Or Congress. This agency started its life with a leader that used blackmail as a standard law enforcement technique. The FBI should have been dismantled from the ground up when Hoover died as his ghost still haunts the agency in all it's actions.
John Kennedy threatened to do that with the CIA, and you see where that got him...
All though I would also argue that the government should be liable if the signing keys are leaked
Liable for what, exactly?
Apparently you have never heard of "Sovereign Immunity". I GUARANTEE the Gummint would argue that if Apple tried to get (what kind of) meaningful relief from its Signing Key being "Leaked". And you know what? The Gummint would almost assuredly win that case. In fact, it wouldn't even survive the first Motion To Dismiss.
Try suing a Government. Tort Claims Act be damned. It is WAY hard to sue the Gumming in a Civil Case. And of course you can't charge the Gummint with any CRIME. So...?
And what "relief" could a Court grant them? An offer to have Superman spin the world backward, so that the "Leak" hadn't occurred?
THAT's why this is ridiculously dangerous. Well, ONE of the reasons...
[1]: Yes, people will say that "real men use a NAS". As of today, Macs can't do iSCSI. Macs can't do 10GigE. Macs can't do FC. The fastest I/O you are going to get is from the Apple-only M.2 wannabee SSD. Even a Thunderbolt drive barely performs better than a USB 3.0 drive.
2. Sonnet has a TWIN 10GigE to Thunderbolt adapter. Pricey, yes; but I think that might be the case with 10GigE overall. And you didn't say "cheaply"...
3. Several companies, including Atto, have both Thunderbolt to FC (as well as TB to 10GigE) adapters. Promise has a TB to TWIN 16 Gig FC adapter, too.
Sure, some of these interfaces cost as much as a cheap used-car; but OTOH, the people that need this stuff are generally not just surfing the web and posting stuff on Facebook. And again, you stated flatly that it couldn't be done; NOT that it couldn't be done on a "beer" budget.
So you see, Apple's decision to throw their design-decisions behind Thunderbolt is (finally) beginning to pay-off. They simply don't have to have a pile of dedicated connectors (not to mention the hardware to support them) for them to be able to offer (mostly through 3rd party vendors) these relatively exotic interfaces, for those who need them.
Did they go too far with the new MacBook's "one connector to rule them all" approach? Hell, yeah! But, all-in-all, Apple has made a very wise decision with ThunderBolt, and the proof is that other computer manufacturers are (finally!) beginning to agree.
The price of free choice is that some people will choose poorly. The price of restricted choice is that sometimes Apple will choose poorly on our behalf.
The problem with "choosing poorly" is that it isn't just "some people"; it is the VAST MAJORITY of people, that have better things to do with their lives than learn the ramifications of clicking "Allow".
Yes, the price of freedom is eternal vigilance; but in this particular case, you can get pwned even if you are extremely vigilant.
The only exceptions I would make are the Amazon app store and F-Droid.
Seriously, I can't see the need for ANY "exceptions" whatsoever.
Think about it: With the pretty much lassez-faire attitude that Google has about "Acceptability" for Apps in the Play Store, why oh why would ANY legit Android Developer NOT want the raw number of potential sales that comes with having your App listed on the "One Stop Shopping, and Approved, 'Safe' " Google Play Store?
So, IMHO, the fact that an App is NOT listed on Google Play should be the #1 Red Flag that something isn't exactly what it seems with an Android App.
Show me how, in any PRACTICAL sense, that I am wrong. I love "freedom" as much as the next person; but sometimes, the risks outweigh the benefits. And as long as we keep keeping private stuff and do private stuff (like banking) on our mobile devices (regardless of platform), any "benefits" must be VERY carefully considered against the possibility of being pwned.
What exactly are the benefits of using an "unofficial app store"? Pirated apps? Apps the Play store won't carry? Because all I've seen about the "unofficial" ones is they seem to be a major source of malware.
In all seriousness, and without a hint of Trolling, the main "advantage", AFAICT, is that it makes you feel superior to users of iOS, because only you have true "freedom".
Unfortunately, like in life, with "freedom" comes responsibility; and up until just recently, Android really didn't give users a fighting chance when it came to its Permissions model.
In fact, the very combination of "Sideloading" (or lack of Walled-Garden-ness) and Android's clearly pathetic "all-or-nothing" Permissions Model (who the F* thought THAT up?!?) (and which will still be in existence in the field for the next half-decade), is very much the Perfect Storm of vulnerability.
Again, without any attempt at Trolling, say what you will about iOS, the proof is in the pudding when it comes to malware on the two respective platforms.
I have a feeling Google tacitly allows Android's design to be pwnable, so that the Play store vetting is the only thing stopping your device from getting malware
If only that were true. But unfortunately, you have only a slightly better chance of actually getting a "clean", well-behaved App from the Play Store than you do from some random.ru site.
That is only if the app developer allows that in the manifest. Otherwise, the app falls back to the all or nothing permission model.
And, more importantly, only if your phone has Android 6 available, which the vast majority in actual use likely don't.
And don't go on and on about installing custom "ROMs", Cyanogen, etc. Only about 1% of Android users outside of Slashdot would even know how to do that, let alone figure out where to get a TRUSTWORTH custom "ROM", etc.
So yeah, good that Android is FINALLY getting something akin to iOS' Security Model; but in reality, it will be half-a-decade before all Android phones are running Android 6 or above.
That wouldn't work. The flash memory is encrypted with AES-256, which is impossible to brute-force with optimal quantum computers using only the resources of the Solar System until the heat death of the Universe. It seems highly unlikely for several reasons that the NSA has a way to crack it. This means that the only way to read the flash is to recreate the key.
Apple has a special piece of hardware that does AES-256 encryption and decryption. It includes a secret 256-bit random number that can't be read, and which can only be accessed by combining a PIN with it to form the key. I don't think the key is exposed either, so the only way to read the flash is to enter the correct PIN into that particular hardware. It cannot easily or reliably be copied, and if it's destroyed the flash is permanently unreadable.
Brute-forcing a 4-digit or 6-digit PIN is trivial, except that iOS permits only ten tries before wiping the random number and rendering the flash permanently unreadable, and the tries have to come through iOS on that particular device. The FBI wants a changed version of iOS that doesn't have the wipe, the lockout delay, and which accepts access through new means.
For someone pretending to have some knowledge of the situation, you have (deliberately or ignorantly) left out a few key points, to wit:
1. The phone in question is a iPhone 5C. So no "Special piece of hardware" (Secure Enclave chip). This is VERY significant.
2. As I pointed out in another post, AES256 is actually weaker than other forms of the algorithm. So it may be breakable in less time than the Universe has left.
3. When you Brute Forrce a key on a pile of static data in situ, all you need is the data. All the "random numbers" (salt), special hardware (Secure Enclave), secret keys burned into chips, user PINs, etc. are rendered moot. All you need is the encryption algorithm, and the data. Then you "solve" for "key". Since you know the (real) key-length (256 bits), and you know the encryption algorithm, the rest is simply a matter of processing speed. And with multiple copies of the data, you effectively multiply your processing speed by the number of copies.
So, sorry. The problem is trivial (tedious, but trivial), once the data is sitting outside of the control of the phone.
Not that bullshit again. If you remove the flash memory, you separate it from one of the 256 bit keys required for decryption which is locked inside the CPU.
Which still allows you to sicc your fancy Quantum computers on about 10,000 copies of the data, without fear of the "Wipe", the "Timeout", or having to have 10,000 interns manually keying in "0001, 0002, 0003..."
Um, have you read anything at all about this case? No one has asked Apple to insert a backdoor. The FBI has asked Apple to write a program that:
1. Remains in memory
2. Only Apple has access to
3. Allows the FBI to use unlimited guesses on the PIN
4. Allows the FBI to use a custom pin entry through the Thunderbolt port
Nowhere in there is a request for a backdoor on every phone, nor does the FBI want a generic backdoor they can use whenever they want. In fact, the first item is because the FBI doesn't want anyone to claim they can reverse engineer the backdoor from the firmware.
And all they have to do is to desolder the Flash, read the (encrypted) contents (I'm sure Apple will give them the pinout for the flash chip, even if proprietary (which I don't believe it is. In fact, Step 10 in this iFixit Teardown reveals that it is a Toshiba THGBX2G7B2JLA01 128 Gb (16 GB) NAND flash)), and have at it. No "remote wipe", no "10 tries", no "escalating timeout" issues.
Also NIST wanted 256 bit keys for all entrants into the AES competition for that exact reason so AES, SERPENT, and TWOFISH should all be ok unless there is a break that is discovered in any of them and then you would be screwed
What I am saying is that if there were any serious talk about quantum computers in 1972 then there's a good chance I'd have heard about it.
Sorry, not this time...
According to that esteemed, peer-reviewed (and CIA-owned) publication, Wired, David Deutsch is the father of Quantum Computing, and first postulated same "in the 1970s".
In all fairness, I never heard about Quantum Computing until the 1990s; so what do I know?
And while I agree that Apple users have not been able to take advantage of every little micro-improvement in performance, what they make they make well, and capable enough, to mire than cover 90-95% of use-cases.
But what is being pointed out here is that Oculus won't be on the Mac because they don't make decent systems for gaming. Even their highest spec system that would be the closest to a decent gaming system is rubbish for gaming (yes we all know it wasn't "designed" for it). So Luckey's point is perfectly correct, Apple makes no good computers for gaming.
Building to the minimum is essentially the problem described in the article about Mac's. All the iMac's are capable until you look at the GPU. The highest end model has a Radeon R9 M395X, which is 10% below a nVidia Geforce 970 GTX in performance. So you're spending 3000$ on a computer for looks, not performance.
You're basing your ridiculous tirade on a cherry-picked (and thoroughly unimpressive) "10%" "performance" (What Performance? Which Benchmark? Etc.) figure?!?!?
So, we're not seeing anything impressive because we're acclimated to the scene. If we could see today's hardware back in, say, 2000 (or 1995) we'd be pretty damned impressed. They're good computers, they're excellent computers. We're desensitized, so to speak, so we're not thinking of them as good computers but, really, they're fantastic machines.
That is exactly the point!
We are getting to the end of Moore's Law (actually have been there for a few years), and every manufacturer is feeling the pressure of "stuff just not getting much faster". And while I agree that Apple users have not been able to take advantage of every little micro-improvement in performance, what they make they make well, and capable enough, to mire than cover 90-95% of use-cases.
Do I secretly wish that Apple would finally make a nice, medium-sized Tower? Yeah, probably so; but never so much that I have given more than a passing thought to scratching that itch with a Hacintosh, let alone jumping to another Platform.
Quite frankly, I firmly believe that's why Apple has taken such a laissez-Faire attitude to the Hackintosh community. They could easily lock OS X to Apple hardware; but allowing the small amount of piracy that the Hack community represents gives them some very useful data on how many users are actually feeling that Apple's hardware isn't meeting their needs for one reason or another.
The files may have been open so they weren't able to back them up so it wasn't within the period of one backup. DOS/Windows isn't like UNIX. You can't work with open files. That's why Windows has to crash completely for even minor updates.
I am not a real Windows Admin, but that just isn't true. Modern backups of Windows servers take advantage of a snapshot-ting capability (I think it's called VSS) so that all files can be backed-up. I have no idea how it actually works, but I know that it does.
Yeah, because we all know jihadis use their work-supplied phones to communicate with their bretheren, and NOT their own personally-owned phones that they DESTROYED
By the way, all I ever hear about the other phones was that they were "smashed to bits"; but what does that REALLY mean?
What I am getting at is that there is every chance in the world that at least ONE of those phones' microcontrollers/flash memory was not actually "smashed" (epoxy IC packages are QUITE robust!), and with BGA packages (that have no "pins" to break), it wouldn't be that hard for someone with the Gummint's resources and budget to do some forensics on THOSE phones (which, as you allude-to, are probably the phones with the REAL data).
But yet, NO ONE brings up the other phones, other than they were "Smashed to bits". Wonder why...?
i dont get it. nsa broke ios years ago, as well as android.
actually, i do get it. the never ending pursuit of limitless power.
imagine the next target...
Depending on the number of "years ago", that could be completely irrelevant to an iPhone 5C running iOS 9.
Also, from what I have heard, the NSA doesn't exactly like to share with other Agencies.
Slashdot Mirror
Or Congress. This agency started its life with a leader that used blackmail as a standard law enforcement technique. The FBI should have been dismantled from the ground up when Hoover died as his ghost still haunts the agency in all it's actions.
John Kennedy threatened to do that with the CIA, and you see where that got him...
All though I would also argue that the government should be liable if the signing keys are leaked
Liable for what, exactly?
Apparently you have never heard of "Sovereign Immunity". I GUARANTEE the Gummint would argue that if Apple tried to get (what kind of) meaningful relief from its Signing Key being "Leaked". And you know what? The Gummint would almost assuredly win that case. In fact, it wouldn't even survive the first Motion To Dismiss.
Try suing a Government. Tort Claims Act be damned. It is WAY hard to sue the Gumming in a Civil Case. And of course you can't charge the Gummint with any CRIME. So...?
And what "relief" could a Court grant them? An offer to have Superman spin the world backward, so that the "Leak" hadn't occurred?
THAT's why this is ridiculously dangerous. Well, ONE of the reasons...
[1]: Yes, people will say that "real men use a NAS". As of today, Macs can't do iSCSI. Macs can't do 10GigE. Macs can't do FC. The fastest I/O you are going to get is from the Apple-only M.2 wannabee SSD. Even a Thunderbolt drive barely performs better than a USB 3.0 drive.
1. GlobalSAN iSCSI Initiator allows iSCSI Targets. Or if you like F/OSS Solutions, iSCSIIntiator does it, too.
2. Sonnet has a TWIN 10GigE to Thunderbolt adapter. Pricey, yes; but I think that might be the case with 10GigE overall. And you didn't say "cheaply"...
3. Several companies, including Atto, have both Thunderbolt to FC (as well as TB to 10GigE) adapters. Promise has a TB to TWIN 16 Gig FC adapter, too.
Sure, some of these interfaces cost as much as a cheap used-car; but OTOH, the people that need this stuff are generally not just surfing the web and posting stuff on Facebook. And again, you stated flatly that it couldn't be done; NOT that it couldn't be done on a "beer" budget.
So you see, Apple's decision to throw their design-decisions behind Thunderbolt is (finally) beginning to pay-off. They simply don't have to have a pile of dedicated connectors (not to mention the hardware to support them) for them to be able to offer (mostly through 3rd party vendors) these relatively exotic interfaces, for those who need them.
Did they go too far with the new MacBook's "one connector to rule them all" approach? Hell, yeah! But, all-in-all, Apple has made a very wise decision with ThunderBolt, and the proof is that other computer manufacturers are (finally!) beginning to agree.
The price of free choice is that some people will choose poorly. The price of restricted choice is that sometimes Apple will choose poorly on our behalf.
The problem with "choosing poorly" is that it isn't just "some people"; it is the VAST MAJORITY of people, that have better things to do with their lives than learn the ramifications of clicking "Allow".
Yes, the price of freedom is eternal vigilance; but in this particular case, you can get pwned even if you are extremely vigilant.
The only exceptions I would make are the Amazon app store and F-Droid.
Seriously, I can't see the need for ANY "exceptions" whatsoever.
Think about it: With the pretty much lassez-faire attitude that Google has about "Acceptability" for Apps in the Play Store, why oh why would ANY legit Android Developer NOT want the raw number of potential sales that comes with having your App listed on the "One Stop Shopping, and Approved, 'Safe' " Google Play Store?
So, IMHO, the fact that an App is NOT listed on Google Play should be the #1 Red Flag that something isn't exactly what it seems with an Android App.
Show me how, in any PRACTICAL sense, that I am wrong. I love "freedom" as much as the next person; but sometimes, the risks outweigh the benefits. And as long as we keep keeping private stuff and do private stuff (like banking) on our mobile devices (regardless of platform), any "benefits" must be VERY carefully considered against the possibility of being pwned.
What exactly are the benefits of using an "unofficial app store"? Pirated apps? Apps the Play store won't carry? Because all I've seen about the "unofficial" ones is they seem to be a major source of malware.
In all seriousness, and without a hint of Trolling, the main "advantage", AFAICT, is that it makes you feel superior to users of iOS, because only you have true "freedom".
Unfortunately, like in life, with "freedom" comes responsibility; and up until just recently, Android really didn't give users a fighting chance when it came to its Permissions model.
In fact, the very combination of "Sideloading" (or lack of Walled-Garden-ness) and Android's clearly pathetic "all-or-nothing" Permissions Model (who the F* thought THAT up?!?) (and which will still be in existence in the field for the next half-decade), is very much the Perfect Storm of vulnerability.
Again, without any attempt at Trolling, say what you will about iOS, the proof is in the pudding when it comes to malware on the two respective platforms.
I have a feeling Google tacitly allows Android's design to be pwnable, so that the Play store vetting is the only thing stopping your device from getting malware
If only that were true. But unfortunately, you have only a slightly better chance of actually getting a "clean", well-behaved App from the Play Store than you do from some random .ru site.
That is only if the app developer allows that in the manifest. Otherwise, the app falls back to the all or nothing permission model.
And, more importantly, only if your phone has Android 6 available, which the vast majority in actual use likely don't.
And don't go on and on about installing custom "ROMs", Cyanogen, etc. Only about 1% of Android users outside of Slashdot would even know how to do that, let alone figure out where to get a TRUSTWORTH custom "ROM", etc.
So yeah, good that Android is FINALLY getting something akin to iOS' Security Model; but in reality, it will be half-a-decade before all Android phones are running Android 6 or above.
That wouldn't work. The flash memory is encrypted with AES-256, which is impossible to brute-force with optimal quantum computers using only the resources of the Solar System until the heat death of the Universe. It seems highly unlikely for several reasons that the NSA has a way to crack it. This means that the only way to read the flash is to recreate the key.
Apple has a special piece of hardware that does AES-256 encryption and decryption. It includes a secret 256-bit random number that can't be read, and which can only be accessed by combining a PIN with it to form the key. I don't think the key is exposed either, so the only way to read the flash is to enter the correct PIN into that particular hardware. It cannot easily or reliably be copied, and if it's destroyed the flash is permanently unreadable.
Brute-forcing a 4-digit or 6-digit PIN is trivial, except that iOS permits only ten tries before wiping the random number and rendering the flash permanently unreadable, and the tries have to come through iOS on that particular device. The FBI wants a changed version of iOS that doesn't have the wipe, the lockout delay, and which accepts access through new means.
For someone pretending to have some knowledge of the situation, you have (deliberately or ignorantly) left out a few key points, to wit:
1. The phone in question is a iPhone 5C. So no "Special piece of hardware" (Secure Enclave chip). This is VERY significant.
2. As I pointed out in another post, AES256 is actually weaker than other forms of the algorithm. So it may be breakable in less time than the Universe has left.
3. When you Brute Forrce a key on a pile of static data in situ, all you need is the data. All the "random numbers" (salt), special hardware (Secure Enclave), secret keys burned into chips, user PINs, etc. are rendered moot. All you need is the encryption algorithm, and the data. Then you "solve" for "key". Since you know the (real) key-length (256 bits), and you know the encryption algorithm, the rest is simply a matter of processing speed. And with multiple copies of the data, you effectively multiply your processing speed by the number of copies.
So, sorry. The problem is trivial (tedious, but trivial), once the data is sitting outside of the control of the phone.
Not that bullshit again. If you remove the flash memory, you separate it from one of the 256 bit keys required for decryption which is locked inside the CPU.
Which still allows you to sicc your fancy Quantum computers on about 10,000 copies of the data, without fear of the "Wipe", the "Timeout", or having to have 10,000 interns manually keying in "0001, 0002, 0003..."
So yeah, that bullshit again.
Um, have you read anything at all about this case? No one has asked Apple to insert a backdoor. The FBI has asked Apple to write a program that:
1. Remains in memory 2. Only Apple has access to 3. Allows the FBI to use unlimited guesses on the PIN 4. Allows the FBI to use a custom pin entry through the Thunderbolt port
Nowhere in there is a request for a backdoor on every phone, nor does the FBI want a generic backdoor they can use whenever they want. In fact, the first item is because the FBI doesn't want anyone to claim they can reverse engineer the backdoor from the firmware.
And all they have to do is to desolder the Flash, read the (encrypted) contents (I'm sure Apple will give them the pinout for the flash chip, even if proprietary (which I don't believe it is. In fact, Step 10 in this iFixit Teardown reveals that it is a Toshiba THGBX2G7B2JLA01 128 Gb (16 GB) NAND flash)), and have at it. No "remote wipe", no "10 tries", no "escalating timeout" issues.
Parent may already have been replaced by a cunning Eliza program.
Say, do you have any psychological problems?
Also NIST wanted 256 bit keys for all entrants into the AES competition for that exact reason so AES, SERPENT, and TWOFISH should all be ok unless there is a break that is discovered in any of them and then you would be screwed
Yes, but isn't AES 256 actually weaker than AES 128?
Disclaimer: I am in no way a cryptologist, or a math expert
What I am saying is that if there were any serious talk about quantum computers in 1972 then there's a good chance I'd have heard about it.
Sorry, not this time...
According to that esteemed, peer-reviewed (and CIA-owned) publication, Wired, David Deutsch is the father of Quantum Computing, and first postulated same "in the 1970s".
In all fairness, I never heard about Quantum Computing until the 1990s; so what do I know?
And while I agree that Apple users have not been able to take advantage of every little micro-improvement in performance, what they make they make well, and capable enough, to mire than cover 90-95% of use-cases.
But what is being pointed out here is that Oculus won't be on the Mac because they don't make decent systems for gaming. Even their highest spec system that would be the closest to a decent gaming system is rubbish for gaming (yes we all know it wasn't "designed" for it). So Luckey's point is perfectly correct, Apple makes no good computers for gaming.
But perhaps with the introduction of the "Metal" API in OS X and iOS, Apple is trying to change that.
Yes, the non Jony Ive editions.
Jony designed those, too.
Building to the minimum is essentially the problem described in the article about Mac's. All the iMac's are capable until you look at the GPU. The highest end model has a Radeon R9 M395X, which is 10% below a nVidia Geforce 970 GTX in performance. So you're spending 3000$ on a computer for looks, not performance.
You're basing your ridiculous tirade on a cherry-picked (and thoroughly unimpressive) "10%" "performance" (What Performance? Which Benchmark? Etc.) figure?!?!?
Gotta try harder than that.
I have. It still sucks, and I have a high performance Linux laptop running a well known distribution that doesn't sleep properly.
I also have a Linux VM on a Mac that is running fine.
That's probably because most of the authors of the Linux support are using MacBook Pros. Seriously.
So, we're not seeing anything impressive because we're acclimated to the scene. If we could see today's hardware back in, say, 2000 (or 1995) we'd be pretty damned impressed. They're good computers, they're excellent computers. We're desensitized, so to speak, so we're not thinking of them as good computers but, really, they're fantastic machines.
That is exactly the point!
We are getting to the end of Moore's Law (actually have been there for a few years), and every manufacturer is feeling the pressure of "stuff just not getting much faster". And while I agree that Apple users have not been able to take advantage of every little micro-improvement in performance, what they make they make well, and capable enough, to mire than cover 90-95% of use-cases.
Do I secretly wish that Apple would finally make a nice, medium-sized Tower? Yeah, probably so; but never so much that I have given more than a passing thought to scratching that itch with a Hacintosh, let alone jumping to another Platform.
Quite frankly, I firmly believe that's why Apple has taken such a laissez-Faire attitude to the Hackintosh community. They could easily lock OS X to Apple hardware; but allowing the small amount of piracy that the Hack community represents gives them some very useful data on how many users are actually feeling that Apple's hardware isn't meeting their needs for one reason or another.
The files may have been open so they weren't able to back them up so it wasn't within the period of one backup. DOS/Windows isn't like UNIX. You can't work with open files. That's why Windows has to crash completely for even minor updates.
I am not a real Windows Admin, but that just isn't true. Modern backups of Windows servers take advantage of a snapshot-ting capability (I think it's called VSS) so that all files can be backed-up. I have no idea how it actually works, but I know that it does.
... rolled back to an old backup. As a result, we lost audit data for about 147 roots.
How the fuck are there that many changes for root CA's withing the period of one backup?
Because they only backed up the system once, and then never actually started the backups running on their regular schedule, I'll bet.
Actually, what seems to have happened is that they _did_ have a backup. But, they had to roll back to an old one.
Sounds like the excuse I'd give if I was worrying about keeping my job.
Seriously. No backup?
I know. And this is the company that has been one of the most aggressive about pushing their products into "the Cloud".
Yeah, because we all know jihadis use their work-supplied phones to communicate with their bretheren, and NOT their own personally-owned phones that they DESTROYED
By the way, all I ever hear about the other phones was that they were "smashed to bits"; but what does that REALLY mean?
What I am getting at is that there is every chance in the world that at least ONE of those phones' microcontrollers/flash memory was not actually "smashed" (epoxy IC packages are QUITE robust!), and with BGA packages (that have no "pins" to break), it wouldn't be that hard for someone with the Gummint's resources and budget to do some forensics on THOSE phones (which, as you allude-to, are probably the phones with the REAL data).
But yet, NO ONE brings up the other phones, other than they were "Smashed to bits". Wonder why...?
i dont get it. nsa broke ios years ago, as well as android. actually, i do get it. the never ending pursuit of limitless power. imagine the next target...
Depending on the number of "years ago", that could be completely irrelevant to an iPhone 5C running iOS 9.
Also, from what I have heard, the NSA doesn't exactly like to share with other Agencies.