A while back the US Post Office was running public service ads to inform people "If you didn't order it but they sent it to you anyway, it's yours free."
If DC sent me a scanner, I would take it as a gift, to do with as I please.
This guy's idea seems like a workable solution: many companies pull stuff like this knowing that 90% of the sheeple will pay, 9% will complain to the company but take no other action, and.0009% (maybe) will actually try to get the company in legal trouble. If we can/. that last number up in a significant number of cases, we may change the cost/risk ratio for these companies and get them to change their ways.
Having been unfortunate enough to have worked with Windows NT (writing drivers for custom hardware my employer makes) I can tell you that NT is NOT written in C++. Actually, the original NT kernel was written in Pascal (now there's a thought to cause you to awaken screaming).
However, the biggest problem with using C++ for shared libraries under any platform is that there is NO spec for
How to mangle a name
how to pass data on the stack (pure C++ linkage functions need not follow the C calling standard: it is perfectly valid for the compiler to generate them all callee-pops, or pass args in registers, or anything else. You want C style arg passing, you have to use extern "C")
how to lay out structures with base classes or virtual functions
As a result, it is not possible to guarantee that file x.so, created with g++, is linkable to y.so, created with Borland C++. It's a shame that with all the other CRAP the ANSI committee dragged into the language during standardization, they couldn't have spelled out at least how to mangle the damn names!
As a big fan of C++ myself, I'd LOVE to see the kernel, X, and the windowing toolkit be true C++ classes: this is systems programming and that's what Stroustrup created the language for.
However, I take exception (pun intended) at the claim that QT is C++: It isn't. It is another language, that you then must run through Qt's MOC preprocessor to make C++. If TrollTech had truly make Qt C++, they would have used virtual functions or functors to implement the connection to the windowing message system, not this mutant "Signals and Slots" stuff. That's what got me when I looked at KDE: I want to write in C++, not some mutant version of something that is almost C++. It's almost as bad as working with somebody who uses
#define AND &&
#define OR ||
#define NOT !
#define BEGIN {
#define END }
Take a kilo of $element (measured to an accuracy of 1 ppm): Given the atomic weight of $element, you know how many atoms of $element you have (to an accuracy of 1 ppm). You know that when $element decays it emits N $particles: so set up a particle counter and count how many $particles are emitted in $time: This give you the number of atoms that decay per unit time for that many atoms. You derive the decay rate of $element from that.
Given that $element1 decays into $element2, how do we know that there wasn't any $element2 in the mix to start with?
In the cases that are used, the decay product of $element1 isn't any old isotope of $element2, but an isotope that is not produced by any other form of decay. Therefor, you know that every atom of $element2<isotope-x> was produced by the decay of an atom of $element1.
Couple of things you've all missed
on
Focusing Audio
·
· Score: 2
The sound level is 80-90dbA. This isn't very loud: about the level of normal conversation at best. You throw that at a wall, and let it disperse, and you now have a whisper. They'll have to make it a lot louder, and that will be difficult because...
The sound is created by generating a LOT of ultrasound, then letting it create mixing products in the audible range. Let me explain:
Given two signals at frequencies f1 and f2 impinging upon a non-linear transmission element, the resultant signal will be composed of four frequency elements: the original f1 and f2, and two new elements (f1-f2) and (f1+f2). The power in the original signal will be split between the signals. The amount of the split depends upon the magnitude of the non-linearity: a completely non-linear transfer element will completely remove the origional elements, while medium that is only slightly non-linear will transfer only a little power.
Now, the problem is this: air isn't very non-linear. So, to get a watt of audible sound, they are going to have to push, ohh, say 10 watts of ultrasound out of the speakers. One watt goes into sound, nine goes into heating of the air as it absorbs the ultrasound. Just what I want at the next (already smoky and hot) concert I go to: a megawatt of heat being added to the air!
Given the general level of paranoia around here, I'm suprised nobody picked up on the fact that the same phased-array techniques that allow this to work can be used to make a highly directional MICROPHONE. Imagine one of these set up in a public place, suitably camoflaged behind ceiling tiles.
Hey Joe, let's punch up what those two are saying: they look like dissidents
The only reason I don't use Google more and Altavista less it that AV allows more precision in searching by the "near" keyword and the not keyword:
Linux near Slashdot and not (Portman or grits or "ick your as")
I do wish Google would get these, as they really let me thin out the garbage: I usually append "and not (homepage or jumppage or links or "link page")" and remove about a zillion wastes of my time from any search.
it now seems to be legal to export strong crypto like PGP from the USA..
Maybe PGP, but I make the test gear for the radios the FBI uses, which use a 56 bit single DES, and we cannot export that equipment with the DES functions in place.
Look, if you are going to claim you are trying to predict the future, have some freaking content! This article had nothing about how we would get from here to there, nothing about what advances would make this machine possible (other than vague platitudes) and nothing about how this would fit into the world it lived in. To the authors of this article: Go read Heinlein, or Niven, or Asimov, or Clarke, or any other decent futurist before operating your keyboard! (because I seriously doubt any professional journalist would want to dictate his stories to a speech recognition system.)
rather than holding the cellular companies responsible for their appalling lack of security
Actually, the cellular industry would have loved to deploy encryption. All GSM phones (PCS here in the US) have a very strong encryption algorithm in them, but here in the US the cellular carriers are prohibited by law from turning it on.
Furthermore, a US manufacturer who wishes to incorporate encryption in their product, even 64 bit single DES which is trivial to crack, cannot export it (a fact I know from personal experience).
You all have broadband, right? Why not just pick a time, and use one of the streaming audio protocols to chat, and capture the stream. It wouldn't be any worse than the other GIS's;^].
If you've never read Robert A. Heinlein's The Moon Is A Harsh Mistress, do so. There's a scene in the book that is quite apropos to this ruling:
In the book, our heros are trying to stage a revolution freeing the Moon from earth's tyranny. They are meeting with a group of The Powers That Be on earth to discuss the Moon's future. One of the main characters, Prof, who is the second most politically astute of Our Hereos, says something to a judge that the main character (Mannie) does not understand. Later, Mannie asks Prof what he said. Prof indicates that he asked the Honorable muckitimuck if his wife still worked in the whorehouse. Prof then explains that the whole goal of the meeting from the earth's side was to present a comprimise that would prevent the Moon from revolting, and that by stooping to personal insult, unreasonable behavior, and every other trick in the book to prevent that reasonable comprimise from being offered, the many residents of the Moon were kept fired up about the revolt, and therefor unwilling to accept a comprimise that in the long run would hurt them.
Now, in light of that summary, consider:
The judge's ruling was absolutely prejudicial, and will be overturned in a cocaine heartbeat the next level up.
We are all fired up about this, and some of us will write our congressdrones about this issue
I keep hearing/reading people saying that we need online voting to make voting easier.
BULLSHIT!
(and I'd use <font size=+5> if/. would let me do so!) <attire type="flamesuit">
The last thing the system needs is a bunch of morons who are too lazy to go down to the local polling place being able to vote. Much of the bad legislation around comes from politicians knowing that the average voter is clueless and will vote how he's told by TV.
Personally, I like Robert Heinlein's suggestion: you should not be allowed to vote unless you can find the roots of y=9x^2+12x+4. Now I know a lot of people will point to Jim Crowe laws (discrimination at the polling place to prevent blacks from voting) but guess what? Those days are OVER. GET OVER IT!
Also, I see a lot of teenagers bitching that they cannot vote. OK kid, tell you what. When you have proven that you have some degree of responsiblity, that you understand what it is to support yourself and see large amounts of your money being taken from you by force by well-meaning fools who will use it to assuage their own guilt, when you understand the consequences of your actions and the fact that some mistakes will outlive you, then you can vote.
I think you miss the point. Unfortunately, the political parties in large part control whom I may vote for on election day. Therefor, I should have a say in their selections.
Consider how it is now: the bulk of people tend to be moderate. The two major parties are anything but. Now, if I am a moderate, I have two choices: I register as a party member, select one candidate from one party in the primary, re-register as an independant on election day, and vote for the lesser of N evils, of which I had a hand in selecting only one. Or, I can NOT vote in the primary, and vote for the lesser of N evils, of which I had NO say in selecting. Either way, what happens is the parties avoid the more moderate candidates, and I get to pick Dumb or Dumber.
Now, what if everybody could vote in the primaries: the parties would then be forced to select individuals who were more in line with the majority in the primary. The average Joe/Josephine would have a better selection on Election Day. We might actually see participation increase.
Besides, why in THE HELL should the political parties (entities that have no especial standing under the Constitution) have access to whether I've registers Republican, Democrat, or Slashdot Party?
In short, I assert that the formation of the party machines has been one of the most detrimental developments in the American political scene: candidates need to campaign for the primary, then for the real election, more than doubling the amount of money they need (increasing their ties to PACs), and reducing the choices the voter has, driving down participation.
(I'm glad to see that you, like I, check your messages for replies. So few do....)
I'm glad to see you feel that freedom of speech is important. However, isn't it better by far to allow (those who you believe to be) fools to speak, and then demonstrate their foolishness, than to drive them out of town and prevent others from seeing their foolishness? That was the point I was trying to make: that "making a [Republican|Democrat|whatever] convention something a city wouldn't want to host" was counterproductive.
Also, while I support anybody who believe strongly enough in something to protest and risk arrest, too often "protesters" at these sort of things sound something like this:
Yeah man, they're like, the man, man, and they suck, man. Stick it to the man, man!
In other words, no clue why they are protesting, just being a tool of one or the other faction.
Personally, I'd like to see the electoral college replaced by direct election, the parties denied the use of public facilities for the primaries (which are not a part of the electoral system as set forth in the Constitution, as I am sure you know but many/.'ers may not), and the parties forced to allow independants like myself to vote in BOTH primaries. That, and denying incumbants the option of running for re-election (force them to sit out a term of office in the real world, or as close as that sort ever get to it) would go a long ways toward fixing the current mess we are in.
Sounds like we are a lot closer that we are apart. Shake?
So good to see you are very well versed in the art of debate....
The Republican convention is nothing about freedom of speech. It is an orgy of big business pouring money into the political system, buying politicians, and creating a fabricated and imaginary hope for the people.
And this would differ from the Democrat primary how?
People have the freedom to speak out against them just as much as they have the freedom to spread lies.
OK, so you say you aren't a Democrat. However, you still seem to feel that Republicans are less deserving of freedom of speech than the general populus, so my previous comment still stands. Sorry, but it is far more important to support the freedom of speech of those you disagree with than to support the freedom of speech of those you agree with.
The point is to make a Republican convention something that no city wants. Because obviously Republicans have no right to freedom of speech, only Democrats.
Amazing how narrowly most people define "freedom".
What Rob should have done was to post the story with no comment, then post a comment to the story with his political beliefs. This would have done three things:
It would have kept the editorial content seperate from the news content
It would have allowed Rob to actually state why he feels the way he does, thus allowing others to think about their own opinions with respect to any information in Rob's post
It would have screwed the "first post" morons.
Really Rob, if you want/. to be respected, you must be respectable. Keep your editorial opinions as seperate as possible from the news. Failure to do so is why I cancelled my local paper.
I'm glad to see your reply. Personally, DoubleClick doesn't bother me, because I run a filtering proxy (btw, I don't filter/.'s own ad server). I also don't care that/. uses DC on occasion. However, I just thought it interesting that Hemos was saying that he couldn't do anything about it. While I suspected something much like what you described, I thought it best to ask and allow/. to remove all doubt.
And just try that with most of the rest of the media!
I've managed to aquire a fragment of the source for the latest Microsoft program. However, I am a Linux and embedded programmer, so can anybody here help me decipher it?
I too have been programming professionally for over a decade (in C++), and here's what I've learned:
Don't be afraid to fail.
Sometimes, the problem definition is too vague, or there are too many unknowns in the mix. You have to follow the scientific method, and the first step of the scientific method is gather data.
You have a module that you cannot figure out how to do? cvs commit, then say to yourself, "Self, we're going to write something to see what is what. When we are done, we will delete what we have written, and using what we've learned, write it properly". You'll be amazed at how much clearer the hidden requirements become when you start bashing code against them.
Now, as to the C++ issue: I don't know if part of the problem is your level of skill in C++, or if that is just a red herring. If you really aren't comfortable using OOD, then for Torvalds' sake DON'T! Just write the code proceedurally. Again, I've often found that once I've written a code block, I can then better see where the objects are within the code. Then, I re-write the code to be OO.
In my experience, no object is reusable from the beginning, no matter how much work you put into it, no matter how long you worked on the design. Only after the third time a class has been used have you come close enough to defining what functions need to be virtual, what members non-private, what args need to be passed to the constructor, etc., to make the class really reusable. Accept it and move on.
Like I said, don't be afraid to fail. It's the only way you will succeed.
We're not talking kilonewtons of thrust here: probably closer to decinewtons. True, the cable will pull a little in the east/west direction, but gravitational tidal forces will tend to correct this. The thing to remember is that a little force over a long time is better than a lot of force over a little time.
The shuttle's solid rocket boosters basically use thermite: aluminum mixed with ammonium perchlorate if I remember correctly, with a poylmer stabilizer to control the burn rate. The result is aluminum oxide plus other stuff. Aluminum is a toxic chemical if it gets into your system.
The earth's magnetic field extends north and south, and rotates at 2pi radians/day (same as the earth).
The tether is vertical.
The motion is east/west
So far, you were quite correct. Now, the system operates in one of 2 modes: generator or motor.
In Generator mode, the current is the cross product of velocity and field direction, in other words up and down along the wire.
In motor mode, the force is the cross product of field direction and current flow, in other words east/west along the orbital path.
Now, below geosync orbit, generating power will bring the satelite down earth, but above geosync the sat will actually be pushed higher. Much the same sort of thing is happening with the Moon: Earth's tidal forces are dragging the Moon around and pushing it away. Different cause (gravity vs. electromagnetism), but the same effect.
Also, at geosync, you want the sat to say put. If it drifts away from geosync, then you are moving, and can use the earth's field to get back.
Slashdot Mirror
A while back the US Post Office was running public service ads to inform people "If you didn't order it but they sent it to you anyway, it's yours free."
.0009% (maybe) will actually try to get the company in legal trouble. If we can /. that last number up in a significant number of cases, we may change the cost/risk ratio for these companies and get them to change their ways.
If DC sent me a scanner, I would take it as a gift, to do with as I please.
This guy's idea seems like a workable solution: many companies pull stuff like this knowing that 90% of the sheeple will pay, 9% will complain to the company but take no other action, and
But then again, probably not.
However, the biggest problem with using C++ for shared libraries under any platform is that there is NO spec for
As a result, it is not possible to guarantee that file x.so, created with g++, is linkable to y.so, created with Borland C++. It's a shame that with all the other CRAP the ANSI committee dragged into the language during standardization, they couldn't have spelled out at least how to mangle the damn names!
As a big fan of C++ myself, I'd LOVE to see the kernel, X, and the windowing toolkit be true C++ classes: this is systems programming and that's what Stroustrup created the language for.
However, I take exception (pun intended) at the claim that QT is C++: It isn't. It is another language, that you then must run through Qt's MOC preprocessor to make C++. If TrollTech had truly make Qt C++, they would have used virtual functions or functors to implement the connection to the windowing message system, not this mutant "Signals and Slots" stuff. That's what got me when I looked at KDE: I want to write in C++, not some mutant version of something that is almost C++. It's almost as bad as working with somebody who uses
There. I said it. I feel better.
me->Attire(Attire::Flamesuit);
How do we know the half-life of $element
Take a kilo of $element (measured to an accuracy of 1 ppm): Given the atomic weight of $element, you know how many atoms of $element you have (to an accuracy of 1 ppm). You know that when $element decays it emits N $particles: so set up a particle counter and count how many $particles are emitted in $time: This give you the number of atoms that decay per unit time for that many atoms. You derive the decay rate of $element from that.
Given that $element1 decays into $element2, how do we know that there wasn't any $element2 in the mix to start with?
In the cases that are used, the decay product of $element1 isn't any old isotope of $element2, but an isotope that is not produced by any other form of decay. Therefor, you know that every atom of $element2<isotope-x> was produced by the decay of an atom of $element1.
Now, the problem is this: air isn't very non-linear. So, to get a watt of audible sound, they are going to have to push, ohh, say 10 watts of ultrasound out of the speakers. One watt goes into sound, nine goes into heating of the air as it absorbs the ultrasound. Just what I want at the next (already smoky and hot) concert I go to: a megawatt of heat being added to the air!
Hmmm....
I do wish Google would get these, as they really let me thin out the garbage: I usually append "and not (homepage or jumppage or links or "link page")" and remove about a zillion wastes of my time from any search.
Maybe PGP, but I make the test gear for the radios the FBI uses, which use a 56 bit single DES, and we cannot export that equipment with the DES functions in place.
Look, if you are going to claim you are trying to predict the future, have some freaking content! This article had nothing about how we would get from here to there, nothing about what advances would make this machine possible (other than vague platitudes) and nothing about how this would fit into the world it lived in. To the authors of this article: Go read Heinlein, or Niven, or Asimov, or Clarke, or any other decent futurist before operating your keyboard! (because I seriously doubt any professional journalist would want to dictate his stories to a speech recognition system.)
Mr. Victor Borge had this one nailed back in the '70s with his phonetic pronunciation
Actually, the cellular industry would have loved to deploy encryption. All GSM phones (PCS here in the US) have a very strong encryption algorithm in them, but here in the US the cellular carriers are prohibited by law from turning it on.
Furthermore, a US manufacturer who wishes to incorporate encryption in their product, even 64 bit single DES which is trivial to crack, cannot export it (a fact I know from personal experience).
You all have broadband, right? Why not just pick a time, and use one of the streaming audio protocols to chat, and capture the stream. It wouldn't be any worse than the other GIS's ;^].
In the book, our heros are trying to stage a revolution freeing the Moon from earth's tyranny. They are meeting with a group of The Powers That Be on earth to discuss the Moon's future. One of the main characters, Prof, who is the second most politically astute of Our Hereos, says something to a judge that the main character (Mannie) does not understand. Later, Mannie asks Prof what he said. Prof indicates that he asked the Honorable muckitimuck if his wife still worked in the whorehouse. Prof then explains that the whole goal of the meeting from the earth's side was to present a comprimise that would prevent the Moon from revolting, and that by stooping to personal insult, unreasonable behavior, and every other trick in the book to prevent that reasonable comprimise from being offered, the many residents of the Moon were kept fired up about the revolt, and therefor unwilling to accept a comprimise that in the long run would hurt them.
Now, in light of that summary, consider:
In many ways, this ruling was a good thing.
(and I'd use <font size=+5> if
<attire type="flamesuit">
The last thing the system needs is a bunch of morons who are too lazy to go down to the local polling place being able to vote. Much of the bad legislation around comes from politicians knowing that the average voter is clueless and will vote how he's told by TV.
Personally, I like Robert Heinlein's suggestion: you should not be allowed to vote unless you can find the roots of y=9x^2+12x+4. Now I know a lot of people will point to Jim Crowe laws (discrimination at the polling place to prevent blacks from voting) but guess what? Those days are OVER. GET OVER IT!
Also, I see a lot of teenagers bitching that they cannot vote. OK kid, tell you what. When you have proven that you have some degree of responsiblity, that you understand what it is to support yourself and see large amounts of your money being taken from you by force by well-meaning fools who will use it to assuage their own guilt, when you understand the consequences of your actions and the fact that some mistakes will outlive you, then you can vote.
</attire>
I think you miss the point. Unfortunately, the political parties in large part control whom I may vote for on election day. Therefor, I should have a say in their selections.
Consider how it is now: the bulk of people tend to be moderate. The two major parties are anything but. Now, if I am a moderate, I have two choices: I register as a party member, select one candidate from one party in the primary, re-register as an independant on election day, and vote for the lesser of N evils, of which I had a hand in selecting only one. Or, I can NOT vote in the primary, and vote for the lesser of N evils, of which I had NO say in selecting. Either way, what happens is the parties avoid the more moderate candidates, and I get to pick Dumb or Dumber.
Now, what if everybody could vote in the primaries: the parties would then be forced to select individuals who were more in line with the majority in the primary. The average Joe/Josephine would have a better selection on Election Day. We might actually see participation increase.
Besides, why in THE HELL should the political parties (entities that have no especial standing under the Constitution) have access to whether I've registers Republican, Democrat, or Slashdot Party?
In short, I assert that the formation of the party machines has been one of the most detrimental developments in the American political scene: candidates need to campaign for the primary, then for the real election, more than doubling the amount of money they need (increasing their ties to PACs), and reducing the choices the voter has, driving down participation.
I'm glad to see you feel that freedom of speech is important. However, isn't it better by far to allow (those who you believe to be) fools to speak, and then demonstrate their foolishness, than to drive them out of town and prevent others from seeing their foolishness? That was the point I was trying to make: that "making a [Republican|Democrat|whatever] convention something a city wouldn't want to host" was counterproductive.
Also, while I support anybody who believe strongly enough in something to protest and risk arrest, too often "protesters" at these sort of things sound something like this:
In other words, no clue why they are protesting, just being a tool of one or the other faction.
Personally, I'd like to see the electoral college replaced by direct election, the parties denied the use of public facilities for the primaries (which are not a part of the electoral system as set forth in the Constitution, as I am sure you know but many
Sounds like we are a lot closer that we are apart. Shake?
So good to see you are very well versed in the art of debate....
And this would differ from the Democrat primary how?
OK, so you say you aren't a Democrat. However, you still seem to feel that Republicans are less deserving of freedom of speech than the general populus, so my previous comment still stands. Sorry, but it is far more important to support the freedom of speech of those you disagree with than to support the freedom of speech of those you agree with.
The point is to make a Republican convention something that no city wants.
Because obviously Republicans have no right to freedom of speech, only Democrats.
Amazing how narrowly most people define "freedom".
Really Rob, if you want
I'm glad to see your reply. Personally, DoubleClick doesn't bother me, because I run a filtering proxy (btw, I don't filter /.'s own ad server). I also don't care that /. uses DC on occasion. However, I just thought it interesting that Hemos was saying that he couldn't do anything about it. While I suspected something much like what you described, I thought it best to ask and allow /. to remove all doubt.
And just try that with most of the rest of the media!
I've managed to aquire a fragment of the source for the latest Microsoft program. However, I am a Linux and embedded programmer, so can anybody here help me decipher it?
winGunHandle pGun = WinGetGunHandle(WIN_DESERTEAGLE_50);
winClipHandle pClip = WinGetClipHandle(TEN_ROUND_CLIP);
winFootHandle pFoot = winGetAppendageHandle(WIN_LEFT_FOOT);
winLockClip(pGun,pClip);
winCycleAction(pGun);
winSafety(pGun,WIN_FIRE);
winAimGun(pGun,pFoot);
while (winClipcount(pClip) > 0)
winSqueezeTrigger(pGun);
I think this has something to do with the new pricing/licensing system, but I'm not sure....
Believe me, if I had my way, we wouldn't be using it.
If it's not up to you, User ID #2, then who is it up to?
Seriously, cannot you and Rob make your own decisions on who's banner ad service your system runs?
I too have been programming professionally for over a decade (in C++), and here's what I've learned:
Don't be afraid to fail.
Sometimes, the problem definition is too vague, or there are too many unknowns in the mix. You have to follow the scientific method, and the first step of the scientific method is gather data.
You have a module that you cannot figure out how to do? cvs commit, then say to yourself, "Self, we're going to write something to see what is what. When we are done, we will delete what we have written, and using what we've learned, write it properly". You'll be amazed at how much clearer the hidden requirements become when you start bashing code against them.
Now, as to the C++ issue: I don't know if part of the problem is your level of skill in C++, or if that is just a red herring. If you really aren't comfortable using OOD, then for Torvalds' sake DON'T! Just write the code proceedurally. Again, I've often found that once I've written a code block, I can then better see where the objects are within the code. Then, I re-write the code to be OO.
In my experience, no object is reusable from the beginning, no matter how much work you put into it, no matter how long you worked on the design. Only after the third time a class has been used have you come close enough to defining what functions need to be virtual, what members non-private, what args need to be passed to the constructor, etc., to make the class really reusable. Accept it and move on.
Like I said, don't be afraid to fail. It's the only way you will succeed.
What keeps a script kiddie from sending spoofed itrace packets implicating every machine on the planet from a comprimised machine?
We're not talking kilonewtons of thrust here: probably closer to decinewtons. True, the cable will pull a little in the east/west direction, but gravitational tidal forces will tend to correct this. The thing to remember is that a little force over a long time is better than a lot of force over a little time.
The shuttle's solid rocket boosters basically use thermite: aluminum mixed with ammonium perchlorate if I remember correctly, with a poylmer stabilizer to control the burn rate. The result is aluminum oxide plus other stuff. Aluminum is a toxic chemical if it gets into your system.
So far, you were quite correct. Now, the system operates in one of 2 modes: generator or motor.
Now, below geosync orbit, generating power will bring the satelite down earth, but above geosync the sat will actually be pushed higher. Much the same sort of thing is happening with the Moon: Earth's tidal forces are dragging the Moon around and pushing it away. Different cause (gravity vs. electromagnetism), but the same effect.
Also, at geosync, you want the sat to say put. If it drifts away from geosync, then you are moving, and can use the earth's field to get back.