When you say 'rail', I start to wonder if you actually did this. The two communicating parts that you're talking about are on the same SoC. The secure enclave is given a key to try. If it's wrong, it increments a counter. If it's correct, it resets the counter. If the counter reaches a threshold value, it deletes the key that's stored internally to the secure enclave. That key never exists outside of the secure enclave. You need a bit more than an an oscilloscope to mess up a signal travelling inside a chip.
Right, and that means make sure that a lawyer has read the EULA for every single piece of software that you run (including all software updates that have a new license). Can't afford that? Maybe that proprietary software has too high a TCO for you and you should look elsewhere.
What insane argument is that? That's like saying there are no Windows apps running on Linux without using WINE, so there is no interoperability argument for writing WINE.
No, it's like saying that there are no Windows apps running on Linux using WINE, which would be false. Android Java does not implement any of the defined subsets of the Java standard library and so you can not take an unmodified Java program and run it on Android. They could have made a case that there are Java libraries that are used on both Android and J2SE implementations, but that would have required that they explain libraries to a non-technical court.
Is there actually much practical difference between Mercurial and Git?
Yes, the mercurial CLI is just about good enough, so people use it. The git CLI is an abomination that is so bad that people are willing to invest time avoiding using it. As a result, there are a lot of nice GUI tools for working with git.
Google people have been investing a lot of time and effort over the last year or so in Windows support for clang. Interestingly, Microsoft has also been contributing and has some odd combinations (MSVC++ front end with LLVM back end and Clang front end with MSVC++ back end) working.
A lot of these are likely easier with mobile devices. There's a fairly limited set of CPU and GPU combinations for Android and an even more limited set for iOS. Not sure about iOS, but Android devices pretty much all support bluetooth gamepads (and most support HDMI output, for big screen playback), so if a company like Sony or Nintendo wants to release a standard gamepad for their games then it would be quite easy to do.
Emulation is not necessary for Sony original titles - they have the source code, they can do a port. Most modern mobile devices have a more powerful GPU than the PS2 (and an increasing number have a more powerful on than the PS3), though things that use the Cell SPU are probably harder to port.
No, the lesson is that if you have a single point of failure then you should be aware of it. It's one of the (many) reasons that language-specific package managers are a terrible idea. If you depend on a library then you want something that is packaged as a library, can be easily mirrored for your own local development, and will show up in package managers for your targets or can be statically linked / bundled when deploying on a platform without a package manager.
What is being said is "dependencies are bad, you shouldn't have them"
It's true. It's also true to say 'reinventing the wheel is bad, you shouldn't do it'. Eventually, you have to pick the lesser of two evils, but neither solution is particularly enjoyable.
GIF is pretty simple. PNG is more complex. JPEG is a little more complex than GIF, if you can do even the core mathematical libraries for JPEG2000 in 160 hours I'd be moderately impressed. TIFF includes vast numbers of possible image encodings, including some weird and wonderful uncompressed encodings (floating point channels, weird channel interleavings) and many compressed variants (including embedded other formats, layers, different resolution previews, and so on). Oh, and most existing image libraries have been reasonably well fuzz-tested recently and had hundreds of security holes fixed, because parsing binary formats in C without introducing exploits turns out to be hard.
I remember the kids who went into "climate science" in college
Really? Because most places don't offer it as an undergraduate course, it's something that people generally move into as a PhD specialism, or often later after a PhD, typically in physics (or, often, computer science, because building models of complex systems involves a lot of algorithmic knowledge).
The Christians only stopped being in the news in the UK every day killing each other for being the wrong flavour of Christian when some time around 2001 the citizens of New York suddenly decided that funding terrorism wasn't cool anymore.
It uses exactly the same EMV protocol as the chip on my credit card. The only difference is that my credit card is a lot more portable than my phone and doesn't need a battery.
Go on, just once, it will give the conspiracy theorists something to point to for the next 3-5 years and keep them entertained. Then you can auto-mod anything that links to the admission to -1, thus keeping the trolls away from the rest of us and also giving them something else to complain about. It's win-win!
Mir exists, in part, because there actually are some real problems with X11, such as the complete lack of anything resembling security on the input path. These problems were not things that the Wayland developers decided to fix. I'm not a huge fan of Mir, but if you're going to replace X11 then replacing it with Wayland is at best a sideways step.
Huh? SMS Backup+ works fine with any IMAP server. I'm using it with my own server and have not had any problems. All of my SMSs show up in my mail client as I'd expect.
I sort of see Japan as the place to look how to make something new , whereas I look at China as the place to look how to make something cheap and in high volume
You might ask someone who remembers the '70s how they used to see Japan.
No, performance still sucks because the designers still have absolutely no idea how to design a network filesystem. If they bothered to read the AFS papers then they'd know that all of the 'hard' problems that they're struggling with were solved decades ago (and have open source implementations). Like so many other related projects, they're far too concerned about building a platform without thinking about the underlying protocols. If I had to pick one out of open source or open protocols, I'd pick open protocols: if you start from there then you're far more likely to end up with both.
It is a bit of a stretch. There was a nice entry into the IOCC a few years ago that was a program that was valid as C program, a shell script, or a makefile. Running it as either a shell script or makefile would compile the C program, which would then print its output. There's been some interesting recent research involving isolating instructions that are NOPs on various architectures and writing exploit code that is a valid executable on both x86 and ARM (it doesn't have to be long, because you can encode a jump to the architecture-specific version in the portable code).
It's worth noting that this is even (almost) the official and documented way of writing a cross-architecture Windows binary: you have a little.NET stub that P/Invoke's the native binary for the architecture that it detects.
The problem with nuclear weapons is that the fallout is not contained. Hitting North Korea, without effectively hitting China is very difficult. If the Chinese say that dropping a nuke within a certain radius of their borders would be considered an act of war, do you think that a sane US President would still do so? Retaliation is far likely to be using conventional weapons (and conventional missiles can still do a lot of damage). China would probably welcome a chance to show off how much damage its military can do in a conventional conflict against an opponent that is unambiguously the bad guy to the international audience.
Slashdot Mirror
When you say 'rail', I start to wonder if you actually did this. The two communicating parts that you're talking about are on the same SoC. The secure enclave is given a key to try. If it's wrong, it increments a counter. If it's correct, it resets the counter. If the counter reaches a threshold value, it deletes the key that's stored internally to the secure enclave. That key never exists outside of the secure enclave. You need a bit more than an an oscilloscope to mess up a signal travelling inside a chip.
Right, and that means make sure that a lawyer has read the EULA for every single piece of software that you run (including all software updates that have a new license). Can't afford that? Maybe that proprietary software has too high a TCO for you and you should look elsewhere.
What insane argument is that? That's like saying there are no Windows apps running on Linux without using WINE, so there is no interoperability argument for writing WINE.
No, it's like saying that there are no Windows apps running on Linux using WINE, which would be false. Android Java does not implement any of the defined subsets of the Java standard library and so you can not take an unmodified Java program and run it on Android. They could have made a case that there are Java libraries that are used on both Android and J2SE implementations, but that would have required that they explain libraries to a non-technical court.
Is there actually much practical difference between Mercurial and Git?
Yes, the mercurial CLI is just about good enough, so people use it. The git CLI is an abomination that is so bad that people are willing to invest time avoiding using it. As a result, there are a lot of nice GUI tools for working with git.
Google people have been investing a lot of time and effort over the last year or so in Windows support for clang. Interestingly, Microsoft has also been contributing and has some odd combinations (MSVC++ front end with LLVM back end and Clang front end with MSVC++ back end) working.
A lot of these are likely easier with mobile devices. There's a fairly limited set of CPU and GPU combinations for Android and an even more limited set for iOS. Not sure about iOS, but Android devices pretty much all support bluetooth gamepads (and most support HDMI output, for big screen playback), so if a company like Sony or Nintendo wants to release a standard gamepad for their games then it would be quite easy to do.
Emulation is not necessary for Sony original titles - they have the source code, they can do a port. Most modern mobile devices have a more powerful GPU than the PS2 (and an increasing number have a more powerful on than the PS3), though things that use the Cell SPU are probably harder to port.
It's because women are smaller.
No, the lesson is that if you have a single point of failure then you should be aware of it. It's one of the (many) reasons that language-specific package managers are a terrible idea. If you depend on a library then you want something that is packaged as a library, can be easily mirrored for your own local development, and will show up in package managers for your targets or can be statically linked / bundled when deploying on a platform without a package manager.
What is being said is "dependencies are bad, you shouldn't have them"
It's true. It's also true to say 'reinventing the wheel is bad, you shouldn't do it'. Eventually, you have to pick the lesser of two evils, but neither solution is particularly enjoyable.
it's unlikely to be worth re-implementing them
Very true.
I remember the kids who went into "climate science" in college
Really? Because most places don't offer it as an undergraduate course, it's something that people generally move into as a PhD specialism, or often later after a PhD, typically in physics (or, often, computer science, because building models of complex systems involves a lot of algorithmic knowledge).
The Christians only stopped being in the news in the UK every day killing each other for being the wrong flavour of Christian when some time around 2001 the citizens of New York suddenly decided that funding terrorism wasn't cool anymore.
It uses exactly the same EMV protocol as the chip on my credit card. The only difference is that my credit card is a lot more portable than my phone and doesn't need a battery.
Go on, just once, it will give the conspiracy theorists something to point to for the next 3-5 years and keep them entertained. Then you can auto-mod anything that links to the admission to -1, thus keeping the trolls away from the rest of us and also giving them something else to complain about. It's win-win!
Mir exists, in part, because there actually are some real problems with X11, such as the complete lack of anything resembling security on the input path. These problems were not things that the Wayland developers decided to fix. I'm not a huge fan of Mir, but if you're going to replace X11 then replacing it with Wayland is at best a sideways step.
Huh? SMS Backup+ works fine with any IMAP server. I'm using it with my own server and have not had any problems. All of my SMSs show up in my mail client as I'd expect.
Because people now see China as they saw Japan in the '70s, right before Japan ate their lunch in the '80s.
I sort of see Japan as the place to look how to make something new , whereas I look at China as the place to look how to make something cheap and in high volume
You might ask someone who remembers the '70s how they used to see Japan.
No, performance still sucks because the designers still have absolutely no idea how to design a network filesystem. If they bothered to read the AFS papers then they'd know that all of the 'hard' problems that they're struggling with were solved decades ago (and have open source implementations). Like so many other related projects, they're far too concerned about building a platform without thinking about the underlying protocols. If I had to pick one out of open source or open protocols, I'd pick open protocols: if you start from there then you're far more likely to end up with both.
It is a bit of a stretch. There was a nice entry into the IOCC a few years ago that was a program that was valid as C program, a shell script, or a makefile. Running it as either a shell script or makefile would compile the C program, which would then print its output. There's been some interesting recent research involving isolating instructions that are NOPs on various architectures and writing exploit code that is a valid executable on both x86 and ARM (it doesn't have to be long, because you can encode a jump to the architecture-specific version in the portable code).
It's worth noting that this is even (almost) the official and documented way of writing a cross-architecture Windows binary: you have a little .NET stub that P/Invoke's the native binary for the architecture that it detects.
Once all the Germans were warlike and mean,
But that couldn't happen again.
We taught them a lesson in 1918
And they've hardly bothered us since then.
The problem with nuclear weapons is that the fallout is not contained. Hitting North Korea, without effectively hitting China is very difficult. If the Chinese say that dropping a nuke within a certain radius of their borders would be considered an act of war, do you think that a sane US President would still do so? Retaliation is far likely to be using conventional weapons (and conventional missiles can still do a lot of damage). China would probably welcome a chance to show off how much damage its military can do in a conventional conflict against an opponent that is unambiguously the bad guy to the international audience.
No. They're not a problem for the country that they're leaving, they're a problem for the country that they're arriving in.
Yup, and then shop them to the MPAA for copyright infringement!