The reasoning is precisely because we have cheap and fast CPUs we can now sacrifice some performance (it doesn't take much to deal with fast ethernet) in order to get more stability. In the case of network and sound, driver faults, for instance, won't (shouldn't anyways) take down the system anymore.
It's all theory of course, but I can see the reasoning at least.
It won't even really work in the manner you've described it unless you propose injecting millions upon millions of such a "vaccine" (assuming it even does the whole attach and deny attachment thing).
From the moment I press the power button on my Vista laptop, to the time when the desktop finally appears, including _tapping_ out my password (since it's a TabletPC) the total elapsed time is about 45 seconds.
If breeder reactors were used, then the wastes only stay radioactive for a couple hundred years at worst. Plus you also get 10x the energy out as well as less than 10% of the waste.
That's not memory, that's battery aging.
Li-Ion batteries, instead of having memory, simply age and lost capacity over time. If your battery is warm and at anything significantly above or below ~60%, then it loses capacity at a much higher rate.
Wait, all of your programs are limited by DISK I/O?
Is your program really ripping through a gigabyte of data per second?
You're chunking your data into segments that fit into memory and loading them all at once enmasse right? Perhaps you could have two threads, one to load data while the other does processing in memory.
There's almost always a way to reduce the disk I/O in these days with so much RAM and cache (on the CPUs and disk themselves).
You missed the fully compatible with USB 2.0 part. USB 3.0 still has the copper component since, as you've mentioned, it's still useful and important. So not only can you continue to use existing USB parts, new parts can still just use the copper part.
And it indeed is 5.0 Gigabits per second, not bytes.
That wasn't the real issue with USB 2.0 anyways. The real issue is that USB 2.0 was suited for burst data transfers (where you got the theoretical speed), but simply can't do streaming at the maximum speed due to the enormous overhead of the protocol.
With that said, CPU utilization will indeed be a HUGE concern since USB 3.0 is so fast. The relatively minor CPU overhead of USB 2.0 will give way to CPU stalling overhead unless USB 3.0 addresses it. There's not enough information to make a statement about this, so we'll have to wait for more information to be released.
Ever heard of green algae? Those nasty little critters started releasing this toxic waste called Oxygen into the atmosphere poisoning practically the entire biosystem. The effects of their actions persist even to today.
Even with that said, the firewall in Vista has been updated to also allow outbound blocking in the same manner of Linux. That is, it doesn't by default.
Slashdot Mirror
Yeah, they whine and angst instead =P
And yet people go on about longer boot times for Vista.
(I agree with you btw)
Interestingly enough VS2008 starts up in 1 second while Eclipse takes 6 seconds on my machine.
Both are too low for it to be a factor in which I prefer but Eclipse definitely takes longer to load.
The reasoning is precisely because we have cheap and fast CPUs we can now sacrifice some performance (it doesn't take much to deal with fast ethernet) in order to get more stability. In the case of network and sound, driver faults, for instance, won't (shouldn't anyways) take down the system anymore. It's all theory of course, but I can see the reasoning at least.
It won't even really work in the manner you've described it unless you propose injecting millions upon millions of such a "vaccine" (assuming it even does the whole attach and deny attachment thing).
From the moment I press the power button on my Vista laptop, to the time when the desktop finally appears, including _tapping_ out my password (since it's a TabletPC) the total elapsed time is about 45 seconds.
Well, SP1 is and SP3 will also be exception. Oops that's just every single SP for XP.
That's not how vaccines work...
Actually, it depends on the type of infinity you're talking about (yes there are different types of infinities).
For instance, Let [A] be the set of all integers and [B] be the set of all integers except 1.
Both are infinite sets.
Subtracting [B] from [A] leaves us with a decidedly non-infinite set that contains only 1.
If breeder reactors were used, then the wastes only stay radioactive for a couple hundred years at worst. Plus you also get 10x the energy out as well as less than 10% of the waste.
Before you even think that Solar is clean, you should do a bit of research on what it takes to make solar cells. Quite eye-opening.
That's not memory, that's battery aging. Li-Ion batteries, instead of having memory, simply age and lost capacity over time. If your battery is warm and at anything significantly above or below ~60%, then it loses capacity at a much higher rate.
Wait, all of your programs are limited by DISK I/O?
Is your program really ripping through a gigabyte of data per second?
You're chunking your data into segments that fit into memory and loading them all at once enmasse right? Perhaps you could have two threads, one to load data while the other does processing in memory.
There's almost always a way to reduce the disk I/O in these days with so much RAM and cache (on the CPUs and disk themselves).
You missed the fully compatible with USB 2.0 part. USB 3.0 still has the copper component since, as you've mentioned, it's still useful and important. So not only can you continue to use existing USB parts, new parts can still just use the copper part. And it indeed is 5.0 Gigabits per second, not bytes.
That wasn't the real issue with USB 2.0 anyways. The real issue is that USB 2.0 was suited for burst data transfers (where you got the theoretical speed), but simply can't do streaming at the maximum speed due to the enormous overhead of the protocol.
With that said, CPU utilization will indeed be a HUGE concern since USB 3.0 is so fast. The relatively minor CPU overhead of USB 2.0 will give way to CPU stalling overhead unless USB 3.0 addresses it. There's not enough information to make a statement about this, so we'll have to wait for more information to be released.
Actually, USB 3.0 will have optical connects (while retaining the metal connectors for USB 2.0 compatibility (the socket, not the cable).
USB 3.0 is slated for 2008. Optical connect, fully USB 2.0 compatible, 5GBPS. Barring a major mistake, Firewire is, again, DOA.
Probably depends on the locale. The flashing red light means exactly what you deduced where I live.
Ever heard of green algae? Those nasty little critters started releasing this toxic waste called Oxygen into the atmosphere poisoning practically the entire biosystem. The effects of their actions persist even to today.
With scissors I'd presume. Slow moving sharp edges should cut through this easily.
Hopefully bullets and shrapnel, being fast moving, won't.
Why would you need to update Firefox if it's only going to be used for browsing known-safe internal sites?
And there's always...
Password:
(sudo prompt for password)
I believe you mean no more cooperative multitasking. The modern desktop OS's are all preemptive IIRC.
Depends. If you're an administrator, it'll give a continue prompt. If you're a normal user, it'll pop up the login prompt.
Even with that said, the firewall in Vista has been updated to also allow outbound blocking in the same manner of Linux. That is, it doesn't by default.