tend to argue that sexual orientation is genetically predetermined
No-one with an ounce of sense argues that. Because the argument is irrelevant. People get to sleep with whomever they chose, provided all parties are consenting adults. End of conversation.
I don't know man, it just seems to me that doing it in firmware costs zero (yes, yes, I know software isn't free, but you know what I mean). Whereas doing it with parts, costs non-zero. I happen to work with hardware designers and firmware engineers, and they argue over cents, and square mm of board area. There is simply no way that they would waste parts and board space when the same job can be done with a few lines of code in a microcontroller that's already on the board. Have you seen the size of the board in the latest macbooks? How about in the iphone, or the ipod touch? It's bonkers. Square millimeters count. And, of course, there is no reliability issue. If that microcontroller crashes (!!) or isn't working in some other way, then your entire PC is dead, and being unable to use the power button to turn it off (or on) is a moot point. Nothing's ever actually off these days anyway, it just doesn't need to be. My most recent project used a single pushbutton to turn on and off, and of course, I did that in software. A timer when you hold the button down, driven by interrupts, and a sleep mode that waits for a falling edge on the button, with an internal pullup. Parts count, one, the button itself. Debouncing in software. I even got the pullup resistor for free, since it's just a matter of setting a register inside the chip. When the chip wakes up, it turns on a step-up DC converter in order to power the rest of the board, which would be much harder to achieve with discrete logic, and a timing capacitor, and so-on. Power usage when "off", 9 nanoamps. That's less than the internal leakage of the batteries. You just can't get that with discrete components.
Added to which is the fact that you can fiddle around with the software right up until you ship (well.... ish...), but if you decide that you need a different timing for your power switch, or different behaviour altogether, you're probably looking at a board revision. And no-one likes those.
this could easily be printed onto silicon as part of an IC package that may do many other things
Capacitors on ICs are of the order of picofarads, and cost quite alot of package space. No-one is using them for timing.
have all the hardware manufuacturers of audio input & output chipsets filter out supersonic & subsonic frequencies before the rest of the machine even sees them?
As has already been mentioned, this is exactly what all existing audio recording hardware does. Anti-aliasing filters are placed in the analog path, before digitization, and they're normally set to cut off around 20Khz, since that's the upper limit of human hearing. Leaving these filters out results in unusable audio, they are an essential component of any analog-to-digital conversion of any sort. Unless you're talking about pro-level audio recording hardware, there is no way consumer cellphones can pick up actual "ultrasound". They can pick up signals encoded in audible audio in other ways, but that couldn't be filtered out, and it isn't ultrasound.
A cellphone to unlock your house is a godsend for any hacker who wants to rob your place easily without ever leaving a trace.
If it is implemented even slightly properly, it will be much more secure than a key. Burglars don't bother with the fiddling with the key anyway, they just open things you forgot to lock, or force things that aren't very robust.
Meanwhile, back in the real world, the power button on a macbook pro is managed by this chip. In actual fact, the power button is part of the keyboard itself, on the macbook air it's not even a separate-looking button, it's just another key. Since the microcontroller part is on the board already, and in addition to the capacitor you'd also need a voltage reference, a comparator, a discharge circuit (for when the power button is released), etcetera, do you really believe that a hardware designer is going to bother with any of those parts when it can be done in firmware for zero board cost?
Reliability-wise, if the SMSC chip dies, or contains broken firmware, there would be little point in being able to "turn the laptop off" - which as we know doesn't really mean 'off', since power will still be being provided to that chip, but anyway - since the laptop would be completely dead.
So, is holding the button down for more than one second implemented by charging a capacitor until it reaches a certain voltage, and triggering the shutdown that way? That seems extremely unlikely.
The biggest difference is that the iMessage-to-iMessage conversations consist of Blue bubbles
And are free (microscopic data usage charges notwithstanding), and you can play games through them, and send all manner of pointless images, animations, drawings and make the bubbles bounce up and down. Et-Effing-Cetera. AND:
The youngsters all use various types of vendor locked in technologies.
It only works if apps support it. If it's not an expected feature available to the end user as default then it's not likely to be exploited by anyone.
Unlike Windows, where right-click is automatically built into any app, and the OS scans the executable looking for likely candidates to populate a little popup window with....
Of course it only "works if apps support it". That's a given. Apps have to support the features they support. But every single app on OSX that I've ever used supports it, especially anything aimed at developers.
Clean the fans out. The only macbook that I ever had that actually died was one of those older ones with the little catches the popped out of the screen when you closed it. Real ancient. The newer ones seem pretty damn reliable to me.
Yes, I will suggest that the switching action of transistors is caused by software. And, further, that software is caused by the switching action of transistors.
When you hold down the power key, some firmware somewhere, in a little microcontroller, starts a timer. When that timer expires, that firmware causes a transistor to be switched, which then cuts the power. It doesn't cut the power from everything, of course, since that same microcontroller is the one that causes the 'on switch' to actually turn the power back on again.
There's even a microcontroller in the charger, that's responsible for controlling current, and presumably for turning that LED orange or green, or whatever. It's an MSP430, so let's assume that they've got an MSP430 looking after the power button too. Some MSP430's have built-in touch controllers, and since they run in sleep mode on nanoamps, it's safe to keep them "on", and waiting for the power button to be touched.
"The announcements were made by a new non-profit foundation that is taking over the educational project, formerly led by the BBC."
So, it's not even the BBC anymore. And during the world's various wars, all media companies got themselves involved in propaganda. It's not as though anyone who was in charge of the BBC then, is in charge of it now.
A) "Linux 4.7 was released on Sun, 24 Jul 2016." It's released, making it not bleeding edge alpha.
B) Typical attitude. Outrageous really. Broken wifi support is not the end-users 'fault'.
C) Unsupported assumption. Maybe he's got a really good wifi adapter - you certainly don't know.
Exactly, it's like those deals that have shops guaranteeing the lowest price, or they'll beat it by ten percent. Trick is that while it may appear that different stores are selling the same thing, they actually all collude to make sure they sell *slightly* different models, meaning you can't ever claim that ten percent. Clever.
They use micro USB. Micro USB is shit, and it's always been shit, and it was a terrible terrible choice. Yes I know it's the standard for charging non-Apple phone, but that doesn't change the fact that it's uniformly terrible.
But the world is full of devices on which you can't just install whatever software you like. If a lawsuit like this succeeds, then what does that mean for all those devices? Will all manufacturers of computing devices have to provide SDKs for them? And remove whatever cryptographic restrictions have been placed on them? That would certainly make for a pretty interesting world.
Interesting theory, but I don't understand how it's supposed to work. Are you saying that your phone would only work with certain types of headphones, or something? What exactly would Apple prevent here? No-one was copying music via the headphone jack, it's not exactly a pathway that needs any copy protection on it. When people want to copy MP3s, they just press Ctrl+C and Ctrl+V.
You have to have really, really thick skin to work at that level.
Why? I should have thought that you need to be highly technically skilled, and be able to take constructive criticism when it's justified. Why you would need a 'really really really' thick skin to work on an operating system kernel is a bit beyond me.
Oh, that's right. It's Saint Linus. He gets to say what he wants to people.... why again? There are other OSs out there, with rather better designed kernels, that appear to be run perfectly well without this public naming-and-shaming that Linux seems to get off on.
There's no excuse for it. He's a bully. And I doubt very much in Andrew Morton would dare to chime in once Linus has spoken.
I've been writing code for thirty years, in a fairly wide range of languages, and on platforms from 8 bit micros to modern desktop machines. When I first discovered programming, the only language available to me was BASIC. Shortly after that I discovered C (via this weird C compiler for an Amstrad CPC that appeared to produce code that was even slower than BASIC, but that's another story). Shortly after *that* I stopped using goto, without any encouragement from Dijkstra, whom I've never read. Since that time, the only 'goto's' I've used, were jumps in assembly language. That's it.
If people's eyes glaze over when you ramble on about how "if's" and "gotos" are the same thing, it's because you're not making any sense. If they wander away mute, it's because they've heard this from you before, and have grown tired of the argument. "Ifs" and "gotos" are obviously not the same, which is why they have different names. Put your code into a function, and return from it instead of using a goto. There is no universe in which this is not a better solution. Prove me wrong:).
Because it is a very large area of the ocean in which plastic particulates float. It probably doesn't look much different from the rest of the ocean to the naked eye.
You don't realise that, because you haven't read the article, nor any of the linked articles that might help further your understanding of the problem. That's ok, you're probably busy. I've taken the following quote from here, to help you out a bit.
The debris is continuously mixed by wind and wave action and widely dispersed both over huge surface areas and throughout the top portion of the water column. It is possible to sail through the “garbage patch” area and see very little or no debris on the water’s surface. It is also difficult to estimate the size of these “patches,” because the borders and content constantly change with ocean currents and winds. Regardless of the exact size, mass, and location of the “garbage patch,” manmade debris does not belong in our oceans and waterways and must be addressed.
Slashdot Mirror
tend to argue that sexual orientation is genetically predetermined
No-one with an ounce of sense argues that. Because the argument is irrelevant. People get to sleep with whomever they chose, provided all parties are consenting adults. End of conversation.
Too much star trek or something.
I disagree. This is not possible.
I don't know man, it just seems to me that doing it in firmware costs zero (yes, yes, I know software isn't free, but you know what I mean). Whereas doing it with parts, costs non-zero. I happen to work with hardware designers and firmware engineers, and they argue over cents, and square mm of board area. There is simply no way that they would waste parts and board space when the same job can be done with a few lines of code in a microcontroller that's already on the board. Have you seen the size of the board in the latest macbooks? How about in the iphone, or the ipod touch? It's bonkers. Square millimeters count. And, of course, there is no reliability issue. If that microcontroller crashes (!!) or isn't working in some other way, then your entire PC is dead, and being unable to use the power button to turn it off (or on) is a moot point. Nothing's ever actually off these days anyway, it just doesn't need to be. My most recent project used a single pushbutton to turn on and off, and of course, I did that in software. A timer when you hold the button down, driven by interrupts, and a sleep mode that waits for a falling edge on the button, with an internal pullup. Parts count, one, the button itself. Debouncing in software. I even got the pullup resistor for free, since it's just a matter of setting a register inside the chip. When the chip wakes up, it turns on a step-up DC converter in order to power the rest of the board, which would be much harder to achieve with discrete logic, and a timing capacitor, and so-on. Power usage when "off", 9 nanoamps. That's less than the internal leakage of the batteries. You just can't get that with discrete components.
Added to which is the fact that you can fiddle around with the software right up until you ship (well.... ish...), but if you decide that you need a different timing for your power switch, or different behaviour altogether, you're probably looking at a board revision. And no-one likes those.
this could easily be printed onto silicon as part of an IC package that may do many other things
Capacitors on ICs are of the order of picofarads, and cost quite alot of package space. No-one is using them for timing.
have all the hardware manufuacturers of audio input & output chipsets filter out supersonic & subsonic frequencies before the rest of the machine even sees them?
As has already been mentioned, this is exactly what all existing audio recording hardware does. Anti-aliasing filters are placed in the analog path, before digitization, and they're normally set to cut off around 20Khz, since that's the upper limit of human hearing. Leaving these filters out results in unusable audio, they are an essential component of any analog-to-digital conversion of any sort. Unless you're talking about pro-level audio recording hardware, there is no way consumer cellphones can pick up actual "ultrasound". They can pick up signals encoded in audible audio in other ways, but that couldn't be filtered out, and it isn't ultrasound.
A cellphone to unlock your house is a godsend for any hacker who wants to rob your place easily without ever leaving a trace.
If it is implemented even slightly properly, it will be much more secure than a key. Burglars don't bother with the fiddling with the key anyway, they just open things you forgot to lock, or force things that aren't very robust.
For real?
Meanwhile, back in the real world, the power button on a macbook pro is managed by this chip. In actual fact, the power button is part of the keyboard itself, on the macbook air it's not even a separate-looking button, it's just another key. Since the microcontroller part is on the board already, and in addition to the capacitor you'd also need a voltage reference, a comparator, a discharge circuit (for when the power button is released), etcetera, do you really believe that a hardware designer is going to bother with any of those parts when it can be done in firmware for zero board cost?
Reliability-wise, if the SMSC chip dies, or contains broken firmware, there would be little point in being able to "turn the laptop off" - which as we know doesn't really mean 'off', since power will still be being provided to that chip, but anyway - since the laptop would be completely dead.
Possibly this might entail removing some screws.
So, is holding the button down for more than one second implemented by charging a capacitor until it reaches a certain voltage, and triggering the shutdown that way? That seems extremely unlikely.
The biggest difference is that the iMessage-to-iMessage conversations consist of Blue bubbles
And are free (microscopic data usage charges notwithstanding), and you can play games through them, and send all manner of pointless images, animations, drawings and make the bubbles bounce up and down. Et-Effing-Cetera. AND:
The youngsters all use various types of vendor locked in technologies.
Is basically true.
Or, they might start to use nano instead....
Crazy people.
It only works if apps support it. If it's not an expected feature available to the end user as default then it's not likely to be exploited by anyone.
Unlike Windows, where right-click is automatically built into any app, and the OS scans the executable looking for likely candidates to populate a little popup window with....
Of course it only "works if apps support it". That's a given. Apps have to support the features they support. But every single app on OSX that I've ever used supports it, especially anything aimed at developers.
Clean the fans out. The only macbook that I ever had that actually died was one of those older ones with the little catches the popped out of the screen when you closed it. Real ancient. The newer ones seem pretty damn reliable to me.
Yes, I will suggest that the switching action of transistors is caused by software. And, further, that software is caused by the switching action of transistors.
When you hold down the power key, some firmware somewhere, in a little microcontroller, starts a timer. When that timer expires, that firmware causes a transistor to be switched, which then cuts the power. It doesn't cut the power from everything, of course, since that same microcontroller is the one that causes the 'on switch' to actually turn the power back on again.
There's even a microcontroller in the charger, that's responsible for controlling current, and presumably for turning that LED orange or green, or whatever. It's an MSP430, so let's assume that they've got an MSP430 looking after the power button too. Some MSP430's have built-in touch controllers, and since they run in sleep mode on nanoamps, it's safe to keep them "on", and waiting for the power button to be touched.
"The announcements were made by a new non-profit foundation that is taking over the educational project, formerly led by the BBC."
So, it's not even the BBC anymore. And during the world's various wars, all media companies got themselves involved in propaganda. It's not as though anyone who was in charge of the BBC then, is in charge of it now.
What "current actions"?
A) "Linux 4.7 was released on Sun, 24 Jul 2016." It's released, making it not bleeding edge alpha. B) Typical attitude. Outrageous really. Broken wifi support is not the end-users 'fault'. C) Unsupported assumption. Maybe he's got a really good wifi adapter - you certainly don't know.
Exactly, it's like those deals that have shops guaranteeing the lowest price, or they'll beat it by ten percent. Trick is that while it may appear that different stores are selling the same thing, they actually all collude to make sure they sell *slightly* different models, meaning you can't ever claim that ten percent. Clever.
They use micro USB. Micro USB is shit, and it's always been shit, and it was a terrible terrible choice. Yes I know it's the standard for charging non-Apple phone, but that doesn't change the fact that it's uniformly terrible.
But the world is full of devices on which you can't just install whatever software you like. If a lawsuit like this succeeds, then what does that mean for all those devices? Will all manufacturers of computing devices have to provide SDKs for them? And remove whatever cryptographic restrictions have been placed on them? That would certainly make for a pretty interesting world.
Interesting theory, but I don't understand how it's supposed to work. Are you saying that your phone would only work with certain types of headphones, or something? What exactly would Apple prevent here? No-one was copying music via the headphone jack, it's not exactly a pathway that needs any copy protection on it. When people want to copy MP3s, they just press Ctrl+C and Ctrl+V.
He is leading the most complex and arguably most successful software project in the world.
The Linux Kernel is very far from being the most complex software project in the world.
You have to have really, really thick skin to work at that level.
Why? I should have thought that you need to be highly technically skilled, and be able to take constructive criticism when it's justified. Why you would need a 'really really really' thick skin to work on an operating system kernel is a bit beyond me.
Oh, that's right. It's Saint Linus. He gets to say what he wants to people.... why again? There are other OSs out there, with rather better designed kernels, that appear to be run perfectly well without this public naming-and-shaming that Linux seems to get off on.
There's no excuse for it. He's a bully. And I doubt very much in Andrew Morton would dare to chime in once Linus has spoken.
I've been writing code for thirty years, in a fairly wide range of languages, and on platforms from 8 bit micros to modern desktop machines. When I first discovered programming, the only language available to me was BASIC. Shortly after that I discovered C (via this weird C compiler for an Amstrad CPC that appeared to produce code that was even slower than BASIC, but that's another story). Shortly after *that* I stopped using goto, without any encouragement from Dijkstra, whom I've never read. Since that time, the only 'goto's' I've used, were jumps in assembly language. That's it.
If people's eyes glaze over when you ramble on about how "if's" and "gotos" are the same thing, it's because you're not making any sense. If they wander away mute, it's because they've heard this from you before, and have grown tired of the argument. "Ifs" and "gotos" are obviously not the same, which is why they have different names. Put your code into a function, and return from it instead of using a goto. There is no universe in which this is not a better solution. Prove me wrong :).
It's tiny fragments of plastic floating in the sea. Please try to keep up.
Because it is a very large area of the ocean in which plastic particulates float. It probably doesn't look much different from the rest of the ocean to the naked eye.
You don't realise that, because you haven't read the article, nor any of the linked articles that might help further your understanding of the problem. That's ok, you're probably busy. I've taken the following quote from here, to help you out a bit.
The debris is continuously mixed by wind and wave action and widely dispersed both over huge surface areas and throughout the top portion of the water column. It is possible to sail through the “garbage patch” area and see very little or no debris on the water’s surface. It is also difficult to estimate the size of these “patches,” because the borders and content constantly change with ocean currents and winds. Regardless of the exact size, mass, and location of the “garbage patch,” manmade debris does not belong in our oceans and waterways and must be addressed.
I did try that, but I could never get it to work properly. Maybe it's time I tried again.
You can do that? I have MacOS server installed, can I just build an iOS app and put it on my kids ipads (for instance) alongside all their other apps?