"No there certainly is no analogy between security tools like SAINT that can be misused and guns..."
Too bad you meant this bit ironically, because I agree with it literally. The day I see a widely-available computer prog (security or otherwise) made with the express purpose of killing people (and/or things), the way guns are, I'm going to move to Burkina Faso and live in a cave.
Funny, my Glock 19 pistol has put holes in paper targets, not people, so it does not meet the design criteria alleged by Interrobang.
A weapon is designed to deliver the round to the point of aim. The point of aim, and the intent of the shot, is entirely up to the shooter.
That being said, let me recite one of the rules of gun safety: "Only point a weapon at something you intend to destroy". You can defend yourself, or blow your own foot off -- it's up to you.
The problem I see with multicast (either alone or with some client/server mixed in) happens when the network connection to the client is almost full. If all your connection can handle is n players, what happens when the nth+1 player joins your multicast group? Or go back to the n players case (client/server)-- suppose you start an ftp connection to download the latest game patch. Or, what happens when a router at your isp decides to hiccup for a few seconds?
In all these cases, there's no way for multicast to gracefully handle the bandwidth demand that exceeds your network connection. If a server was the sole "owner" of your network connection, then it could gracefully degrade the information overload (say, by removing the player furthest from you, reducing the update rate for some number of players, or reducing the ftp transfer rate.)
I'd sure love to see a multicast that could handle bandwidth overloads, but I don't think it exists yet.
If anything, DIS is a fine example of what not to do on the real internet.
DIS is a broadcast technology; all entities receive complete information for all other entities. Even if an entity can't interact with another entity (say, the distance between two aircraft is 1000 miles), those two aircraft will know about each other. This is wasteful, both of bandwidth and processing power. Some work has been done using multicast instead of broadcast, but bandwidth is still wasted.
Further, there is no concept of a partial client update. If a client changes it appearance, it must send the client's position, velocity, acceleration, orientation, etc., along with its appearance. The size of an entity state PDU (the packet that announces a change of state) is 144 bytes (IIRC.)
The reason the US Armed Forces can get away with this inefficient use of bandwith is that they throw expensive technology at the problem. The Defense Simulation Internet (DSI) is a private backbone used for hosting large scale simulations. When you have a dedicated DS3, you tend to not care about the network. Also, the Army doesn't have to deal with modems.
On the real Internet, you have to build what Turbine and Origin have done with their MMPs: client/multi-server architectures. Each client receives information about relevant objects (clients, events) _only_. Each server handles a different part of the virual world.
Signed,
A former developer on the SIMNET project at BBN (the creators of DIS)
While Nader is spot-on with his diagnosis of the problem (corporations and pacs buying politicians, increasing their power), his solution doesn't diffuse or remove the power. Rather, the power is simply transferred to another group -- Trial Laywers and Union Bosses.
We'd be just as screwed with them as we were with the Corporations.
The Real Solution[tm], is to diffuse the power -- that is, keep the federal goverments restricted by the Ninth and Tenth amendmends.
Vote NADER!! The only alternative to today's "candidate(s)" brought to you by Big Brother corporation. Fight the power!!
Oh please -- Nader is the lapdog of the Trial Lawyers. Unless you happen to be a lawyer, you'd be just as screwed with Nader as President as, for example, Bush.
If you lose power on one rotor (the bearings sieze, or a blade is sheared off, etc.), you start rotating around the C.G. of (you+personal helicopter). Game over, man.
Lose both rotors, and you're a lawn dart. Game very over, man.
The NRA has always argued that "Guns don't kill people. People kill people." I say: "Guns make it a hell of a lot easier to kill people, and only stupid people carry guns."
And for your next trick, you'll join Mothers Against Cars Driven By Drunks. After all, a car makes it easy for a drunk to kill someone.
Next, you'll call cars with fuel capacity in excess of five gallons "assault cars" (can drive farther without refueling, so you can kill more pedestrians.) And, of course, no one "needs" a car with out a 65 mph speed governor.
The bizzare logic of the Anti-Self-Defense lobby always cracks me up.
The difference between "government" and all other groups, is that the government is directly responsible to the people, whereas all other groups are responsible only to themselves, a subset of the people.
If I don't like the way the government is acting, I have recourse, via my vote. If I don't like the way Microsoft is acting, I can go pound sand. My President/Prime Minister/Congrassman/whatever answers to me, but Gates et al answer to no-one.
It is alleged that the government is responsible to the people, but in reality, voting is the rough equivalent of farting at a hurricane. Those with the power and cash to lobby will always win.
Corporations are beholden to their shareholders, and, indirectly, their customers. If you don't like what a corporation is doing, you're free to not do business with them.
I can choose to do business with a certain corporation or not; however, I have no choice but to accept the mediocre and overpriced product that is government. I know where I have more choice.
Re:Joke (was Re:Intel Should Just Buy Transmeta)
on
IBM Wary of Crusoe?
·
· Score: 1
Right, and the board can say no to all offers, because they are the shareholders themselves, not anybody elses nominees.
True enough -- although I'd bet that some of Transmeta's venture capitalists are on the board, and the VCs will require cash (euphemistically called a "liquidation event") at some point. In the face of an offer for the company, the VC will weigh the value of cash now versus the value of (hopefully) more cash, or an IPO, later.
"We've already established that. We're just negotiating the price."
-- joke (sadly, I don't know the author's name)
Taking this to the topic at hand, at some price the idealism of Transmeta's stockholders will give in to cold hard cash. Indeed, the board of directors has a fiduciary duty to examine all offers -- they can't just ignore them because they want to.
Because Transmeta is privately held, Intel could not mount a hostile takeover. This is the only difference in takeovers between publicly and privately held companies.
NASA has been playing the gold-plated space exploration game for far too long. India, and the private launch companies, may show real guts and innovate in space.
I have worked on military flight simulators (A-10, F-16 in SIMNET). These were relatively cheap sims (the goal was team training, not procedures -- how to fly the aircraft -- training), so they had no motion platform. The rendering hardware was from the mid-1980s, so we had a 7 foot tall rack full of hardware that had one-tenth the performance of a TNT2.
So, we had a somewhat-believable, but clearly not real, world. In spite of this, several Air Force pilots had problems with simulator sickness -- to the point that one pilot couldn't fly more than a few minutes before he became nauseous and got a splitting headache.
It is interesting to note that there is a similar problem with pilots in real aircraft -- if you don't have the horizon in view, your inner ear will take over and try to tell you where up and down are. If you believe what your ears ("the seat of your pants") -- rather than your instruments -- tell you, you will end up very dead. It's called the Graveyard Spiral, and it is the favorite way for pilots to kill themselves. It's a fair bet that JFK Jr. killed himself in this way.
lbrlove asked: Has anyone considered the liability issues of this (I am seriously asking)
You can't avoid the lawyers, but you can make it more difficult for them to go after your assets. Strategies include:
1. each spacecraft is held by a separate C corporation. It is alleged that United Airlines hold each aircraft they have in a separate C Corporation. That way, if the aircraft goes down and people sue, the maximum amount the lawyers can get is the value of the aircraft (and insurance on that aircraft, if any.)
2. Offshore asset trusts. Some jurisdictions (e.g., Cook Islands) will not recognize foreign judgements, so the plantiffs will have to try the case again in the jurisdictions of the trust. Further, there are tight time limits to file.
3. somewhat convoluted multi-entity structures which give multiple levels of protection (e.g., a limited partnership with the general partner being a C corporation.)
Note that I Am Not A Lawyer -- I just pay one to cover my own butt with some of these methods. Your Mileage May Vary, yadda, yadda, yadda...
GNU in front of Linux is RMS blowing his own horn
on
Thus Spake Stallman
·
· Score: 1
Many people mistakenly believe that the reason RMS insists on using GNU/Linux is his seek for fame; that he wants people to give him credit for his work
IMHO, Mr Torvalds solved the hard problem -- the kernel. So, his brainchild -- Linux -- should have top billing. "Linux with GNU extensions" is correct; "GNU/Linux" is RMS self-aggrandizement, and is worthy of a marketing droid.
To me, it seems that if I created a missle, and it landed within 100m of my prefered target, it would take out my target.
That's true for just about every target except hardened missile silos.
Back in the Cold War days (and probably now, too), a missile was considered accurate enough to take out a missile silo when its accuracy was better than a 100 meter CEP (Circular Error Probable.)
Consider the case of a random adversary sending a cruise missile to get you. 100m accuracy gets the missile somewhere in your neighborhood; 1m accuracy means you are hit somewhere on your leg. If the missile is accurate, the adversary needs less of a warhead to produce the same effect.
As a pilot, I often get NOTAMS (Notices To Airmen) warning of GPS jamming tests. The tests occur over a region, and appear to use ground-based transmitters.
I'd guess that the US government will retain the ability to dither the signal.
I'm curious -- does anyone know the error rate for Airbus software?
Airbus has taken the position that the computer, rather than the pilots, can fly the airplane better. In fact, the pilots cannot override the computer.
As a pilot and software engineer, I find the Airbus philosophy fundamentally flawed. I can think of a few incidents on Boeing aircraft that would have turned into an "everyone dead" senario on a fly-by-wire Airbus (A320/330/340). You won't catch me dead flying on or acting as Pilot In Command of an aircraft exhibiting the Airbus philosophy.
dfay said: 1) Don't just go after video/audio! Many game developers use DirectX not for the wonderful (ahem, *sarcasm*) APIs in Direct3D, but for the DirectInput and DirectPlay APIs.
DirectPlay has been the bastard stepchild of the DirectX group since day one. Take a look at the past six Game Developer Magazines. Go to the post-mortems. You'll find about half of them say "DirectPlay did not work as advertised".
DP is nothing more than a lobby and send/receive. As we've seen before, simple send/receive is necessary, but not sufficient, for good network gaming (poor networking examples include Unreal and Half Life at release, X-Wing vs. Tie Fighter, etc.)
If you're serious about networking, you have two choices: roll your own on top of winsock/berkeley sockets (taking care not to repeat the mistakes of the past) or use a company that specializes in networking (RTIME or VR-1.)
Too bad you meant this bit ironically, because I agree with it literally. The day I see a widely-available computer prog (security or otherwise) made with the express purpose of killing people (and/or things), the way guns are, I'm going to move to Burkina Faso and live in a cave.
Funny, my Glock 19 pistol has put holes in paper targets, not people, so it does not meet the design criteria alleged by Interrobang.
A weapon is designed to deliver the round to the point of aim. The point of aim, and the intent of the shot, is entirely up to the shooter.
That being said, let me recite one of the rules of gun safety: "Only point a weapon at something you intend to destroy". You can defend yourself, or blow your own foot off -- it's up to you.
or, Marketing Instructions Per Second.
Forgive me, but I can't help remembering the movie line "People called Romanis (sp), they go the house."
I'd be happy to fly one of the ASAT missions.
The problem I see with multicast (either alone or with some client/server mixed in) happens when the network connection to the client is almost full. If all your connection can handle is n players, what happens when the nth+1 player joins your multicast group? Or go back to the n players case (client/server)-- suppose you start an ftp connection to download the latest game patch. Or, what happens when a router at your isp decides to hiccup for a few seconds?
In all these cases, there's no way for multicast to gracefully handle the bandwidth demand that exceeds your network connection. If a server was the sole "owner" of your network connection, then it could gracefully degrade the information overload (say, by removing the player furthest from you, reducing the update rate for some number of players, or reducing the ftp transfer rate.)
I'd sure love to see a multicast that could handle bandwidth overloads, but I don't think it exists yet.
Comments?
DIS is a broadcast technology; all entities receive complete information for all other entities. Even if an entity can't interact with another entity (say, the distance between two aircraft is 1000 miles), those two aircraft will know about each other. This is wasteful, both of bandwidth and processing power. Some work has been done using multicast instead of broadcast, but bandwidth is still wasted.
Further, there is no concept of a partial client update. If a client changes it appearance, it must send the client's position, velocity, acceleration, orientation, etc., along with its appearance. The size of an entity state PDU (the packet that announces a change of state) is 144 bytes (IIRC.)
The reason the US Armed Forces can get away with this inefficient use of bandwith is that they throw expensive technology at the problem. The Defense Simulation Internet (DSI) is a private backbone used for hosting large scale simulations. When you have a dedicated DS3, you tend to not care about the network. Also, the Army doesn't have to deal with modems.
On the real Internet, you have to build what Turbine and Origin have done with their MMPs: client/multi-server architectures. Each client receives information about relevant objects (clients, events) _only_. Each server handles a different part of the virual world.
Signed,
A former developer on the SIMNET project at BBN (the creators of DIS)
We'd be just as screwed with them as we were with the Corporations.
The Real Solution[tm], is to diffuse the power -- that is, keep the federal goverments restricted by the Ninth and Tenth amendmends.
Oh please -- Nader is the lapdog of the Trial Lawyers. Unless you happen to be a lawyer, you'd be just as screwed with Nader as President as, for example, Bush.
Lose both rotors, and you're a lawn dart. Game very over, man.
I think I'll wait on this one.
John (PP-ASEL, -G, -IA)
And for your next trick, you'll join Mothers Against Cars Driven By Drunks. After all, a car makes it easy for a drunk to kill someone.
Next, you'll call cars with fuel capacity in excess of five gallons "assault cars" (can drive farther without refueling, so you can kill more pedestrians.) And, of course, no one "needs" a car with out a 65 mph speed governor.
The bizzare logic of the Anti-Self-Defense lobby always cracks me up.
If I don't like the way the government is acting, I have recourse, via my vote. If I don't like the way Microsoft is acting, I can go pound sand. My President/Prime Minister/Congrassman/whatever answers to me, but Gates et al answer to no-one.
It is alleged that the government is responsible to the people, but in reality, voting is the rough equivalent of farting at a hurricane. Those with the power and cash to lobby will always win.
Corporations are beholden to their shareholders, and, indirectly, their customers. If you don't like what a corporation is doing, you're free to not do business with them.
I can choose to do business with a certain corporation or not; however, I have no choice but to accept the mediocre and overpriced product that is government. I know where I have more choice.
True enough -- although I'd bet that some of Transmeta's venture capitalists are on the board, and the VCs will require cash (euphemistically called a "liquidation event") at some point. In the face of an offer for the company, the VC will weigh the value of cash now versus the value of (hopefully) more cash, or an IPO, later.
"Sure"
"How about for $1?"
"What kind of woman do you think I am?"
"We've already established that. We're just negotiating the price."
-- joke (sadly, I don't know the author's name)
Taking this to the topic at hand, at some price the idealism of Transmeta's stockholders will give in to cold hard cash. Indeed, the board of directors has a fiduciary duty to examine all offers -- they can't just ignore them because they want to.
Because Transmeta is privately held, Intel could not mount a hostile takeover. This is the only difference in takeovers between publicly and privately held companies.
So, we had a somewhat-believable, but clearly not real, world. In spite of this, several Air Force pilots had problems with simulator sickness -- to the point that one pilot couldn't fly more than a few minutes before he became nauseous and got a splitting headache.
It is interesting to note that there is a similar problem with pilots in real aircraft -- if you don't have the horizon in view, your inner ear will take over and try to tell you where up and down are. If you believe what your ears ("the seat of your pants") -- rather than your instruments -- tell you, you will end up very dead. It's called the Graveyard Spiral, and it is the favorite way for pilots to kill themselves. It's a fair bet that JFK Jr. killed himself in this way.
Weird ... really weird.
1. with my cookies, 1-click enabled.
2. close browser, remove amazon cookies.
3. open browser, amazon askes me to log in; no 1-click
4. close browser, put amazon cookies back
5. open browser, amazon recognizes me, 1-click enabled, no password required.
Another reason to turn off 1-click. If you don't, you might find a weird set of books on your doorstep, and one maxed-out credit card.
You can't avoid the lawyers, but you can make it more difficult for them to go after your assets. Strategies include:
1. each spacecraft is held by a separate C corporation. It is alleged that United Airlines hold each aircraft they have in a separate C Corporation. That way, if the aircraft goes down and people sue, the maximum amount the lawyers can get is the value of the aircraft (and insurance on that aircraft, if any.)
2. Offshore asset trusts. Some jurisdictions (e.g., Cook Islands) will not recognize foreign judgements, so the plantiffs will have to try the case again in the jurisdictions of the trust. Further, there are tight time limits to file.
3. somewhat convoluted multi-entity structures which give multiple levels of protection (e.g., a limited partnership with the general partner being a C corporation.)
Note that I Am Not A Lawyer -- I just pay one to cover my own butt with some of these methods. Your Mileage May Vary, yadda, yadda, yadda ...
IMHO, Mr Torvalds solved the hard problem -- the kernel. So, his brainchild -- Linux -- should have top billing. "Linux with GNU extensions" is correct; "GNU/Linux" is RMS self-aggrandizement, and is worthy of a marketing droid.
That's true for just about every target except hardened missile silos.
Back in the Cold War days (and probably now, too), a missile was considered accurate enough to take out a missile silo when its accuracy was better than a 100 meter CEP (Circular Error Probable.)
Consider the case of a random adversary sending a cruise missile to get you. 100m accuracy gets the missile somewhere in your neighborhood; 1m accuracy means you are hit somewhere on your leg. If the missile is accurate, the adversary needs less of a warhead to produce the same effect.
I'd guess that the US government will retain the ability to dither the signal.
Airbus has taken the position that the computer, rather than the pilots, can fly the airplane better. In fact, the pilots cannot override the computer.
As a pilot and software engineer, I find the Airbus philosophy fundamentally flawed. I can think of a few incidents on Boeing aircraft that would have turned into an "everyone dead" senario on a fly-by-wire Airbus (A320/330/340). You won't catch me dead flying on or acting as Pilot In Command of an aircraft exhibiting the Airbus philosophy.
OK, French Persons, flame away. :-)
DirectPlay has been the bastard stepchild of the DirectX group since day one. Take a look at the past six Game Developer Magazines. Go to the post-mortems. You'll find about half of them say "DirectPlay did not work as advertised".
DP is nothing more than a lobby and send/receive. As we've seen before, simple send/receive is necessary, but not sufficient, for good network gaming (poor networking examples include Unreal and Half Life at release, X-Wing vs. Tie Fighter, etc.)
If you're serious about networking, you have two choices: roll your own on top of winsock/berkeley sockets (taking care not to repeat the mistakes of the past) or use a company that specializes in networking (RTIME or VR-1.)