Yes, indeed. And, for that matter, Google tools usually become much less interesting once they hit 1.0. There are a bunch of folks who are probably more excited to be using = 1.0
Heck Mac OS X started at version 10.0 and I can tell you from bitter experience debugging its kernel that it was beta.
Wow James, when Q explained this to you didn't he tell you not to bring attention to it ? Now SPECTRE knows exactly where you are hiding the blueprints to the nuclear submarine base.
Or is this just a ruse to hide your porn habit from your spouse ?
Yes indeed. And it worth remembering exactly what Turing meant when he proposed this "test" - certainly not the competition that now bears his name. He was making a philosophical comment on the nature of intelligence. Essentially: if you can't tell the difference between a computer and a person, then who is to say that the computer is not intelligent. To be more explicit: If A possesses the quality B. If you have no means of distinguishing C from A, then you are unable to answer the question, does C posses the quality B. You might as well assume that C DOES posses B.
Turing did not know how to define intelligence any more than you or I. (And I guarantee that he was much smarter than you and I). This is not a test for intelligence at all.
As a person with a relative who has severe Alzheimer's Disease. I might as well define intelligence as the ability to look me in the eye and hold my gaze. That's what makes me believe that she still posesses "intelligence". Think about it, why not my mouth or my feet ? The eyes are what I am using to look at her...if that is why she is looking at them, then there is a whole heck of sentience left.
Ok, I dropped a word: "I worked on drivers for broadcast video editing and effects [hardware]" Specifically PCI X and Firewire based codecs and effects pipelines (custom ASICs). Actually I also worked on custom UW SCSI drivers that fed the codecs through DMA. You'll find that the high end broadcast Non-Linear Editors are still hardware based. I thought that the fact that I was talking about kernel resident hardware drivers was patently obvious from the rest of my comments.
For what it is worth, framebuffer devices ("Graphics Cards") and Human Interface Devices ("drawing tablets") are ubiquitous devices that have good support in IOKit. Unlike, say, a custom 3D effects ASIC.
Microsoft has no control over the shit quality of drivers released by hardware manufacturers.
Then again, TFA admits that Apple does a better job preventing third parties from giving the user a poor experience.
I think that there is some truth to this. In the case of the kernel, Apple's IOKit uses embedded C++. For ubiquitous drivers, this is an absolute godsend, most of the work of a driver is already done for you. Writing a driver largely consists of just writing the details that are specific to your device. It's Apple's job to build the framework for devices of your class, not yours.
Secondly, Apple's definition of a real time operating system is subtle, but significant: it's difficult to do things in kernel mode that will degrade user mode (application) experience. It's a pain in the ass sometimes: doing work on a true primary hardware interrupt is hard to achieve, but there are few devices that really need this. The upshot is that a "badly" written driver is unlikely to stop iTunes from playing back media.
Thirdly, certain classes of driver can be entirely user mode. This is a good thing !
Apple's approach is not perfect, if you need real time performance in the kernel then it's truly hard work, and there's little documentation to support you. Thankfully, such drivers are not common. (I worked on drivers for broadcast video editing and effects, IOKit doesn't provide a base class to derive such drivers from !)
My point is not (just) fanboiism. I believe that there is a lot that can be done in architecting an API to encourage well behaved drivers, and other code. Microsoft could do something about it. It would just take dumping an entire driver model and replacing it something better though.
This is the internet. If you're not browsing with your credibility threshold set to "999,999/1,000,000 offers are scams," then I've got some little blue pills I'd like to sell you - I'm the manager of an implausibly sounding bank in Nigeria, and if you'd just pay some advance fees, I can increase your bust size while you make money at home !
How much do you want to bet Paystar has the same "donation"? Erm, a million, billion dollars ?
You are smoking crack. Apple is not the same threat to MS that Linux is. MS profits from Apple!
MS's margins on (legitimate) sales to Mac users have higher average margins than average sales to typical PC users: Firstly, there's no such thing as an OEM version of XP or Vista for Mac. Most OS sales in the PC world are low margin OEM bundles with HW. A Mac user must purchase a full retail price version of Windows to install on their Mac. Most unit sales of Office for the Mac are, again, individual sales. Most unit sales of Office for Windows are Corporate.
Then we can go into the famous $150m "donation" that MS made to Apple, and the patent cross licensing agreement between the two companies...
What time of day did they measure it ? The East Coast wakes up 3 hours before the West Coast. We have the internet pretty much to ourselves until noon...
The spammers are quite reasonable. Once your male has been enhanced, they don't bother trying to sell you any more. Whatever you do though, don't confuse them by buying the bust enhancement cream as well...
You're still within the first 30 days. If you don't like the EULA, take the phone back. Diress ? Assuming you mean duress, then what exactly was the duress ? A desperate need to own the iPhone 3G ? Or did the assistant literally hold a gun to your head.
Bah, no need for the FUD. You can run whatever you like on the iPhone. Apple can't disable an application (with this "feature") unless they know about it. The SDK let's you write and run whatever apps you like. You just can't distribute them through the Apple store without obeying the rules.
This is to stop you from, say, downloading an app from the Apple store that secretly reports your location to, I dunno, an advertiser, Microsoft, your Mum, or a jealous spouse...
Yeah, the bootloader operates before anything remotely resembling DNS operates. It would be immensely difficult to prevent someone from restoring the OS. So, not "bricked" but hung until restored.
Apple's not perfect by any means, but this is just going to piss people off: not Apple, so much as Apple's customers and towards the FSF I imagine.
Sure DRM sucks, but in my opinion it was the price Apple had to pay in order to get the record industry to begin to see sense.
Any, while we're at it, Apple stores suit me just fine. I've bought Apple equipment in three continents and gotten over the counter warranty replacement twice at genius bars on the other side of the planet. Show me another consumer electronics company that can do that.
(Not Sony Ericsson, they won't touch my Euro cellphone in the US)
Perhaps the point here is to piss off some overzealous security guards into escorting them out of Apple stores and turn it into an "Apple hates civil rights as well as fair use rights" scandal ?
By reading through the file identifying them, which entails file IO which we've established is slower than decode. And, for reasons pointed out elsewhere, don't try this with MPEG4, H.264 or DiVX. Which pretty much leaves MPEG1 and 2.
The point is that you'd do so much work parsing the stream in order to chunk it that you might as well just decode it. Particularly if you can spend the time while the file IO is blocked decoding frames.
That's the model that the PHBs want, give it to them. Export as ascii so you can actually build the damn thing, and make sure that every single automated formatting feature is turned off: smart quotes, spell checking, etc...Word tracks changes creditably, of course as the chorus correctly says, version control is where this ought to be done, not in the file itself. You don't need the same degree of change tracking in a source file that you do in a legal document, but heck I'm a programmer not a lawyer, or a PHB.
Also, just to stick it to the man, make sure that you order MS Office in the most expensive way possible...when they ask why it costs so much, tell them it's the "idiot tax".
he is pretty high on the list of people that would probably know about it
Especially if contact had been established before the mission. Presumably if that were the case, he would need to be forewarned about prior claims to the moon's cheese.
My personal opinion is that in all likelihood, earth is by no means unusual, life is probably rampant throughout the universe, but that's not to say anyone has, or will, visit us on earth: I agree, he's gone off the rails.
As other commenters have said, decoding video is not, per se, a trivially parallelized algorithm. Especially for modern codecs with lots of temporal encoding. MJPEG would be easily parallelized, buy you'd have to be dealing with fairly ancient sources...MediaComposer 1 for instance.
However, there are different classes of "video app" that are good targets for parallelization. Real world video editing for instance: consider multiple streams of video with overlays, rotations, effects etc. Video and audio decoding can happen in parallel, you can pipeline the effects stages so that each effect is handed off to another core. Modern video editing systems do this with aplomb.
I'm from the commercial end of this so, I can't comment much on open source alternatives. But I will say that a lot of the algorithms in certain products are highly tuned to the particular CPU type. And they're smart enough to distribute work across only as many cores as actually exist.
Finally. Don't forget that optimization is hard. You have to consider the speed of the hard drive, the cost of sharing data between threads and cpu caches and a bunch of other real constraints. Any half decent cpu of the last five years or so can easily decode most video faster than it can be read and written to disk. So long as this is true, you won't get any benefit from parallelization.
Rubbish. A long time ago I lived in RIchmond (West London). Richmond Bridge crosses the Thames and has a sharp turn on the other bank. Every year there were a predictable number of accidents due to motorists speeding over the bridge and hitting pedestrians. Including a number of deaths. They installed one of the first Gatso speed cameras on the bridge, and guess what ? No more accidents.
Even I as someone who will exceed the speed limit on occasion applaud this instance. I despise cameras positioned solely to raise revenue, which today seems to be the norm. But this situation made complete sense.
A good question. I don't have to reboot that often, but sometimes that's what cures my problems. However, I can't honestly say that means that the problem is with the router. It could be with devices upstream or downstream - rebooting the router may be the trick that resets the state machine for the connected device.
I never have to reboot my TiVo, my cell phone, my PSP, my STBs etc...
Slashdot Mirror
It's not the folks descended from criminals that worry me. It's the folks who are descended from the prison wardens who cause all the trouble.
You are freaking joking right ? While you are at it, do all your surface mount soldering with a screwdriver heated over a gas torch.
Yes, indeed. And, for that matter, Google tools usually become much less interesting once they hit 1.0.
There are a bunch of folks who are probably more excited to be using = 1.0
Heck Mac OS X started at version 10.0 and I can tell you from bitter experience debugging its kernel that it was beta.
the boston tea party wasn't an act of terrorism?
Hardly. Civil disobedience certainly. Who was terrorized by this act ? The Indian tribes that the perpetrators tried to impersonate ?
The notorious first shot at tbe Battle of Lexington is a better candidate for accusations of terrorism.
Wow James, when Q explained this to you didn't he tell you not to bring attention to it ? Now SPECTRE knows exactly where you are hiding the blueprints to the nuclear submarine base.
Or is this just a ruse to hide your porn habit from your spouse ?
Yes indeed. And it worth remembering exactly what Turing meant when he proposed this "test" - certainly not the competition that now bears his name.
He was making a philosophical comment on the nature of intelligence. Essentially: if you can't tell the difference between a computer and a person, then who is to say that the computer is not intelligent. To be more explicit: If A possesses the quality B. If you have no means of distinguishing C from A, then you are unable to answer the question, does C posses the quality B.
You might as well assume that C DOES posses B.
Turing did not know how to define intelligence any more than you or I. (And I guarantee that he was much smarter than you and I). This is not a test for intelligence at all.
As a person with a relative who has severe Alzheimer's Disease. I might as well define intelligence as the ability to look me in the eye and hold my gaze. That's what makes me believe that she still posesses "intelligence". Think about it, why not my mouth or my feet ? The eyes are what I am using to look at her...if that is why she is looking at them, then there is a whole heck of sentience left.
Ok, I dropped a word: "I worked on drivers for broadcast video editing and effects [hardware]"
Specifically PCI X and Firewire based codecs and effects pipelines (custom ASICs). Actually I also worked on custom UW SCSI drivers that fed the codecs through DMA. You'll find that the high end broadcast Non-Linear Editors are still hardware based. I thought that the fact that I was talking about kernel resident hardware drivers was patently obvious from the rest of my comments.
For what it is worth, framebuffer devices ("Graphics Cards") and Human Interface Devices ("drawing tablets") are ubiquitous devices that have good support in IOKit. Unlike, say, a custom 3D effects ASIC.
Microsoft has no control over the shit quality of drivers released by hardware manufacturers.
Then again, TFA admits that Apple does a better job preventing third parties from giving the user a poor experience.
I think that there is some truth to this. In the case of the kernel, Apple's IOKit uses embedded C++. For ubiquitous drivers, this is an absolute godsend, most of the work of a driver is already done for you. Writing a driver largely consists of just writing the details that are specific to your device. It's Apple's job to build the framework for devices of your class, not yours.
Secondly, Apple's definition of a real time operating system is subtle, but significant: it's difficult to do things in kernel mode that will degrade user mode (application) experience. It's a pain in the ass sometimes: doing work on a true primary hardware interrupt is hard to achieve, but there are few devices that really need this. The upshot is that a "badly" written driver is unlikely to stop iTunes from playing back media.
Thirdly, certain classes of driver can be entirely user mode. This is a good thing !
Apple's approach is not perfect, if you need real time performance in the kernel then it's truly hard work, and there's little documentation to support you. Thankfully, such drivers are not common. (I worked on drivers for broadcast video editing and effects, IOKit doesn't provide a base class to derive such drivers from !)
My point is not (just) fanboiism. I believe that there is a lot that can be done in architecting an API to encourage well behaved drivers, and other code. Microsoft could do something about it. It would just take dumping an entire driver model and replacing it something better though.
This is the internet. If you're not browsing with your credibility threshold set to "999,999/1,000,000 offers are scams," then I've got some little blue pills I'd like to sell you - I'm the manager of an implausibly sounding bank in Nigeria, and if you'd just pay some advance fees, I can increase your bust size while you make money at home !
How much do you want to bet Paystar has the same "donation"?
Erm, a million, billion dollars ?
You are smoking crack. Apple is not the same threat to MS that Linux is. MS profits from Apple!
MS's margins on (legitimate) sales to Mac users have higher average margins than average sales to typical PC users:
Firstly, there's no such thing as an OEM version of XP or Vista for Mac. Most OS sales in the PC world are low margin OEM bundles with HW.
A Mac user must purchase a full retail price version of Windows to install on their Mac.
Most unit sales of Office for the Mac are, again, individual sales. Most unit sales of Office for Windows are Corporate.
Then we can go into the famous $150m "donation" that MS made to Apple, and the patent cross licensing agreement between the two companies...
What time of day did they measure it ?
The East Coast wakes up 3 hours before the West Coast. We have the internet pretty much to ourselves until noon...
The spammers are quite reasonable. Once your male has been enhanced, they don't bother trying to sell you any more.
Whatever you do though, don't confuse them by buying the bust enhancement cream as well...
You mean:
sudo echo "bad.guys.ip.address iphone-services.apple.com" > /etc/hosts
?
You're still within the first 30 days. If you don't like the EULA, take the phone back.
Diress ? Assuming you mean duress, then what exactly was the duress ? A desperate need to own the iPhone 3G ? Or did the assistant literally hold a gun to your head.
Bah, no need for the FUD. You can run whatever you like on the iPhone. Apple can't disable an application (with this "feature") unless they know about it. The SDK let's you write and run whatever apps you like. You just can't distribute them through the Apple store without obeying the rules.
This is to stop you from, say, downloading an app from the Apple store that secretly reports your location to, I dunno, an advertiser, Microsoft, your Mum, or a jealous spouse...
Yeah, the bootloader operates before anything remotely resembling DNS operates. It would be immensely difficult to prevent someone from restoring the OS. So, not "bricked" but hung until restored.
Apple's not perfect by any means, but this is just going to piss people off: not Apple, so much as Apple's customers and towards the FSF I imagine.
Sure DRM sucks, but in my opinion it was the price Apple had to pay in order to get the record industry to begin to see sense.
Any, while we're at it, Apple stores suit me just fine. I've bought Apple equipment in three continents and gotten over the counter warranty replacement twice at genius bars on the other side of the planet. Show me another consumer electronics company that can do that.
(Not Sony Ericsson, they won't touch my Euro cellphone in the US)
Perhaps the point here is to piss off some overzealous security guards into escorting them out of Apple stores and turn it into an "Apple hates civil rights as well as fair use rights" scandal ?
Draco Dormiens numquam titilllandus
How do you find the key frames in the data?
By reading through the file identifying them, which entails file IO which we've established is slower than decode.
And, for reasons pointed out elsewhere, don't try this with MPEG4, H.264 or DiVX. Which pretty much leaves MPEG1 and 2.
The point is that you'd do so much work parsing the stream in order to chunk it that you might as well just decode it.
Particularly if you can spend the time while the file IO is blocked decoding frames.
That's the model that the PHBs want, give it to them. Export as ascii so you can actually build the damn thing, and make sure that every single automated formatting feature is turned off: smart quotes, spell checking, etc...Word tracks changes creditably, of course as the chorus correctly says, version control is where this ought to be done, not in the file itself. You don't need the same degree of change tracking in a source file that you do in a legal document, but heck I'm a programmer not a lawyer, or a PHB.
Also, just to stick it to the man, make sure that you order MS Office in the most expensive way possible...when they ask why it costs so much, tell them it's the "idiot tax".
he is pretty high on the list of people that would probably know about it
Especially if contact had been established before the mission. Presumably if that were the case, he would need to be forewarned about prior claims to the moon's cheese.
My personal opinion is that in all likelihood, earth is by no means unusual, life is probably rampant throughout the universe, but that's not to say anyone has, or will, visit us on earth: I agree, he's gone off the rails.
As other commenters have said, decoding video is not, per se, a trivially parallelized algorithm. Especially for modern codecs with lots of temporal encoding. MJPEG would be easily parallelized, buy you'd have to be dealing with fairly ancient sources...MediaComposer 1 for instance.
However, there are different classes of "video app" that are good targets for parallelization. Real world video editing for instance: consider multiple streams of video with overlays, rotations, effects etc. Video and audio decoding can happen in parallel, you can pipeline the effects stages so that each effect is handed off to another core. Modern video editing systems do this with aplomb.
I'm from the commercial end of this so, I can't comment much on open source alternatives. But I will say that a lot of the algorithms in certain products are highly tuned to the particular CPU type.
And they're smart enough to distribute work across only as many cores as actually exist.
Finally. Don't forget that optimization is hard. You have to consider the speed of the hard drive, the cost of sharing data between threads and cpu caches and a bunch of other real constraints. Any half decent cpu of the last five years or so can easily decode most video faster than it can be read and written to disk. So long as this is true, you won't get any benefit from parallelization.
What, won't they have Geiger Muller tubes in the future? Won't the reading be warning enough ?
Rubbish. A long time ago I lived in RIchmond (West London). Richmond Bridge crosses the Thames and has a sharp turn on the other bank. Every year there were a predictable number of accidents due to motorists speeding over the bridge and hitting pedestrians. Including a number of deaths. They installed one of the first Gatso speed cameras on the bridge, and guess what ? No more accidents.
Even I as someone who will exceed the speed limit on occasion applaud this instance. I despise cameras positioned solely to raise revenue, which today seems to be the norm. But this situation made complete sense.
A good question. I don't have to reboot that often, but sometimes that's what cures my problems.
However, I can't honestly say that means that the problem is with the router. It could be with devices upstream or downstream - rebooting the router may be the trick that resets the state machine for the connected device.
I never have to reboot my TiVo, my cell phone, my PSP, my STBs etc...
Yes, but they're not white and middle class.