to be reading this data, so let's get this onto the SSD ASAP.")
It's time to kick ass and browse Usenet, and we're all out of Usenet!
Dude, it's Google. When I worked there I used hundreds of gigabytes as scratch space and then forgot to delete them. How much bandwidth do you think is really used by an average Wifi link? It's not really all that much - I'd be surprised if they ended up with more than a few gigabytes from driving around every day.
And chances are very good that "looking over the data" was a completely automated process. It's not like they're going to have humans pore over gigs of data.
And 1tb of hard drive space costs only about $100, less than 1% of the total copier's cost.
Yes, if they wanted to spend more to do less, that would certainly be an option!
Here's how this sort of thing works.
First, you write some code, and then you test it indoors. Then you debug it until you run out of bugs.
Then you bring it outdoors, with the hardware suite in your passenger seat, and drive it around for a bit. Then you debug that until you run out of bugs.
Then you get someone in a remote location to do the same thing. Then you debug that until you run out of bugs.
Then you hire a few cars to drive around for a day or two. Then you debug that until you run out of bugs.
Then you cover a neighborhood, or a medium-sized city. By this point, it works. Is it missing stuff? Maybe. Is it picking up more than it should be? Maybe. But you think it works, and, honestly, that's all you can ever be sure of.
And then you keep scaling up. You fix bugs as they arise, you carefully test dubious areas, but overall, if it's getting the data you expect, you don't worry about it too much.
I imagine what they did was "record everything to a file", and then, later, "scan that file for beacons". In reality, they should have been doing "record beacons to a file", and then "read that file", but I can certainly see how the mistake was made.
Only because nobody's bothering to mine nuclear fuel - and nobody's bothering to mine it because the stockpiles are a cheaper source of it.
Is that this changes absolutely nothing whatsoever.
Pirated videos? Invariably re-encoded into something smaller. Bam! Checksum completely obliterated!
Videos provided by the PR firm, placed on Youtube? Invariably re-encoded into something smaller. Bam! Checksum completely obliterated!
Videos ripped straight off the DVD or Blu-Ray disc, byte for byte, then redistributed? Data not changed! Bam! Checksum . . . completely intact!
So as I understand it, detecting an unauthorized video with MPEG 7 means you have to download it, determine what it's actually a video of if the checksum is utterly missing, and then, even if the checksum isn't missing, determine if it was authorized. This differs from the current approach, where you have to download it, determine what it's actually a video of no matter what, and then, despite the fact that it never had a checksum which would probably be gone now anyway, determine if it was authorized.
Can anyone out there describe a form of copyright infringement that this actually helps detect?
One that isn't invented for the sole purpose of being detected by this technique?
It's behavior like this that has kept me from buying Space Hulk or any Warhammer 40k minis. I haven't even picked up any of the Dawn of War games since I found out what they've been doing.
You hear me, Games Workshop? I am your ideal customer. I have lots of money and I want to give you some of it. But I'm not going to until you stop being a goddamn asshole.
I'm defending "flash programmers" because I, someone who used C++ for many years but has since switched to Lua due to the spectacularly higher ease of writing code, am hit by the same policy changes.
Yes, Flash is what's making the big waves, but this isn't about Flash - this is about Apple's ability to control what tools you use to develop iPhone apps.
True, admittedly. There's downsides.
Should've just emailed the developer asking for a paypal address, handed him the appropriate amount of money, then torrented it.
Free love?
Free Tibet?
Free Mitnick?
. . . Free Willy?
After all, gas mileage is the most critical thing when people choose their method of transportation. That's why nobody drives cars, and everyone takes public transportation.
C'mon. If I were able to halve my transport time, I would gladly pay an extra 40% for fuel. Those are the numbers your page shows. It's a no-brainer for most people. It would expand our range and save us time, and it turns out that is what people focus on.
Otherwise, we'd all be using public transportation already.
Woman Claims Nerve Damage Caused Persistent Sexual Arousal Syndrome. Doctors Unsurprised.
Woman Further Claims That Falling On The Floor Can Cause Nerve Damage. Doctors Still Unsurprised.
Woman Then Observes That Balancing On A Small Piece Of Plastic Can Result In Falling On The Floor. Doctors Remain Unsurprised.
Hey, you know what else can "cause" persistent sexual arousal syndrome? Basically anything.
People don't get "hypnotized into believing that something is burning them" via a bright light. Your claim is that a trained hypnotist can cause it to occur, presumably via the power of suggestion because that's all hypnotism really is. If the teller wasn't telling her "oh, this is burning you" then nothing of that sort was happening.
She's not an epileptic and she didn't have a seizure. If that were the case, he'd be getting sued for that.
There is no scientific evidence to indicate that anything you're claiming might have occured. There aren't even *unreputable* stories hinting at it, to say nothing of reputable ones. You've taken multiple unrelated concepts and jammed them together into a complete hypothetical.
Here, I'll list them.
* A trained hypnotist can cause burns with an *ice cube*. This is true! However, it's believed that this is because "cold" and "hot" are interpreted very similarly by the human body. LED lights are neither cold nor hot, and therefore this confusion cannot occur.
* A *trained hypnotist* can cause burns with an ice cube. The teller was not a trained hypnotist and did not suggest, in any way, that she should be burned by anything.
* She has claimed to get Tourette's. To the best of my knowledge, Tourette's cannot be gotten via LED lights.
* I have no idea why you brought epilepsy up, considering that epilepsy does not result in any of the symptoms described and was not mentioned by anyone involved.
Unless you can show some reasonable method via which the teller caused this maliciously, or should have somehow known it was a problem - a specific problem, note, which has, to the best of my knowledge, never occured once in the history of humanity - then there is no way this is even remotely a valid case, and waving your hands and talking about magic doesn't change any of that.
If I get someone to hypnotize me into believing that people's gazes burn me, that does not mean I get to sue people for looking at me because omg burns. Perhaps she should be suing whoever hypnotized her into believing that a harmless LED light could burn her.
On the other hand, like it or not, it turns out that money is a really amazing motivator for human behavior. Compare Wal-Mart to FEMA. Wal-Mart gets it right. FEMA doesn't. Yes, Wal-Mart is doing it for a profit, but unlike FEMA they're actually succeeding in it.
So, yes, all of your points are completely true, but I also find them rather irrelevant if the end effect is what we're going for.
There were playable demos at both GDC and PAX. I presume he was at one of those, possibly both.
Getting your botnet removed from the user's computer makes bad financial sense, especially when you can get 95% of the same computational power while being totally invisible. They'd be fools to do otherwise.
You'd be surprised how true that isn't. Even Windows is reasonably good at idletime priority processes - unless they intentionally used CPU that the user was trying to use, people would probably never notice.
A bigger problem is that there are very few useful problems that are practical to calculate on a distributed botnet of that fashion. It's been tried and failed before, more than once - most real-world problems involve large databases of confidential data, which is obviously inappropriate for a network like this on multiple levels.
I don't know about you, but I really see no reason I should have paid an extra $100 over the last few years for ports that I will never, ever use. I mean, what about PS2 keyboard and mouse ports? What about so-called "standard" keyboard ports? I bet you could add SCSI to a modern computer pretty cheaply.
I mean, basically, your cutoff point is "well I might use it at some point in the future". Apparently a serial port is useful to everyone (it's not) while an ISA port is useless to everyone (it's not). As mentioned, the vast majority of people will use, *at most*, one PCIe slot for a flashy graphics card, an onboard Ethernet port, onboard audio out, and USB. And most people will honestly just rely on the onboard video.
That's it. That's all you need for a modern computer. Anything more is a waste of money for most users.
(also I have no idea how you can use modern sound cards, they're all absolute junk and I haven't had one that worked properly for the last five years, my "sound card" is - natch - connected over USB :P)