IIRC from physics classes, is the force making it hard to walk on a moving merry-go-round not the centripetal force?? I thought Coriolis was only a pseudo-force, not a real one.
Ehh... you're 3/4 right. Centripetal force is real and coreolis force is "imaginary". Centripital force is force towards the center of rotation, keeping you from traveling in a streight line. Centripital force doesn't make it hard to walk on the merry-go-round; centripital force allows you to stay on the merry-go-round. You're thinking of the "imaginary" centrifugal force that appears to counter-act the centripital force you are applying with your feet.
Centrifugal force and Coreolis force are both imaginary forces used as short hand for taking second time derrivatives (calclating accelerations) in rotating reference frames using polar coordinates . If you're spinning at a constant speed about the merry-go-round, you keep the same polar cooarinates when in fact, a lot of corce is acting on your body to keep it constantly changing direction at a fairly high rate. In the reference frame you ae always at rest, so you don't say that momentum change is balancing out the force you are using to keep yourself "still" in the rotating reference frame, you say that this imaginary "centrifugal" force is acting on you. The two statements are equivalent, but one is a technical gloss.
Now suppose you try moving in relation to the rotating reference frame. You want to travel in a streight line in the polar coordinates. Well, since the frame of reference is rotating, a streight path in non-rotating space is a curved line in the rotating reference frame, and the amount of aparent curvature is dependent on speed of travel relative to the rotating reference frame. So when you try and walk in a streight line on the merry-go-round with out correcting for rotation, you more or less walk in a streight line in the non-rotating reference frame. In the rotating reference frame, your path is curved. The easiest way to do calculations is to make up frorces that would havepushed your path into that curved shape. It's all just short hand so that everything doesn't need to be translated to and from the stationary reference frame.
Even at the equator, you experience the coreolis effect, it's just that your axis of rotation is parallel to the ground. At the equtor, running East appears to make you lighter, running West appears to make you heavier, jumping up appears to push you West, and dropping off a ledge appears to push you East. One explination of why thy always launch spacecraft in an eastwardly-traveling orbit is that that way the coreolis force helps, rather than hinders the spaceflight. In a non-rotating reference frame this is equivalent to saying that it already has a lot of speed in an easterly direction, so blasting off to the west actually means sloing down a lot rather than using the speed it already has due to traveling at the same speed as the ground.
It's all equivalent, sometimes it's jsut easier to do the math one way. If nobody has done the math to figure out how the imaginary forces get added in in your situation, then you need to translate everything into a non-rotaing, non-accelerating frame of reference and do the calculatins and translate them back into your rotating frame of reference.
It's kinda like special relativity. If you forget the formulas, you can re-derrive them by looking at everyhting in a stationary reference frame and looking at a photon clock and a photon yardstick and figuring out what apears to happen to one secodn and what appearsto happen to one meter and what appears to happen to one kg being acted upon by 1 Newton. It's just a lot easier if you remember the formulas Einstein derrived for you instead of having to transate everything to and from the stationary reference frame.
Slammer was UDP, so people got full traffic even if they only had port 80 open. Unless customers have the option of port-based filtering on the upstream side of the connection and/or putting a cap on total bandwidth usage for the account, it's hard to make the claim that it's a risk the customer should have dealt with. Fluctuations in thousands of percents over the previous month's bill is really painful. It seems irresponsible to open up customers to such risks without giving them any ability whatsoever to mitigate the risks. ISPs also have a responsibility to the community not to be lazy and "piss in the communal pool" by standing by and not offering (via phone or email) to filter out traffic (bi-directionally at the customner's discretion) from these internet-wide security macro-events.
Ideally you'd be able to roll over bandwidth for exactly one month as in subtracting the previous month's rollover at the end of the month. Your bandwith would be continously throttled to the rate at which you'd expend all of your bandwdth at the end of the month. Without rollover, the ISPs would have a huge sawtooth pattern in monthly load and one of the sides of the teeth being nearly vertical. The rollover is more for the benefit ofthe ISPs than anything, so is upstream port blocking, allowing ISPs to blockunwanted traffic at its boarders.
Open relays are necessary for spoofing the sending domain and also act as buffers connected to fat pipes. Several email trojans come with their own SMTP server. They're very easy to write. If having an SMTP server was the bottleneck, they would have put SMTP servers in all of the SPAM software long ago rather than wasting the effort of finding new open relays.
An open relay allows a SPAMer to lie about his/her domain and ofload a batch of emails lighting fast. The SMTP server does the storing and the forwarding with faked headers.
... about 4 years ago to get a free 6 month subscription to a magazine. It's mailing address is my old Fraternity house. The fake company got sent a letter by the BSA last month. The funny thing is, I think it was a Linux magazine. You'd think they'd check for incorporation before sending letters. The fake company gets corporate credit card offers all the time, but I used to think the BSA was more careful than that.
Being and Eagle Scout, I should have sent them a cease and dessist letter asking them to stop infringing on the Boy Scouts of America's trademark, using the business-reply-envelope of course. I thought about sending them Mardi Gras beads from the party and a condom with a note saying "It's a fake company. The address is my old Fraternity house. Good thing I'm in the area and stop in for parties. There's a good Mardi Gras party tonight. Next time tell the girls at the office." I've heard that companies have to pay the extra postage if you make the business reply mail too heavy. Plus, it'd be great in a few years to see something I did getting put on one of those "most outrageous yet genuine recieved mail" joke emails that go around. If I get another letter maybe I'll send them some sand from the next beach party.
It's been a while since I've studied anything electrocal. However, your power lines have non-negligible inductance. Lower frequence means less voltage drop across the inductor. You also get some hysteresis losses (non-ideal inductors) which will be reduced by lowering the frequency. The first factor doesn't actually effect energy efficiency, just your ability to consume electriciy (voltage just doesn't dip as much when you can't keep up). The second factor directly reduces the percentage fo power lost during distribution. Back in HS, they told us that power was brought into the minneapolis area from power plants in the Dakotas via DC transmission lines to minimize losses. DC is better for distribution, but it's much cheaper to build an efficient and powerful AC motor vs. and equivalent DC motor. The same goes for generators.
So where'd you get that sig? I've used it. I may have been the first person to think it up, but I don't think it requires that much creativity. I'm sure someone said it before me, and I'd like to know who.
Not thieves. Not murderous pirates on the high seas. Copyright infringers. There are big differences. Speeding isn't breaking and entering or manslaughter. Copyright infringement isn't thievery or piracy.
What happens when you have to explain to Dad who just bought his kid a $2000 portable that he can't use all that software?
Huh? You mean Office and Excell? Many *nix applications can import/export documents form/to their win32 counterparts. Not to mention cygwin for unning *nix apps on win32. Cygwin XFree86 or that X server from Hummingbird will allow the kid to run *nix apps on the server with local display. I do this all the time and so do plenty of my friends. Granted, we're at MIT, but I'm sure it happens with plenty of average students at many schools.
TCO would probably kill you. This is hearsay, but I've heard that if you try and site-license MS Windows, they charge you as if you had Windows on every x86 box, regarless of the actual OS mix. If you're running *NIX on non-x86, it might not be so bad 'cause you wouldn't be paying for unused licenses.
Then encrypt the packages and require authentication to get the keys. If this can be spoofed so can their access-control system. Unless they're trying to keep it a secret which software needs updating... which would be futile. Secret updates tend to cause migrations to other platforms, or at least cause really mad usenet postings and very bad press when discovered. The keys can be as fine or coarse grained as you like, in both time and coverage, even overlapping in both. "The 3.5.2003 DESX key for the word.dll update is.... The 1Q2003 DESX key for all Office2003-realted updates is..." You can then use hybrid encryption with both steps using symetric key algorithms so that many keys can be used for the same package. BTW, MS still has a DESX fetish, right?
If a flat list doesn't scale to the size you need, then use a tree. Group packages into trees. At each level have a donloadable list of branches and the date of the last update on that branch. You can have the client reach only the relavent leaf nodes by pruning away the branches that do not contain any software installed on the client system or that have been visited since the date of the last update. Priority flags for a depth-first search would also be good if you have lots of large updates that need to get done. It's simple and obvious enough that I'm sure you thought of it shortly after you posted.
It would seem that they have grouped things improperly if they have to track 100,000 packages. They seem to have made their packaging too fine grained. The Debian packages span more functionality than the products supported by Windows Update, so I don't buy the bredth argument. If they're concerened about update size, they can make the new packages diffs from the previous packages, with preferrred alternative dependencies for cumulative packages in order to end the back-chaining of downloads.
156 bits? Umm... DESX is 64+64 bits, but 8 of those are parity bits, so it's really 120 bit encryption. Or did they come up with some new goofy encrption algorithm that doesn't even use an integral number of bytes as its key?
Anyway, my point is that as implemented by default, the EFS is really only about 35-bit encryption.
The NT hash is also unsalted, greatly reducing the effort and space requirements for hosting a "master list" of passwords. The US government and certain companies are really stupid if they're not working on such a "master list". MD4 is pretty fast in software on 32-bit CPUs, so a nice big array of MIPS (either 32 or 64 bit versions) or Xscale CPUs would probably be much cheaper than Deep Crack. I think for under $50,000 and two years you could generate a lookup table for 99.9 % of the passwords out there.
Unencrypted metadata can also be problematic in some instances. With loop-AES you can tell them you overwrote the parition with ranodm data 256 times, and they have know way of knowing if you're lying or you ran a script to destroy the data when they kicked down the door. Also, if I wanted, I could wite a perl script to poll/proc for the status of the "Internet Shopping" button on my keyboard (did this for my desktop picture) that would unmount/dev/hda7 and securely overwrite hda7_key.gpg at the touch of a button. You can't securely overwrite all of your encrypted files that quickly, but destroying my 256-bit AES keyfile is effectively as strong while being literally about a billion times faster.
You're much better off using PGP-disk, or using a GPG-encrypted keyfile with loop-AES. Besides, both my root password and my GPG password are too long to be typed into the WinXP login screen. (Up to 128 characters, but it starts beeping at you after 14.)
all i can confirm that, krazy8 has not been seen in a long, long time... and that he was busted FOR selling modchips.
the details of his plea bargain are not out, and nobody is speaking about anything.
Oh No! Camp X-ray! I knew it! Poor poor krazy8.
A Taliban general once used heroin money to buy a pack of cigarettes from a guy whose grand-nephwey once shined the shoes of an executive of the company that manufactured the chips. Now he's in prison for having Al Qaeda links. SSL online constitute the use of encryption in conjuction with a crime. Poor poor stupid krazy8.
The obvious and more scalable way to do this is the way Debian, and several other *NIX vendors do it... Have a list of available updates that the client downloads from the vendor. This is static content and easily served. Then you let the client figure out what's relavent and download the individual static content update files it needs. The server load is much lower (and therefore also the latency if everything else is equal) compared to the way MS is doing it. If your copy protection gets broken by the Debian scheme, you're screwed anyway. (Nobody could ever reverse-engineer a client to lie about what software is already installed... right?)
Assuming the MS engineers are smart, the logical explination is that they actively want to collect your information.
I'm not aware of any plans to make Palladium check signatures of binaries in memory. That would be a huge drain on computational resources. I'm not aware of anyone seriously suggesting this at any time in the future. Signatures are checked by the PE loader as the application is started. Don't tell MS, but I stronly suspect implementation problems in the first few versions. (There are still about 40 of the 240 Win2K system calls that don't properly check inputs and will result in BSOD with proper inputs from any user. This is due to be fixed in SP4.)
10 Free Checking accounts with ATM cards... $0
1000 FPGAa at 200 MHz... $5,000
Beer and Pizza for 100 all-nighters... $1,000
Everyone's PINs... priceless
Some things money can't buy, for everything else there's CitiBank.
Damn, mod parent (+1, big brass ones). Shorting traces on your motherboard with a screwdriver that you *think* will reset the CMOS? I would have been too chicken and would have just popped out the EEPROM and re-burned the BIOS.
Do you ccasionally load one chamber of a 44 magnum revolver, give her a spin, point at your computer case and pull the trigger? Do you occasionally boot MS OSes just for the danger factor? I mean, I know all too well the sound of a HD head sticking and I keep my 3rd HD connected even though I've twice had it jam the IDE controller (kernel panic due to losing sight of drives, BIOS couldn't find any drives until I unplugged offending drive). However, there are some things I don't play with. Shorting undocumented motherboard traces is one of those things.
Linux isn't modular? Does the sun not shine in your world? Does the air always smell like farts? Get your head out of your ass!
Almost all of the drivers can be compiled as modules (which means they can be left of the FS or not compiled in the first place). I hear they're working on making ALL of the drivers modules and placing the modules necessary for booting in an initial compressed ramdisk. I think QNX and VxWorks are smaller. Do Symbian and PalmOS support premptive multitasking in protected memory spaces?
In any case, Linux scales very well down into a PDA environment due to its modularity.
They realy should have the backup version be non-flashable, preferably in a ROM. You'd want the backup BIOS to actually call the upgradable BIOS by defult. I've seen a machine hang during a BIOS update. It isn't pretty. There are also some viruses that attempt to overwrite the BIOS. It would be great to have an absolutely unbreakable BIOS. In the backup mode, you might have some bugs in obsure features and might lack the most modern features, but you most likely retain something that can boot at least a recovery floppy. More importantly, it'd be really hard to lose the ability to flash the flashable version of the BIOS.
or is Forth just particularly well-suited to that time of application?
I have no special knowledge of the FreeBSD bootloader. However, Forth was actually originally designed as part of a "system-on-a-chip" CAD system. Forth bytecode is extrememly compact, the VM is very light, and the language itself greatly encourages code reuse. (I forget the statistics, but Forth programs on average are very much characterized by many short functions that get used over and over.) Forth doesn't ncessarily lend itself to fast native binaries on modern superscalar CPUs, but the BIOS needs to wait for the HD to spin up anyway. A 200% faster language for the BIOS won't cut your boot time by 66%. In summary, Forth is simple and compact and designed for embedded applications. This is why it finds its way into bootloaders/firmware.
... as soon as CC companies foot the bill for fraud. Smart card technology is very slowly being adopted, but it would be adopted within 6 months if the powerful CC companies rather than the pwerless merchants pad for the fraud, or at least split the cost of fraud with the merchants.
This would be a bad business move for Apple. Apple wants to support one platform. No need to confuse customers and no need to dilute thier developer base. Not all of thier developers would compile for both platforms, further reducing software support for OS/PPC. In short, they don't want to compete with themselves.
Slashdot Mirror
Ehh... you're 3/4 right. Centripetal force is real and coreolis force is "imaginary". Centripital force is force towards the center of rotation, keeping you from traveling in a streight line. Centripital force doesn't make it hard to walk on the merry-go-round; centripital force allows you to stay on the merry-go-round. You're thinking of the "imaginary" centrifugal force that appears to counter-act the centripital force you are applying with your feet.
Centrifugal force and Coreolis force are both imaginary forces used as short hand for taking second time derrivatives (calclating accelerations) in rotating reference frames using polar coordinates . If you're spinning at a constant speed about the merry-go-round, you keep the same polar cooarinates when in fact, a lot of corce is acting on your body to keep it constantly changing direction at a fairly high rate. In the reference frame you ae always at rest, so you don't say that momentum change is balancing out the force you are using to keep yourself "still" in the rotating reference frame, you say that this imaginary "centrifugal" force is acting on you. The two statements are equivalent, but one is a technical gloss.
Now suppose you try moving in relation to the rotating reference frame. You want to travel in a streight line in the polar coordinates. Well, since the frame of reference is rotating, a streight path in non-rotating space is a curved line in the rotating reference frame, and the amount of aparent curvature is dependent on speed of travel relative to the rotating reference frame. So when you try and walk in a streight line on the merry-go-round with out correcting for rotation, you more or less walk in a streight line in the non-rotating reference frame. In the rotating reference frame, your path is curved. The easiest way to do calculations is to make up frorces that would havepushed your path into that curved shape. It's all just short hand so that everything doesn't need to be translated to and from the stationary reference frame.
Even at the equator, you experience the coreolis effect, it's just that your axis of rotation is parallel to the ground. At the equtor, running East appears to make you lighter, running West appears to make you heavier, jumping up appears to push you West, and dropping off a ledge appears to push you East. One explination of why thy always launch spacecraft in an eastwardly-traveling orbit is that that way the coreolis force helps, rather than hinders the spaceflight. In a non-rotating reference frame this is equivalent to saying that it already has a lot of speed in an easterly direction, so blasting off to the west actually means sloing down a lot rather than using the speed it already has due to traveling at the same speed as the ground.
It's all equivalent, sometimes it's jsut easier to do the math one way. If nobody has done the math to figure out how the imaginary forces get added in in your situation, then you need to translate everything into a non-rotaing, non-accelerating frame of reference and do the calculatins and translate them back into your rotating frame of reference.
It's kinda like special relativity. If you forget the formulas, you can re-derrive them by looking at everyhting in a stationary reference frame and looking at a photon clock and a photon yardstick and figuring out what apears to happen to one secodn and what appearsto happen to one meter and what appears to happen to one kg being acted upon by 1 Newton. It's just a lot easier if you remember the formulas Einstein derrived for you instead of having to transate everything to and from the stationary reference frame.
Ideally you'd be able to roll over bandwidth for exactly one month as in subtracting the previous month's rollover at the end of the month. Your bandwith would be continously throttled to the rate at which you'd expend all of your bandwdth at the end of the month. Without rollover, the ISPs would have a huge sawtooth pattern in monthly load and one of the sides of the teeth being nearly vertical. The rollover is more for the benefit ofthe ISPs than anything, so is upstream port blocking, allowing ISPs to blockunwanted traffic at its boarders.
Anonymous, but a good idea for an improvement to the grandparent, at least on the surface.
An open relay allows a SPAMer to lie about his/her domain and ofload a batch of emails lighting fast. The SMTP server does the storing and the forwarding with faked headers.
Being and Eagle Scout, I should have sent them a cease and dessist letter asking them to stop infringing on the Boy Scouts of America's trademark, using the business-reply-envelope of course. I thought about sending them Mardi Gras beads from the party and a condom with a note saying "It's a fake company. The address is my old Fraternity house. Good thing I'm in the area and stop in for parties. There's a good Mardi Gras party tonight. Next time tell the girls at the office." I've heard that companies have to pay the extra postage if you make the business reply mail too heavy. Plus, it'd be great in a few years to see something I did getting put on one of those "most outrageous yet genuine recieved mail" joke emails that go around. If I get another letter maybe I'll send them some sand from the next beach party.
It's been a while since I've studied anything electrocal. However, your power lines have non-negligible inductance. Lower frequence means less voltage drop across the inductor. You also get some hysteresis losses (non-ideal inductors) which will be reduced by lowering the frequency. The first factor doesn't actually effect energy efficiency, just your ability to consume electriciy (voltage just doesn't dip as much when you can't keep up). The second factor directly reduces the percentage fo power lost during distribution. Back in HS, they told us that power was brought into the minneapolis area from power plants in the Dakotas via DC transmission lines to minimize losses. DC is better for distribution, but it's much cheaper to build an efficient and powerful AC motor vs. and equivalent DC motor. The same goes for generators.
So where'd you get that sig? I've used it. I may have been the first person to think it up, but I don't think it requires that much creativity. I'm sure someone said it before me, and I'd like to know who.
Not just Alpha. Windows NT 3.51 ran on x86, Alpha, MIPS, and PPC. (And the i860, which never made it past being a CPU prototype.)
Not thieves. Not murderous pirates on the high seas. Copyright infringers. There are big differences. Speeding isn't breaking and entering or manslaughter. Copyright infringement isn't thievery or piracy.
TCO would probably kill you. This is hearsay, but I've heard that if you try and site-license MS Windows, they charge you as if you had Windows on every x86 box, regarless of the actual OS mix. If you're running *NIX on non-x86, it might not be so bad 'cause you wouldn't be paying for unused licenses.
Then encrypt the packages and require authentication to get the keys. If this can be spoofed so can their access-control system. Unless they're trying to keep it a secret which software needs updating... which would be futile. Secret updates tend to cause migrations to other platforms, or at least cause really mad usenet postings and very bad press when discovered. The keys can be as fine or coarse grained as you like, in both time and coverage, even overlapping in both. "The 3.5.2003 DESX key for the word.dll update is.... The 1Q2003 DESX key for all Office2003-realted updates is..." You can then use hybrid encryption with both steps using symetric key algorithms so that many keys can be used for the same package. BTW, MS still has a DESX fetish, right?
It would seem that they have grouped things improperly if they have to track 100,000 packages. They seem to have made their packaging too fine grained. The Debian packages span more functionality than the products supported by Windows Update, so I don't buy the bredth argument. If they're concerened about update size, they can make the new packages diffs from the previous packages, with preferrred alternative dependencies for cumulative packages in order to end the back-chaining of downloads.
Anyway, my point is that as implemented by default, the EFS is really only about 35-bit encryption.
The NT hash is also unsalted, greatly reducing the effort and space requirements for hosting a "master list" of passwords. The US government and certain companies are really stupid if they're not working on such a "master list". MD4 is pretty fast in software on 32-bit CPUs, so a nice big array of MIPS (either 32 or 64 bit versions) or Xscale CPUs would probably be much cheaper than Deep Crack. I think for under $50,000 and two years you could generate a lookup table for 99.9 % of the passwords out there.
Unencrypted metadata can also be problematic in some instances. With loop-AES you can tell them you overwrote the parition with ranodm data 256 times, and they have know way of knowing if you're lying or you ran a script to destroy the data when they kicked down the door. Also, if I wanted, I could wite a perl script to poll /proc for the status of the "Internet Shopping" button on my keyboard (did this for my desktop picture) that would unmount /dev/hda7 and securely overwrite hda7_key.gpg at the touch of a button. You can't securely overwrite all of your encrypted files that quickly, but destroying my 256-bit AES keyfile is effectively as strong while being literally about a billion times faster.
You're much better off using PGP-disk, or using a GPG-encrypted keyfile with loop-AES. Besides, both my root password and my GPG password are too long to be typed into the WinXP login screen. (Up to 128 characters, but it starts beeping at you after 14.)
A Taliban general once used heroin money to buy a pack of cigarettes from a guy whose grand-nephwey once shined the shoes of an executive of the company that manufactured the chips. Now he's in prison for having Al Qaeda links. SSL online constitute the use of encryption in conjuction with a crime. Poor poor stupid krazy8.
Assuming the MS engineers are smart, the logical explination is that they actively want to collect your information.
I'm not aware of any plans to make Palladium check signatures of binaries in memory. That would be a huge drain on computational resources. I'm not aware of anyone seriously suggesting this at any time in the future. Signatures are checked by the PE loader as the application is started. Don't tell MS, but I stronly suspect implementation problems in the first few versions. (There are still about 40 of the 240 Win2K system calls that don't properly check inputs and will result in BSOD with proper inputs from any user. This is due to be fixed in SP4.)
10 Free Checking accounts with ATM cards ... $0 ... $5,000 ... $1,000
1000 FPGAa at 200 MHz
Beer and Pizza for 100 all-nighters
Everyone's PINs... priceless
Some things money can't buy, for everything else there's CitiBank.
Do you ccasionally load one chamber of a 44 magnum revolver, give her a spin, point at your computer case and pull the trigger? Do you occasionally boot MS OSes just for the danger factor? I mean, I know all too well the sound of a HD head sticking and I keep my 3rd HD connected even though I've twice had it jam the IDE controller (kernel panic due to losing sight of drives, BIOS couldn't find any drives until I unplugged offending drive). However, there are some things I don't play with. Shorting undocumented motherboard traces is one of those things.
"Don't fuck with the Jesus" -- The Big Labowski (sp?)
Almost all of the drivers can be compiled as modules (which means they can be left of the FS or not compiled in the first place). I hear they're working on making ALL of the drivers modules and placing the modules necessary for booting in an initial compressed ramdisk. I think QNX and VxWorks are smaller. Do Symbian and PalmOS support premptive multitasking in protected memory spaces?
In any case, Linux scales very well down into a PDA environment due to its modularity.
They realy should have the backup version be non-flashable, preferably in a ROM. You'd want the backup BIOS to actually call the upgradable BIOS by defult. I've seen a machine hang during a BIOS update. It isn't pretty. There are also some viruses that attempt to overwrite the BIOS. It would be great to have an absolutely unbreakable BIOS. In the backup mode, you might have some bugs in obsure features and might lack the most modern features, but you most likely retain something that can boot at least a recovery floppy. More importantly, it'd be really hard to lose the ability to flash the flashable version of the BIOS.
I have no special knowledge of the FreeBSD bootloader. However, Forth was actually originally designed as part of a "system-on-a-chip" CAD system. Forth bytecode is extrememly compact, the VM is very light, and the language itself greatly encourages code reuse. (I forget the statistics, but Forth programs on average are very much characterized by many short functions that get used over and over.) Forth doesn't ncessarily lend itself to fast native binaries on modern superscalar CPUs, but the BIOS needs to wait for the HD to spin up anyway. A 200% faster language for the BIOS won't cut your boot time by 66%. In summary, Forth is simple and compact and designed for embedded applications. This is why it finds its way into bootloaders/firmware.
... as soon as CC companies foot the bill for fraud. Smart card technology is very slowly being adopted, but it would be adopted within 6 months if the powerful CC companies rather than the pwerless merchants pad for the fraud, or at least split the cost of fraud with the merchants.
This would be a bad business move for Apple. Apple wants to support one platform. No need to confuse customers and no need to dilute thier developer base. Not all of thier developers would compile for both platforms, further reducing software support for OS/PPC. In short, they don't want to compete with themselves.