IIRC, most comets are still discovered by lucky amateurs.
That used to be true, but nowadays the LINEAR project picks up most of them. Amateur astronomers still account for a few a year, though. There are a lot of comets out there to be found.
Sea water is nearly incompressible, so the viscosity wouldn't be appreciably higher. If it was, deep-marine fish like ratfish would have a tough time (especially considering how few calories there are down there for the poor things to eat). Sea water viscosity actually depends chiefly on the temperature (colder is more viscous), and to a lesser degree on salinity (more saline is more viscous - this obviously implies that sea water is slightly more viscous than fresh water).
There are still all kinds of other effects that make diving to that depth very, very difficult. One is the corrosive effect of oxygen in the water at that pressure.
Anyway, wouldn't a more viscous fluid be better for this thing's method of submergence? It doesn't need to go fast, it just needs to generate downward pressure with the reverse equivalent of an airfoil (or something). Planes get more lift in more viscous air, at least to the point where they can still move forward quickly enough.
It's likely a very small amount. These windows might not even be active-matrix, and they're monochrome. I'd guess less than a watt for even a large pane. Though I still agree with your point, we'd probably all do better to replace aging refrigerators than worry too much about the power demand from this.;)
Re:Old Tech, new (but in hindsight, obvious) use.
on
Multimedia Windowpanes
·
· Score: 1
Which does make some sense, considering to use a projection TV (or other projector) you really have to make the room dark. It has a certain elegance. The window-speakers (which aren't all that new, I think there was a/. story about them at least a year or two ago) are a more questionable idea, and likely wouldn't fly in any burgh that gives a crap about neighbourhood sound pollution.
A low-voltage electric current runs through the window. When the current is on, the window is clear. But flip a switch to turn the current off and the glass goes opaque, allowing it to be used as a projection screen for watching television or DVDs.
The article doesn't mention the underlying technology by name, but it's probably simply an LCD panel similar to that in a notebook display. Whether the relaxed liquid crystal strands uncurl with (white, transparent) or against (black, opaque) the plane of the polarisation filter is always just a manufacturing choice. In notebooks it can be a matter of saving power by minimising the amount of screen you need to change from the default, on average. A white-on-black character display should probably relax to black, while a black-on-white Mac-style windowed display might better relax to white. Of course I'm not claiming they always use that much logic in the decision (Apple maybe since they seem to put a lot of thought into powersaving, but more likely they just buy what's cheapest like everyone else).
The tendency of the windows in the article to relax to opacity could be thought of as a privacy feature, I suppose. It does seem odd when the device is being sold as a window that can be used as a screen, though (as opposed to the converse).
Re:They used to have to look in the windows
on
Multimedia Windowpanes
·
· Score: 2, Funny
Of course, they'll see a horizontally flipped representation. That could colour their impression of you too, especially if you prefer that sort of porn where it's already difficult to tell who's doing what to whom.
Except, of course, that it's not somebody else's stuff anymore. It's YOURS after you buy it. I think you would have every right to look at (and inspect) everything you buy, specially second hand.
Doubtful. Consider that you never own the copyrighted-by-others software on your drive (including the GPL stuff), only an explicit licence to use it. Data on the drive that you author or even just compile will likely still belong to you, regardless of the fact that you've sold the media. In the case of deleted or overwritten data this is even more defencible, especially with legislation like the DMCA - deletion here would be the "copy-protection" method (remember that the data must be copied to be represented in a manner that humans can use, even if only into RAM).
In the case of your collection of TS/TG porn images and movies, the drive purchaser certainly won't own them, because you probably didn't in the first place (most porn being pirated almost ad infinitum from the moment it first appears on the net). Unless, of course, you make original porn of yourself - in which case, my hat's off.
The internet side of the Bernardo-Homolka "ban" was even less effective. At the time what that really meant was Usenet discussion, and the ban amounted to the removal of a couple of alt groups, primarily alt.fan.karla-homolka. Mostly this only happened at a couple of universities, initially U. Waterloo as I recall (there weren't really a lot of commercial ISPs at the time - maybe mindspring in canada had a reasonable Usenet feed). It was also purely a CYA action, not at the direct request of law enforcement. Of course, on Usenet then as on Slashdot now, actual topics of discussion on a newsgroup rarely reflected the name or hierarchy of the group, particularly for a darkly-named alt group. The discussion simply continued on other groups, like the general Canadian city toplevels that were appearing at the time. In the end the discussion and reposting of US published stuff wasn't suppressed, but dispersed - which of course is exactly how the net is designed.
I didn't support the ban on campus servers then, but I've learned that there are good reasons for these on occasion. It would not have been a good thing if Bernardo had walked on a technicality (well, aside from the possibility of good old-fashioned vigilante justice). Regardless, an Internet ban is even more of a joke now than it was then. When you think about it, what this will increasingly do is effectively exclude the net-savvy from jury pools in highly publicised cases (with or without a ban - simple standard jury selection questioning will expose that they "know too much").
Yup. Fire suppression causes fires. You wouldn't leave high-energy fuel and tinder lying around in your basement (well, you shouldn't), but for centuries that's what people have been effectively hoarding on the landscape. Periodic burns are the natural state of most forests, especially conifers. Monoculture second growth and other lack of diversity hurts too (you get a dryer forest) but eternal fire suppression is the biggie.
You really can't cap anything in nature and not expect it to blow its top eventually.
But if people can't install and try out their hacks and bugfixes, there's a similarly slim risk of them bothering to look at the source at all.;)
I think you overestimate the difficulty of modifying embedded systems these days, especially when there'll be an obvious need for a firmware update mechanism for people to discover and adapt to their own uses. Depending on the difficulty of doing this is really just STO. So is closed-source on its own, but in this case I'd fully expect manufacturers and/or legislators to implement a cryptographic scheme so that only "official" code could run in the car. They're just not going to want to take the risk.
And I still don't see how an open source base will help the company meet its obligation to provide support. They can't very well say "ask around, it's open source".
I don't see how. Having newsgroup support for your old release of FreeGMC ten years after release won't help the company support it. If anything, extensive car control software might present an argument against open source, at least in the minds of legislators - having people hack the operating system of their home computer only endangers the nearby network, but people being able to try out hacks to their car on the highway? As it is there are legal limits to what mods you can apply to your car, but when those mods are software-based and hidden - well, suffice it to say that it'll be a can of worms.
Don't need a shocker. Just program the phone to start the vibrator when the speech level is too high. Gives the user the necessary tactile feedback without requiring an additional mechanism.
Better still, use audio feedback. Literally. Or for that matter, just simulate the cheap mics and A/D convertors of yore - too loud, and their voice gets crunchy and unintelligible. Feed this back into their ear and they'll naturally lower their voice to keep it under control.
Feedback is always better in the same sensory mode. Forcing the average preoccupied yapper to interpret a tactile response to an auditory stimulus will never be as automatic as keeping things in the same mode (even if it's only a carefully constructed simulation). I really do think this idea would work - you can see the mechanism in action every time someone initially speaks too loudly into a microphone. The trick would be creating adaptive logic so the phone only creates this effect when it's necessary and useful. If it ended up making someone whisper during their 911 call there'd be trouble.
What I can never get over is how incredibly inexpensive it is for these companies to massively influence legislation. From another excellent Reason article, linked from that marsupial interview:
...the company exploited its connections to get the copyright extension passed. The very day Senate Majority Leader Trent Lott became a co-sponsor of the bill, the Center for Responsive Politics reports, the Disney Political Action Committee donated $1,000 to his campaign chest; within a month, it had also sent $20,000 in soft money to the National Republican Senatorial Committee.
Say it with me in your best Dr. Evil accent: "One *thousand* dollars!". And from Disney's bank account? Boy, that's gotta smart. For that matter, why are the legislators and parties affected in the least by these paltry sums?
They may not be paid huge salaries, but they can't be that broke. If this is all it takes to get laws passed, perhaps all we need to do is take up a collection. Even I can afford $1000 for some juicy bill.
(The same thing impressed me with the Salt Lake City / IOC scandal - so you can get your own Olympics for a few pizzas now?).
Cute thought experiment, but I don't think it matters. The end result of the vector math is still a point in between the bodies (the centre if they're of identical mass). The orbits will appear the same at all times regardless of whether gravity is constrained within light cones or not, and this is why it's such a pickle of a problem. In fact I suspect it's the sort of problem that will lead to other unexpected understanding simply because one has to be so devious to try and measure it.
If you don't believe in the vector math method (that the bodies orbit a central gravitic point, just as, say, a dust ring or ringworld would) try thinking of it this way: each body is orbiting the [displaced phantom of] the other, but because their orbits are complimentary it still doesn't matter. That is, if only one body was affected then the binary system would go spinning crazily away, but because their respective motions necessarily complement one another, it again doesn't matter - with either method, the phantomicitys you're concerned about will exactly cancel each other out.
Same applies if the bodies differ in mass, of course, though the math is a bit harder.;)
It is not your IQ, dick length/cup size, value as a human being, or a score in a video game.
Claiming it's not a score in a video game is just plain denial. Try and formulate an argument that truly distinguishes/. from any other MUD. Everyone knows when you get to level 50 and become a "demigod", all the fun goes out of it. All this talk about +2 and +3 bonuses (not to mention "swords" of various length - and girth) only confirms where we are.
... all so that CostaRicans can use less pesticides, make more money from all of us banana loving Westerners
Actually, it's more about securing the food supply in places where, if bananas succumbed to disease, millions of people would starve (most bananas aren't for fat Westerners, but for skinny Africans). This is in fact more-or-less guaranteed the way things stand now, and it'll make the Irish potato famine look like mild dyspepsia if it happens.
Normally I'm pretty dead-set against GM foods, but in the specific case of organisms that can't reproduce on their own (so you're not unleashing such a potential frankenfruit on the world) it might be ok. When the GM organism can and will run amuck in the real world, that's a different story.
Since the usual cash incentive is less pronounced here I'd also expect there to be less danger of this just creating a different monoculture problem. There are more likely to be smaller, more divergent approaches. But it's also likely to take longer, and there's probably a bit of a time limit.
Oh, you can pretty easily argue that it's evil for a company to profit. Think of where profit actually comes from:
(a) by underpaying your employees (b) by overcharging your customers
There really isn't any alternative - for a company (or individual capitalist) to profit isn't the same sort of "wealth creation" as individual human labour and ingenuity provides. We could have an economy of zero-sum employment collectives; we just happen not to. Complicating matters with notions like "public" ownership of companies doesn't change the basic equation from the simplest case. Profit is necessarily skimmed off the top. We accept this because of the belief that the skimmed funds (aka human effort) will be wisely reused - when the capitalist entity is an individual, this is called philanthropy. The problem is that you're at that individual's whims - if Andrew Carnegie happens to like libraries at the moment, that's what yer gonna get. Most of the laws governing public companies are in fact an attempt to limit that sort of power.
There doesn't seem to be much point in overwriting more than once with the same zero pattern (the article makes this mistake too, though the original authors probably don't). There are really two levels of sophistication we're hoping to elude here:
a) People using the drive's own interface to retrieve "deleted" data
b) People doing direct signal analysis of the magnetic media to find successive generations of overwritten data
Once you've overwritten the disk once (whether with dd, a real SCSI low-level format, or some other means), you're in regime (b). Assuming you're paranoid and/or justifiably concerned enough to bother with repeated writes, using the same bit pattern does little - and zeroing is especially non-optimal, from what I've read. Random bit patterns seem a likely candidate, but randomness is actually particularly easy to divine in a signal.
People have experimented with instead writing various repetitions of constant strings with good success, but what might be ideal is a chaotic pattern that approximates the look of the expected data without divulging anything real (interesting thought - perhaps this is what some of the porn they found was for!). Write that a few times and you have a honeypot that might mislead a naive investigator into thinking there's nothing more to be found - but even this is difficult because the "freshness" of the bit patterns can be determined by their relative signal strength, and you can't simulate age using the default write current no matter how many new patterns you lay on. You can only hope you've made the old, real data so faint that it disappears into the background noise. Since there's no real way to guarantee this, people with real secrets to hide have to physically destroy the media. So much for reduce, reuse, recycle.;)
The technique of extracting the data is akin to the work of deep-sky astronomers, military listening posts, or even sedimentary archaeology. It's quite an interesting problem, as is making the data unrecognisable. The parallel with copy-protection is obvious, and the outcome is the same - an escalating war of technique between intrigued hackers, where the party acting later in time (the deprotector / signal analyst) always has an advantage.
As an aside, when using dd to copy large amounts of data to disk you can often speed things up immensely by tailoring the (output) block size to the destination device.
You can do this on Mac IE as well, co-opting the inbuilt address bar '?' command to go to Google instead of MSN. It's just a resedit away (a quickconvert & resedit away on Mac OS X).
Slashdot Mirror
That used to be true, but nowadays the LINEAR project picks up most of them. Amateur astronomers still account for a few a year, though. There are a lot of comets out there to be found.
Sea water is nearly incompressible, so the viscosity wouldn't be appreciably higher. If it was, deep-marine fish like ratfish would have a tough time (especially considering how few calories there are down there for the poor things to eat). Sea water viscosity actually depends chiefly on the temperature (colder is more viscous), and to a lesser degree on salinity (more saline is more viscous - this obviously implies that sea water is slightly more viscous than fresh water).
There are still all kinds of other effects that make diving to that depth very, very difficult. One is the corrosive effect of oxygen in the water at that pressure.
Anyway, wouldn't a more viscous fluid be better for this thing's method of submergence? It doesn't need to go fast, it just needs to generate downward pressure with the reverse equivalent of an airfoil (or something). Planes get more lift in more viscous air, at least to the point where they can still move forward quickly enough.
And of course the GPL'd mldonkey, for those who aren't partial to Windows.
Actually, this is a reason to avoid buying new hardware.
Remember kids, "Recycle" is a distant third among the three Rs. They say "Reduce, Reuse, Recycle" in that order for a reason.
Only if you quaintly assume that being president has something to do with having a bountiful clue, or being a kind reasonable person.
Woz would make a great technical or education advisor, but probably a lousy president.
It's likely a very small amount. These windows might not even be active-matrix, and they're monochrome. I'd guess less than a watt for even a large pane. Though I still agree with your point, we'd probably all do better to replace aging refrigerators than worry too much about the power demand from this. ;)
Which does make some sense, considering to use a projection TV (or other projector) you really have to make the room dark. It has a certain elegance. The window-speakers (which aren't all that new, I think there was a /. story about them at least a year or two ago) are a more questionable idea, and likely wouldn't fly in any burgh that gives a crap about neighbourhood sound pollution.
Half-right. From the FA:
The article doesn't mention the underlying technology by name, but it's probably simply an LCD panel similar to that in a notebook display. Whether the relaxed liquid crystal strands uncurl with (white, transparent) or against (black, opaque) the plane of the polarisation filter is always just a manufacturing choice. In notebooks it can be a matter of saving power by minimising the amount of screen you need to change from the default, on average. A white-on-black character display should probably relax to black, while a black-on-white Mac-style windowed display might better relax to white. Of course I'm not claiming they always use that much logic in the decision (Apple maybe since they seem to put a lot of thought into powersaving, but more likely they just buy what's cheapest like everyone else).
The tendency of the windows in the article to relax to opacity could be thought of as a privacy feature, I suppose. It does seem odd when the device is being sold as a window that can be used as a screen, though (as opposed to the converse).
Of course, they'll see a horizontally flipped representation. That could colour their impression of you too, especially if you prefer that sort of porn where it's already difficult to tell who's doing what to whom.
They probably do. With 32k games, it could only be...
...an Atari 2600 with a broadband adaptor! Yes! The return of Gameline!
Doubtful. Consider that you never own the copyrighted-by-others software on your drive (including the GPL stuff), only an explicit licence to use it. Data on the drive that you author or even just compile will likely still belong to you, regardless of the fact that you've sold the media. In the case of deleted or overwritten data this is even more defencible, especially with legislation like the DMCA - deletion here would be the "copy-protection" method (remember that the data must be copied to be represented in a manner that humans can use, even if only into RAM).
In the case of your collection of TS/TG porn images and movies, the drive purchaser certainly won't own them, because you probably didn't in the first place (most porn being pirated almost ad infinitum from the moment it first appears on the net). Unless, of course, you make original porn of yourself - in which case, my hat's off.
The irony was in demonstrating that you can push and lobby a federal judge into affirming any damn thing you like...
...but not until a couple of years after it ceases to matter.
Thus the system is corrupt, but so slow that it's all fair in the end. Didn't Frank Herbert foresee this with the Bureau of Sabotage?
The internet side of the Bernardo-Homolka "ban" was even less effective. At the time what that really meant was Usenet discussion, and the ban amounted to the removal of a couple of alt groups, primarily alt.fan.karla-homolka. Mostly this only happened at a couple of universities, initially U. Waterloo as I recall (there weren't really a lot of commercial ISPs at the time - maybe mindspring in canada had a reasonable Usenet feed). It was also purely a CYA action, not at the direct request of law enforcement. Of course, on Usenet then as on Slashdot now, actual topics of discussion on a newsgroup rarely reflected the name or hierarchy of the group, particularly for a darkly-named alt group. The discussion simply continued on other groups, like the general Canadian city toplevels that were appearing at the time. In the end the discussion and reposting of US published stuff wasn't suppressed, but dispersed - which of course is exactly how the net is designed.
I didn't support the ban on campus servers then, but I've learned that there are good reasons for these on occasion. It would not have been a good thing if Bernardo had walked on a technicality (well, aside from the possibility of good old-fashioned vigilante justice). Regardless, an Internet ban is even more of a joke now than it was then. When you think about it, what this will increasingly do is effectively exclude the net-savvy from jury pools in highly publicised cases (with or without a ban - simple standard jury selection questioning will expose that they "know too much").
Yup. Fire suppression causes fires. You wouldn't leave high-energy fuel and tinder lying around in your basement (well, you shouldn't), but for centuries that's what people have been effectively hoarding on the landscape. Periodic burns are the natural state of most forests, especially conifers. Monoculture second growth and other lack of diversity hurts too (you get a dryer forest) but eternal fire suppression is the biggie.
You really can't cap anything in nature and not expect it to blow its top eventually.
But if people can't install and try out their hacks and bugfixes, there's a similarly slim risk of them bothering to look at the source at all. ;)
I think you overestimate the difficulty of modifying embedded systems these days, especially when there'll be an obvious need for a firmware update mechanism for people to discover and adapt to their own uses. Depending on the difficulty of doing this is really just STO. So is closed-source on its own, but in this case I'd fully expect manufacturers and/or legislators to implement a cryptographic scheme so that only "official" code could run in the car. They're just not going to want to take the risk.
And I still don't see how an open source base will help the company meet its obligation to provide support. They can't very well say "ask around, it's open source".
Isometric 3D display? Mouse-driven?
Obviously these rogues have just ripped off Diablo.
I don't see how. Having newsgroup support for your old release of FreeGMC ten years after release won't help the company support it. If anything, extensive car control software might present an argument against open source, at least in the minds of legislators - having people hack the operating system of their home computer only endangers the nearby network, but people being able to try out hacks to their car on the highway? As it is there are legal limits to what mods you can apply to your car, but when those mods are software-based and hidden - well, suffice it to say that it'll be a can of worms.
Better still, use audio feedback. Literally. Or for that matter, just simulate the cheap mics and A/D convertors of yore - too loud, and their voice gets crunchy and unintelligible. Feed this back into their ear and they'll naturally lower their voice to keep it under control.
Feedback is always better in the same sensory mode. Forcing the average preoccupied yapper to interpret a tactile response to an auditory stimulus will never be as automatic as keeping things in the same mode (even if it's only a carefully constructed simulation). I really do think this idea would work - you can see the mechanism in action every time someone initially speaks too loudly into a microphone. The trick would be creating adaptive logic so the phone only creates this effect when it's necessary and useful. If it ended up making someone whisper during their 911 call there'd be trouble.
What I can never get over is how incredibly inexpensive it is for these companies to massively influence legislation. From another excellent Reason article, linked from that marsupial interview:
Say it with me in your best Dr. Evil accent: "One *thousand* dollars!". And from Disney's bank account? Boy, that's gotta smart. For that matter, why are the legislators and parties affected in the least by these paltry sums? They may not be paid huge salaries, but they can't be that broke. If this is all it takes to get laws passed, perhaps all we need to do is take up a collection. Even I can afford $1000 for some juicy bill.
(The same thing impressed me with the Salt Lake City / IOC scandal - so you can get your own Olympics for a few pizzas now?).
Cute thought experiment, but I don't think it matters. The end result of the vector math is still a point in between the bodies (the centre if they're of identical mass). The orbits will appear the same at all times regardless of whether gravity is constrained within light cones or not, and this is why it's such a pickle of a problem. In fact I suspect it's the sort of problem that will lead to other unexpected understanding simply because one has to be so devious to try and measure it.
;)
If you don't believe in the vector math method (that the bodies orbit a central gravitic point, just as, say, a dust ring or ringworld would) try thinking of it this way: each body is orbiting the [displaced phantom of] the other, but because their orbits are complimentary it still doesn't matter. That is, if only one body was affected then the binary system would go spinning crazily away, but because their respective motions necessarily complement one another, it again doesn't matter - with either method, the phantomicitys you're concerned about will exactly cancel each other out.
Same applies if the bodies differ in mass, of course, though the math is a bit harder.
Claiming it's not a score in a video game is just plain denial. Try and formulate an argument that truly distinguishes /. from any other MUD. Everyone knows when you get to level 50 and become a "demigod", all the fun goes out of it. All this talk about +2 and +3 bonuses (not to mention "swords" of various length - and girth) only confirms where we are.
Time to create yet another character...
Actually, it's more about securing the food supply in places where, if bananas succumbed to disease, millions of people would starve (most bananas aren't for fat Westerners, but for skinny Africans). This is in fact more-or-less guaranteed the way things stand now, and it'll make the Irish potato famine look like mild dyspepsia if it happens.
Normally I'm pretty dead-set against GM foods, but in the specific case of organisms that can't reproduce on their own (so you're not unleashing such a potential frankenfruit on the world) it might be ok. When the GM organism can and will run amuck in the real world, that's a different story.
Since the usual cash incentive is less pronounced here I'd also expect there to be less danger of this just creating a different monoculture problem. There are more likely to be smaller, more divergent approaches. But it's also likely to take longer, and there's probably a bit of a time limit.
The BBC has more on the subject.
After this maybe we can start to tackle the apple problem (currently the #1 fruit for pesticide use, similarly because of massive monoculture).
Oh, you can pretty easily argue that it's evil for a company to profit. Think of where profit actually comes from:
(a) by underpaying your employees
(b) by overcharging your customers
There really isn't any alternative - for a company (or individual capitalist) to profit isn't the same sort of "wealth creation" as individual human labour and ingenuity provides. We could have an economy of zero-sum employment collectives; we just happen not to. Complicating matters with notions like "public" ownership of companies doesn't change the basic equation from the simplest case. Profit is necessarily skimmed off the top. We accept this because of the belief that the skimmed funds (aka human effort) will be wisely reused - when the capitalist entity is an individual, this is called philanthropy. The problem is that you're at that individual's whims - if Andrew Carnegie happens to like libraries at the moment, that's what yer gonna get. Most of the laws governing public companies are in fact an attempt to limit that sort of power.
There doesn't seem to be much point in overwriting more than once with the same zero pattern (the article makes this mistake too, though the original authors probably don't). There are really two levels of sophistication we're hoping to elude here:
a) People using the drive's own interface to retrieve "deleted" datab) People doing direct signal analysis of the magnetic media to find successive generations of overwritten data
Once you've overwritten the disk once (whether with dd, a real SCSI low-level format, or some other means), you're in regime (b). Assuming you're paranoid and/or justifiably concerned enough to bother with repeated writes, using the same bit pattern does little - and zeroing is especially non-optimal, from what I've read. Random bit patterns seem a likely candidate, but randomness is actually particularly easy to divine in a signal.
People have experimented with instead writing various repetitions of constant strings with good success, but what might be ideal is a chaotic pattern that approximates the look of the expected data without divulging anything real (interesting thought - perhaps this is what some of the porn they found was for!). Write that a few times and you have a honeypot that might mislead a naive investigator into thinking there's nothing more to be found - but even this is difficult because the "freshness" of the bit patterns can be determined by their relative signal strength, and you can't simulate age using the default write current no matter how many new patterns you lay on. You can only hope you've made the old, real data so faint that it disappears into the background noise. Since there's no real way to guarantee this, people with real secrets to hide have to physically destroy the media. So much for reduce, reuse, recycle. ;)
The technique of extracting the data is akin to the work of deep-sky astronomers, military listening posts, or even sedimentary archaeology. It's quite an interesting problem, as is making the data unrecognisable. The parallel with copy-protection is obvious, and the outcome is the same - an escalating war of technique between intrigued hackers, where the party acting later in time (the deprotector / signal analyst) always has an advantage.
As an aside, when using dd to copy large amounts of data to disk you can often speed things up immensely by tailoring the (output) block size to the destination device.
You can do this on Mac IE as well, co-opting the inbuilt address bar '?' command to go to Google instead of MSN. It's just a resedit away (a quickconvert & resedit away on Mac OS X).