Obviously, they're getting files made available for download, and indexing them.
They should also include SHA-x (SHA-512 or somesuch) hashes of the files.
This would make it very simple to prove that something really isn't infringing. No judge in their right mind would believe that yes, Joe User spent their time calculating a 1 in (2^512) hash collision, assuming such a possiblity even exists in that function.
The original NT kernel could run on MIPS, Alpha, and x86.
Problem was, none of the applications could. So you'd have a kernel and maybe some old NT services, and not much else.
In NT, the HAL is a replaceable module which can be abstracted out and replaced for different kinds of hardware. In fact -- it already does this! NT kernels today (2000, XP) have completely different HALs for PIC vs APIC, and Single vs Multiprocessor in both of those configurations. They're different enough that they have to have different ways of being addressed.
It would merely require some engineering to write an NT 5/NT5.1 HAL to run on a different operating system. However, all the other applications which are compiled to x86 bytecode, would no longer function, requiring either an emulator (slow) or a recompile (long and expensive, relatively.)
Note that a recompile *is* about all it would take -- since it'd still be Windows, the Windows API would still be exposed -- just the words used to represent it internally would be different.
Re:We had to deal with this...
on
Less Might Be More
·
· Score: 2, Interesting
Quantex had the best customer support of any company, ever.
However, that did drive them into the ground.
At their time of bankruptcy, they had $1.7 million dollars in outstanding cross-shipped hardware replacements or outstanding purchase orders, that they would never, ever call to collect on.
I kept a replacement monitor for 7 months on a 30-day return before they called to ask about it, and told me I had another 90 days to actually get it to them. (I never did, they went out of business first.)
I'm thinking that XFS probably accomplishes this transparently...
fwrite() or whatever the system call is starts a new transaction in the FS driver. If any time during the transaction, the data is corrupt, the stream dies, anything that would corrupt the data, I imagine the transaction is discarded, and the system call would return a failed result.
on Windows apps, a failed write-to-disk generates an error message, but generally it's recoverable (i.e. press "Save" again and it will work, if the error was a random one. if it reoccurs, there's something seriously wrong.)
I think the most I've kept a Windows 2003 Server box up for (running on comodity hardware...my dually box died and I needed it fixed RIGHT NOW, so I went to Circuit City and bought a $299 eMachine) is 108 days.
It "failed" at 108 because of environmental problems: the power went out for longer than the UPS could hold it up. It was going strong, however.
(For the record, I'm very surprised at how well it's doing...I stopped using the box and gave it to someone else as a "turn-key" solution preconfigured for their environment. Any time something goes wrong (power failure, etc) the UPS holds it up long enough for the power to come back on, and the "repair" procedure is: "if it screws up so bad your Internet stops working (never happened yet), hold down the power button, wait 5 minutes, turn it back on, wait 10 more minutes, try to use the Internet again. I've tested that artificially but it works perfectly in production.
I believe their services should be *more* open than they are, but *not* interoperable.
Why should people running Microsoft's service have some right to connect to a service owned by another company?
There's no reason except that tech people are a bunch of hippies.
I *do* believe people should be free to use other client software...and with the right client software, you'll have the illusion of transparency. Trillian gives the illusion of transparency through the ability to connect to multiple services at once. Thus, the user gets the illusion of it working the way they probably don't really even care about, and the content providers are permitted to keep their exclusivity.
It's obvious to me that your application won't run if it doesn't have its dependancies, regardless of where system calls, opcodes, and instruction sets are translated: if Word doesn't have the comdlg32.dll library, Common Visual Controls, it won't run.
That'd be like trying to a Linux app with dynamic bindings to glibc....without glibc being on the system.
Unfortunately due to the nature of precompiled binaries, yes, it would be x86 only.
There is, however, no reason the HAL ABI couldn't be the same regardless of the platforms, so a driver would just need to be recompiled with a different target platform, unchanged.
I think the killer app for the phone-drive might be locally stored voicemails.
I'd *love* to have several greetings recorded, and be able to select them as I see fit. "In Class", "At Work", "Busy", "Generic", "Forward to Carrier Voicemail" sorta thing.
While the phone was in-range of a tower or whatever, it handles its own voicemail, recording, etc. When it leaves range or turns off, you get your provider's voicemail.
I bet on the PC, it's a driver hook, i.e. the driver reads open file handles and if it finds a multimedia type going out over the wifi, flags it.
Similar to how ATI's video drivers will capture any video streams being played and output them full-screen on a 2nd display automatically, with Theatre mode enabled.
Multimedia appliances will probably always flag their data.
Instead of, say, limiting the length of time it can be stored, why don't they make it so that (1) once play has started, it must be completed within 48 hours, and (2) once it's finished playing, the file erases itself.
Let the TiVo store unplayed content for an infinite length of time -- but put strict limits on it once it starts to be *used(
This reminds me of when the State of Florida tried to impose a $0.01c/LB tax on sugar, to be paid by the sugar producers, with the revenue (around $433M) to be placed into a fund for environmental restoration, to fix damage caused *by* the sugar producers.
They successfully confused the issue by making it seem like Joe Taxpayer would be responsible for his share of the $433M and the proposition was voted down.
Hell, thinking it an accident is perfectly reasonable.
I heard it on the radio minutes after it happened...my initial reaction was "Oh, shit, that's terrible for the families and such." But "terrorist!" was nowhere near my thinking -- I'd heard of small planes colliding with buildings in the past on accident; thinking that a larger aircraft incapable of the same freak accident was irrational.
Then the second plane hit and I changed my tune a bit. But Bush isn't to be blamed for slow reaction on this one.
Wow, I just defended Bush's actions.
I'm voting for John Kerry, but Bush isn't half as bad as people make him out to be. (However, that half is plenty bad enough, in my opinion.)
Speaking as someone who hasn't watched *any* Trek since TNG was still being placed in new circulation, my take on Nemoy is that he attends Trek events because if he didn't, that would be the death of his public image.
Without that image, he'd probably never get any acting or promotional stints again.
Slashdot Mirror
Can you provide a MIPS, MFLOPS, or Triangles/sec rating to back that up?
The comparison table at the end is the funniest thing I've seen all week. Without a doubt.
Obviously, they're getting files made available for download, and indexing them.
They should also include SHA-x (SHA-512 or somesuch) hashes of the files.
This would make it very simple to prove that something really isn't infringing. No judge in their right mind would believe that yes, Joe User spent their time calculating a 1 in (2^512) hash collision, assuming such a possiblity even exists in that function.
Some content-type munging on the part of a webserver could make this *very* interesting.
.mpg or .avi files to be processed as a web page, with PHP or whatever.
i.e. set up
Send MPEG headers of something known to be non-infringing...with the filesize being some random 256-bit integer. (Overflow potential #1)
Then, simply generate random bytes and throw em down the script.
while(1)
(generate random bytes here)
(write to output stream)
and send them an infinite-length file.
I'm willing to bet it'd kill their crawlers.
What's the line between a product announcement for a technology like this, and a product announcement for, say, AMD Opteron processors?
They're both product announcements.
Except we like AMD here, and like Linux here, and if it's not one of those two, then we don't like it. Apparently.
erm. s/"operating system"/"hardware platform" in paragraph 4.
should've proofed.
The original NT kernel could run on MIPS, Alpha, and x86.
Problem was, none of the applications could. So you'd have a kernel and maybe some old NT services, and not much else.
In NT, the HAL is a replaceable module which can be abstracted out and replaced for different kinds of hardware. In fact -- it already does this! NT kernels today (2000, XP) have completely different HALs for PIC vs APIC, and Single vs Multiprocessor in both of those configurations. They're different enough that they have to have different ways of being addressed.
It would merely require some engineering to write an NT 5/NT5.1 HAL to run on a different operating system. However, all the other applications which are compiled to x86 bytecode, would no longer function, requiring either an emulator (slow) or a recompile (long and expensive, relatively.)
Note that a recompile *is* about all it would take -- since it'd still be Windows, the Windows API would still be exposed -- just the words used to represent it internally would be different.
Quantex had the best customer support of any company, ever.
However, that did drive them into the ground.
At their time of bankruptcy, they had $1.7 million dollars in outstanding cross-shipped hardware replacements or outstanding purchase orders, that they would never, ever call to collect on.
I kept a replacement monitor for 7 months on a 30-day return before they called to ask about it, and told me I had another 90 days to actually get it to them. (I never did, they went out of business first.)
new String("India").endsWith("istan") == FALSE;
Sorry.
Good for them. What's the big deal?
Google, as a corporate entity, and now a public one, has the responsiblity of maximizing profits.
If this means arbitrarily altering their service in certain markets, so be it.
They've done nothing wrong, morally or legally.
You can probably get any remote-trigger flash unit, and wire up a relay from the serial port.
i.e. write a stream of 1s for however long to the serial port so volt is high, trip the relay, trigger the flash, end.
Is this HDMI?
I have a Wega television with an HDML receptical. It'd be very cool to get an all-digital connection with that.
Type %20 instead of space.
It's parsed to space pretty much everywhere, but can be typed out for browsers that don't support spaces.
I'm thinking that XFS probably accomplishes this transparently...
fwrite() or whatever the system call is starts a new transaction in the FS driver. If any time during the transaction, the data is corrupt, the stream dies, anything that would corrupt the data, I imagine the transaction is discarded, and the system call would return a failed result.
on Windows apps, a failed write-to-disk generates an error message, but generally it's recoverable (i.e. press "Save" again and it will work, if the error was a random one. if it reoccurs, there's something seriously wrong.)
I think the most I've kept a Windows 2003 Server box up for (running on comodity hardware...my dually box died and I needed it fixed RIGHT NOW, so I went to Circuit City and bought a $299 eMachine) is 108 days.
It "failed" at 108 because of environmental problems: the power went out for longer than the UPS could hold it up. It was going strong, however.
(For the record, I'm very surprised at how well it's doing...I stopped using the box and gave it to someone else as a "turn-key" solution preconfigured for their environment. Any time something goes wrong (power failure, etc) the UPS holds it up long enough for the power to come back on, and the "repair" procedure is: "if it screws up so bad your Internet stops working (never happened yet), hold down the power button, wait 5 minutes, turn it back on, wait 10 more minutes, try to use the Internet again. I've tested that artificially but it works perfectly in production.
I believe their services should be *more* open than they are, but *not* interoperable.
Why should people running Microsoft's service have some right to connect to a service owned by another company?
There's no reason except that tech people are a bunch of hippies.
I *do* believe people should be free to use other client software...and with the right client software, you'll have the illusion of transparency. Trillian gives the illusion of transparency through the ability to connect to multiple services at once. Thus, the user gets the illusion of it working the way they probably don't really even care about, and the content providers are permitted to keep their exclusivity.
This operates at a lower level than that.
It's obvious to me that your application won't run if it doesn't have its dependancies, regardless of where system calls, opcodes, and instruction sets are translated: if Word doesn't have the comdlg32.dll library, Common Visual Controls, it won't run.
That'd be like trying to a Linux app with dynamic bindings to glibc....without glibc being on the system.
Unfortunately due to the nature of precompiled binaries, yes, it would be x86 only.
There is, however, no reason the HAL ABI couldn't be the same regardless of the platforms, so a driver would just need to be recompiled with a different target platform, unchanged.
I think the killer app for the phone-drive might be locally stored voicemails.
I'd *love* to have several greetings recorded, and be able to select them as I see fit. "In Class", "At Work", "Busy", "Generic", "Forward to Carrier Voicemail" sorta thing.
While the phone was in-range of a tower or whatever, it handles its own voicemail, recording, etc. When it leaves range or turns off, you get your provider's voicemail.
I bet on the PC, it's a driver hook, i.e. the driver reads open file handles and if it finds a multimedia type going out over the wifi, flags it.
Similar to how ATI's video drivers will capture any video streams being played and output them full-screen on a 2nd display automatically, with Theatre mode enabled.
Multimedia appliances will probably always flag their data.
Then you'd have actually purchased the movie, rather than paid to view it once or for x amount of time.
That'd probably cost more.
Instead of, say, limiting the length of time it can be stored, why don't they make it so that (1) once play has started, it must be completed within 48 hours, and (2) once it's finished playing, the file erases itself.
Let the TiVo store unplayed content for an infinite length of time -- but put strict limits on it once it starts to be *used(
This reminds me of when the State of Florida tried to impose a $0.01c/LB tax on sugar, to be paid by the sugar producers, with the revenue (around $433M) to be placed into a fund for environmental restoration, to fix damage caused *by* the sugar producers.
They successfully confused the issue by making it seem like Joe Taxpayer would be responsible for his share of the $433M and the proposition was voted down.
*sigh*
Hell, thinking it an accident is perfectly reasonable.
I heard it on the radio minutes after it happened...my initial reaction was "Oh, shit, that's terrible for the families and such." But "terrorist!" was nowhere near my thinking -- I'd heard of small planes colliding with buildings in the past on accident; thinking that a larger aircraft incapable of the same freak accident was irrational.
Then the second plane hit and I changed my tune a bit. But Bush isn't to be blamed for slow reaction on this one.
Wow, I just defended Bush's actions.
I'm voting for John Kerry, but Bush isn't half as bad as people make him out to be. (However, that half is plenty bad enough, in my opinion.)
Speaking as someone who hasn't watched *any* Trek since TNG was still being placed in new circulation, my take on Nemoy is that he attends Trek events because if he didn't, that would be the death of his public image.
Without that image, he'd probably never get any acting or promotional stints again.