Get over what? And, interesting debate? Sorry bud, but you answered one of his responses with a one sentence response. Then he did the same thing, and finally you did it again and I felt compelled to respond. That's about as interesting as watching paint dry and looks arrogant and annoying.
Jesus christ. He's saying that there's not enough population density to support the system that's in place because there just aren't enough people to make the bus transit efficient. Goddamn...
What. Seriously, you must be retarded. First, I said, to paraphrase, "sure the Amiga or whatever computer you like to masturbate over is better." I'll be the first to say it: The IBM PC is a total hack. I wrote assembler for it for 5 years, I know. So, your whole second paragraph is coming out of nowhere to me.
Second, you've got your timetables off a bit. The PC was HUGE for video games in the 486 era, and companies like Sierra were making PC games back in 1985.
And I don't care what you say, there is no way a production 16MHz chip could ever outrun a production 300MHz chip, period. Find me a 16MHz SGI machine, or whatever RISC hard-on chip you like from back then, and I'll take a 300MHz Pentium 2 and stomp it into shit. That said, I will reiterate that I KNOW the fucking 8086 platform is a hack, always has been. But what YOU seem to fail to acknowledge, is that none of that shit matters if it doesn't have market share. This converstaion started because someone was saying something about virtualized memory, and how since the XBox is a derivative of the 8086 arch, that it has a security hole, which was a response to why the 8086 used 12-bit memory segments (resulting in overlapping segments). I don't even understand how since the XBox should have been faulted into 32-bit protected mode unless MS is stupid, but I never looked at the exploit so I'll state that I must simply misunderstand what was meant by the grandparent. However, when it comes down to it, the IBM PC, despite its shittastic architecture, was the winner. And I was explaining to the grandparent, through a real-life programming example, why you couldn't have these things he was talking about. When it comes down to it, the name of the game was DOS/IBM and to say "well, I only want to work on Amigas" means you weren't employed/weren't as profitable, which in the end (sadly), is all that really matters.
Welcome to the wonderful world of the IBM PC. Those devices did not have nearly the market penetration that the IBM PC had/has, and therefore does it really matter that they were better? I'm sure they were great, but the long and short of it is that if you wanted a graphical program with any kind of speed and wanted to make a living from it, you were writing IBM PC code (and Motorola to a certain extent). Theory is nice but so is money.
If you wanted high performance graphics/sound you had no choice but to directly access memory. Using BIOS calls was out of the question for grahpics due to speed limitations, and you couldn't set up nonblocking sound buffers for any SB/clone larger than 64KB without DMA. Don't forget, that as an application developer in DOS, you had NO OS support for pretty much everything outside of file IO and little petty utilities like string output.
Let's say you wanted a custom timer back in 1990 to draw 320x200x8bit graphics at 25fps in DOS? There were no TimerProc(), KillTimer(), device contexts or any of that to help. You had to write your own service to draw the screen (probably using a back buffer, and then a bitblt over to 0xA000:0000), grab the BIOS timer IRQ interrupt vector (08h on IRQ line 0) from the IV table, reprogram the PIC to the frequency you wanted, do some simple math to determine how often to call the BIOS timer or else the system time went nuts until it was reread from the BIOS, which usually only happens during a reboot (MS Flight Simulator 4 did this), then when your application is finished, you have to undo everything you did or you lose all BIOS services related to the timer IRQ. DMA was worse because it always asked you where to look for its data and it could only fill like 64K at once, so you had to wtrite another interrupt for when it filled up 50% of the way so you could back fill the buffer with more sound without the analog output popping. Whew. Lots of ametuer game programming tips are coming back to me, it's sorta fun.
Something has to change, though. These methods that the MPAA are attempting to employ are both flawed and wasteful. There is no way to prevent people from ripping the content. If pirates have the machine that can turn a recording into analog A/V then the game is already lost. I think the solution is for the Movie studios to figure out how to offer their wares at a low enough cost to compete with pirated videos. I'm not smart enough to say that is a for-sure fix, but if I could legitimately buy a DVD without bullshit copy protection (that costs a ton of money to create) for like $5-$8, maybe $15 for a "super-deluxe" release, I'd be buying them like hot cakes. Right now I don't like buying DVDs because the fucking money I'm paying goes right into the pockets of these assholes. I wish I could pay the artist(s). I like doing that with CD's, but I realize it's far more difficult with movies due to the number of people involved.
The other issue is simply what is happening with PC games, which is that it's easier and usually safer (from a windows/dll hell standpoint) to rip copy protection right out of a game so it actually works. I've had many games refuse to install because their copy protection was either fucking up or just incompatible with my DVD drive. I had one game not let me install because my key was already in use and it wouldn't let me install it. I don't know about you, but when I spend $50 on a game and it doesn't play nice, I tend to get upset. I realize the only thing they lose is a customer, but I won't buy games like that any more. It has seriously turned me into a console gamer. Same with movies, if they make it difficult to play on your hardware, offer tons of restrictions, etc, their DRM will be shot to shit quickly by people who will rip and reburn their content in an unprotected manner.
A story, care of a very smart machinist:
At my grandmother's funeral earlier this year in November, I had the opportunity to meet one of my father's schoolmates who stayed in his hometown. He's 60 now, and to me, he seemed plenty sharp. He was talking to my dad and another one of my dad's friends about the machining trade, and how they're having problems getting help, because American kids aren't getting into machining. The conversation shifted to the foundries he gets his castings from. He explained that when he started getting castings from Chinese foundries a few decades ago, they sucked. Shitty casts, weak alloys, etc. But sure enough, their quality eventually equalled home grown American casts for a fraction of the cost. Guess who went out of work? Same went for his machinist tools. They went from bring short lived, cast pieces of shit to forged, superior tools.
The moral or the story? The Chinese aren't stupid, and anyone who thinks otherwise is a goddamn fool to believe they will not surpass us if we continue down the road where no one can earn a living unless their job is management of some kind. Exporting all labor will turn us back into a country of farmers, once places such as India and China figure out they no longer need us to do their own thing. I almost don't want to have children simply because if I can't escape the United States, I wouldn't want to raise a child into the situation that we both know is about to happen.
I used BSD style for pretty much ever, weening myself on Dr Dobbs journals and Charles Petzold in my youth and not UNIX code... At my current job where we use K&R it took me a while to get used to it, and I will still CR+LF+{ instead of {+CR+LF to this day. Annoying.
Paper mills aren't exactly environmentally friendly. If you've ever go near one, the godawful smell should alert you to the fact that they aren't too healthy for anything.
You specifically brought the initial poster you were referring to into your post, by saying: "[j]ust because YOU can't maintain objectivity", without him ever mentioning himself in any context. Sounds like a personal attack to me, and to probably anyone else, as well.
You know what the funniest thing is though, you're trying to support the use of logical fallacies.
Why are you so adamant about rejecting logic and reason? Why are you trying so hard to get away with relying on logical fallacies to make your point?
What the fuck are you talking about? Listen, I'm supporting the use of stereotypes as a valid method of predicting the intentions of a group we only know a few pieces of data about. With the data we know (most importantly, that Microsoft funded the study) we can state that there is a good likelihood that the study was tampered with, since that has happened with some frequency in the past.
Honestly, man, do you even understand how a logical fallacy is applied to an argument? Here's a lesson for you, for next time: a logical fallacy is simply an algorithm for detecting claims that can't support themselves, *not* for refuting them. That's a logical fallacy in and of itself. Therefore, just because someones argument contains a logical fallacy, doesn't by itself make their claim wrong. It only shows that their basis is not solid. A stereotype is a logical fallacy, "All people I have met of X type do Y, therefore all X people must do Y". It's a useful one though, because it allows us to predict with some accuracy what is happening when we don't have all the data. Exactly like our Microsoft example. Like I said in my earlier post, in a part that you must have skipped over, it is certainly possible that this study was done with the best intentions and Microsoft didn't get their hooks into it, but we all know that many of the studies that Microsoft has funded have been tampered with or have been what I would classify as a mistest in some manner. How is it an issue to have problems trusting this test?
And adamant? I'm sorry to burst your bubble, but one 4-line post on/. hardly defines "adamant", or "trying so hard."
And now, as far as your little "reading comprehension" jab, you can go fuck yourself. The good money says you wouldn't have the balls to argue like that in person, why do it over the internet? Feel the need to piss me off?
Just because it's a logical fallacy doesn't mean the claim isn't false. The fallacy can simply be used as a stereotype, and not as solid proof of determining truth. In this case, a study funded by a company that involves that company as a participant certainly could be unbiased, but in reality, where money talks, you can't say for certain. Why do you think you can't win a raffle where you have a family member employed by the raffling agency?
Regardless, you've committed a logical fallacy in your argument, and tried to bring personal attacks into it. Sorta ironic.
If you've ever seen a PS/2 to AT keyboard connector, that's pretty much exactly what one looks like. Thinkgeek has a hardware logger here:
http://www.thinkgeek.com/gadgets/electronic/5a05/ 128K or memory only though. There are more sophisticaed models that have more memory.
Wait, your goddamn ex got their college money... and isn't using it to help pay for their college? I've said it before, and I'll say it again, fuck marriage.
Off topic a bit, but if you have a MN12 Thunderbird of any year - unless you swapped an engine, computer and other assorted parts - you should have a SOHC 4.6L.
Having served many Saturday detentions, I often wondered about this myself. I am unsure as to the exact way they do this, I would assume they can restrict your entrance to the school the following Monday if you don't show up on Saturday, and that could just escalate into a suspension. For some reason, nobody, from what I can remember, ditched a Saturday and was there on Monday. I don't even remember if anyone really ditched a Saturday. I never did and now I wish I had so I would know what they could actually do about it.
Write it then and make a convincing argument why it should be included in the kernel. Or, another method is to fork your own kernel and implement the API you want to.
Re:Thats the whole point of the "puzzler"
on
Java Puzzlers
·
· Score: 1
Thanks. I have used 7-bit byte machines before, but obviously they are mostly relics now. I'm curious though, why the C drafters chose the word 'char' as the name for that type as opposed to 'byte'.
Re:Thats the whole point of the "puzzler"
on
Java Puzzlers
·
· Score: 1
Agreed. There is no Unicode value -1, so how does it make sense that you should be allowed to store that value in a type that represents a system where that value doesn't exist? I think it makes perfect sense, and I'm a pretty damn hardcore C fanboy. I never did like the unsigned-ness of char, I always thought there should have been a type 'byte' that represented an unsigned 8-bit value, and left char to deal with ASCII stuff.
Off topic, but is there any C that defines sizeof(char) != 1? I'd be interested to know.
There's this civil engineer who has a dream of a super-interstate-highway running from Los Angeles to Washington DC. After years of lobbying, he finally is awarded a contract to construct a perfectly-flat, perfectly-straight 16-lanes-each-direction highway from Los Angeles to Washington. So he employs all the best surveyors, to make sure the highway is a perfect straight-line from LA to DC and to make sure it stays perfectly flat. He subcontracts with only the best construction crews and uses only the best materials. He decides the existing tunnel-diggers and mountain removers just aren't up to the job, so he has bigger ones built, ones that can remove a half-mile swath of the Rockies in a week. They start in LA, and a few months later, they're overlooking the greater Washington metropolis.
Just one more hill to remove before they start work on the terminus and the merging into DC's outer belts. It's a small hill, really, but it's got to go. So they call in one of the smaller mountain-removers to remove it, and just as the machine's getting ready to erase the hill, a snake pops up out of the hill and screams "Wait!"
Obviously, this catches everyone by surprise, so they wait. The snake continues, "My name's Nate. Nate the Snake. You can't destroy this hill! You mustn't!"
"Why not?"
"Because there's a lever buried under this hill, and it's attached to a doomsday device. If the lever is tripped, it'll blow up the entire Earth!"
The engineer consults with his team. "What do you think?" "It's a talking snake." "Yeah, but do we believe him? Do we go around the hill, or do we plow it over?" "Do we believe him!? It's a talking snake! Who'd believe us?" In the end, they decide to err on the side of caution and build the highway around the hill. So, when they're finished, they've got a highway running from LA to DC that's perfectly-flat and perfectly-straight, except for this minor detour around a hill.
So they've got the ribbon-cutting ceremony, and as a perq for designing and building the highway, the engineer gets to be the very first person to use it. So he hops into his Lambourghini in LA at dawn and floors the pedal. 30, 40, 50mph. Shifts into 2d gear. 70, 90, 110mph. He keeps accelerating until the car just can's go any faster.
Shortly before sunset, he's approaching DC, and he remembers the hill. So he slows down to around 225mph to negotiate the slight turn. And he sees Nate crossing the road! He can't squish a talking snake! Especially one that kept them from blowing up the Earth. So he swerves to avoid hitting Nate, and plows into the hill at over 200mph. He trips the lever, and the Earth blows up.
Please, if you know what the halting problem is, you should know that you can set up a threshold of iterations before you determine the data is too deeply linked to be worth mining. Fractal image generation is similar, you choose an arbitrary number of iterations (usually related to the number of colors you can display) before giving up. Regardless, if I wrote a robot, I'd only go one request deep, since you can't get there or any further via a hard link.
Why not? It's not hard to parse out all paths the javascript can take from onClick events and follow the request object. It just means the robots will need to implement javascript if they want to get behind AJAX.
Web crawlers will eventually be released that figure out how to parse the javascript for XMLHttpRequest objects and examine the returned document. Besides, using AJAX for navigation is retarded. Using it for updating a shopping cart, on the other hand, is not. If you use AJAX to direct users around your site and break the back button, god help you.
Slashdot Mirror
Get over what? And, interesting debate? Sorry bud, but you answered one of his responses with a one sentence response. Then he did the same thing, and finally you did it again and I felt compelled to respond. That's about as interesting as watching paint dry and looks arrogant and annoying.
Jesus christ. He's saying that there's not enough population density to support the system that's in place because there just aren't enough people to make the bus transit efficient. Goddamn...
What. Seriously, you must be retarded. First, I said, to paraphrase, "sure the Amiga or whatever computer you like to masturbate over is better." I'll be the first to say it: The IBM PC is a total hack. I wrote assembler for it for 5 years, I know. So, your whole second paragraph is coming out of nowhere to me.
Second, you've got your timetables off a bit. The PC was HUGE for video games in the 486 era, and companies like Sierra were making PC games back in 1985.
And I don't care what you say, there is no way a production 16MHz chip could ever outrun a production 300MHz chip, period. Find me a 16MHz SGI machine, or whatever RISC hard-on chip you like from back then, and I'll take a 300MHz Pentium 2 and stomp it into shit. That said, I will reiterate that I KNOW the fucking 8086 platform is a hack, always has been. But what YOU seem to fail to acknowledge, is that none of that shit matters if it doesn't have market share. This converstaion started because someone was saying something about virtualized memory, and how since the XBox is a derivative of the 8086 arch, that it has a security hole, which was a response to why the 8086 used 12-bit memory segments (resulting in overlapping segments). I don't even understand how since the XBox should have been faulted into 32-bit protected mode unless MS is stupid, but I never looked at the exploit so I'll state that I must simply misunderstand what was meant by the grandparent. However, when it comes down to it, the IBM PC, despite its shittastic architecture, was the winner. And I was explaining to the grandparent, through a real-life programming example, why you couldn't have these things he was talking about. When it comes down to it, the name of the game was DOS/IBM and to say "well, I only want to work on Amigas" means you weren't employed/weren't as profitable, which in the end (sadly), is all that really matters.
Welcome to the wonderful world of the IBM PC. Those devices did not have nearly the market penetration that the IBM PC had/has, and therefore does it really matter that they were better? I'm sure they were great, but the long and short of it is that if you wanted a graphical program with any kind of speed and wanted to make a living from it, you were writing IBM PC code (and Motorola to a certain extent). Theory is nice but so is money.
If you wanted high performance graphics/sound you had no choice but to directly access memory. Using BIOS calls was out of the question for grahpics due to speed limitations, and you couldn't set up nonblocking sound buffers for any SB/clone larger than 64KB without DMA. Don't forget, that as an application developer in DOS, you had NO OS support for pretty much everything outside of file IO and little petty utilities like string output.
Let's say you wanted a custom timer back in 1990 to draw 320x200x8bit graphics at 25fps in DOS? There were no TimerProc(), KillTimer(), device contexts or any of that to help. You had to write your own service to draw the screen (probably using a back buffer, and then a bitblt over to 0xA000:0000), grab the BIOS timer IRQ interrupt vector (08h on IRQ line 0) from the IV table, reprogram the PIC to the frequency you wanted, do some simple math to determine how often to call the BIOS timer or else the system time went nuts until it was reread from the BIOS, which usually only happens during a reboot (MS Flight Simulator 4 did this), then when your application is finished, you have to undo everything you did or you lose all BIOS services related to the timer IRQ. DMA was worse because it always asked you where to look for its data and it could only fill like 64K at once, so you had to wtrite another interrupt for when it filled up 50% of the way so you could back fill the buffer with more sound without the analog output popping. Whew. Lots of ametuer game programming tips are coming back to me, it's sorta fun.
I do not think that quote has been verified by any reliable source.
Something has to change, though. These methods that the MPAA are attempting to employ are both flawed and wasteful. There is no way to prevent people from ripping the content. If pirates have the machine that can turn a recording into analog A/V then the game is already lost. I think the solution is for the Movie studios to figure out how to offer their wares at a low enough cost to compete with pirated videos. I'm not smart enough to say that is a for-sure fix, but if I could legitimately buy a DVD without bullshit copy protection (that costs a ton of money to create) for like $5-$8, maybe $15 for a "super-deluxe" release, I'd be buying them like hot cakes. Right now I don't like buying DVDs because the fucking money I'm paying goes right into the pockets of these assholes. I wish I could pay the artist(s). I like doing that with CD's, but I realize it's far more difficult with movies due to the number of people involved.
The other issue is simply what is happening with PC games, which is that it's easier and usually safer (from a windows/dll hell standpoint) to rip copy protection right out of a game so it actually works. I've had many games refuse to install because their copy protection was either fucking up or just incompatible with my DVD drive. I had one game not let me install because my key was already in use and it wouldn't let me install it. I don't know about you, but when I spend $50 on a game and it doesn't play nice, I tend to get upset. I realize the only thing they lose is a customer, but I won't buy games like that any more. It has seriously turned me into a console gamer. Same with movies, if they make it difficult to play on your hardware, offer tons of restrictions, etc, their DRM will be shot to shit quickly by people who will rip and reburn their content in an unprotected manner.
A story, care of a very smart machinist: At my grandmother's funeral earlier this year in November, I had the opportunity to meet one of my father's schoolmates who stayed in his hometown. He's 60 now, and to me, he seemed plenty sharp. He was talking to my dad and another one of my dad's friends about the machining trade, and how they're having problems getting help, because American kids aren't getting into machining. The conversation shifted to the foundries he gets his castings from. He explained that when he started getting castings from Chinese foundries a few decades ago, they sucked. Shitty casts, weak alloys, etc. But sure enough, their quality eventually equalled home grown American casts for a fraction of the cost. Guess who went out of work? Same went for his machinist tools. They went from bring short lived, cast pieces of shit to forged, superior tools.
The moral or the story? The Chinese aren't stupid, and anyone who thinks otherwise is a goddamn fool to believe they will not surpass us if we continue down the road where no one can earn a living unless their job is management of some kind. Exporting all labor will turn us back into a country of farmers, once places such as India and China figure out they no longer need us to do their own thing. I almost don't want to have children simply because if I can't escape the United States, I wouldn't want to raise a child into the situation that we both know is about to happen.
I used BSD style for pretty much ever, weening myself on Dr Dobbs journals and Charles Petzold in my youth and not UNIX code... At my current job where we use K&R it took me a while to get used to it, and I will still CR+LF+{ instead of {+CR+LF to this day. Annoying.
Paper mills aren't exactly environmentally friendly. If you've ever go near one, the godawful smell should alert you to the fact that they aren't too healthy for anything.
Honestly, man, do you even understand how a logical fallacy is applied to an argument? Here's a lesson for you, for next time: a logical fallacy is simply an algorithm for detecting claims that can't support themselves, *not* for refuting them. That's a logical fallacy in and of itself. Therefore, just because someones argument contains a logical fallacy, doesn't by itself make their claim wrong. It only shows that their basis is not solid. A stereotype is a logical fallacy, "All people I have met of X type do Y, therefore all X people must do Y". It's a useful one though, because it allows us to predict with some accuracy what is happening when we don't have all the data. Exactly like our Microsoft example. Like I said in my earlier post, in a part that you must have skipped over, it is certainly possible that this study was done with the best intentions and Microsoft didn't get their hooks into it, but we all know that many of the studies that Microsoft has funded have been tampered with or have been what I would classify as a mistest in some manner. How is it an issue to have problems trusting this test?
And adamant? I'm sorry to burst your bubble, but one 4-line post on
And now, as far as your little "reading comprehension" jab, you can go fuck yourself. The good money says you wouldn't have the balls to argue like that in person, why do it over the internet? Feel the need to piss me off?
Just because it's a logical fallacy doesn't mean the claim isn't false. The fallacy can simply be used as a stereotype, and not as solid proof of determining truth. In this case, a study funded by a company that involves that company as a participant certainly could be unbiased, but in reality, where money talks, you can't say for certain. Why do you think you can't win a raffle where you have a family member employed by the raffling agency?
Regardless, you've committed a logical fallacy in your argument, and tried to bring personal attacks into it. Sorta ironic.
If you've ever seen a PS/2 to AT keyboard connector, that's pretty much exactly what one looks like. Thinkgeek has a hardware logger here: http://www.thinkgeek.com/gadgets/electronic/5a05/ 128K or memory only though. There are more sophisticaed models that have more memory.
Wait, your goddamn ex got their college money... and isn't using it to help pay for their college? I've said it before, and I'll say it again, fuck marriage.
Off topic a bit, but if you have a MN12 Thunderbird of any year - unless you swapped an engine, computer and other assorted parts - you should have a SOHC 4.6L.
Did someone prove that gravity acts at the speed of light? I thought the answer wasn't known yet...
Too bad the server can't self replicate to handle the load...
Having served many Saturday detentions, I often wondered about this myself. I am unsure as to the exact way they do this, I would assume they can restrict your entrance to the school the following Monday if you don't show up on Saturday, and that could just escalate into a suspension. For some reason, nobody, from what I can remember, ditched a Saturday and was there on Monday. I don't even remember if anyone really ditched a Saturday. I never did and now I wish I had so I would know what they could actually do about it.
Write it then and make a convincing argument why it should be included in the kernel. Or, another method is to fork your own kernel and implement the API you want to.
Thanks. I have used 7-bit byte machines before, but obviously they are mostly relics now. I'm curious though, why the C drafters chose the word 'char' as the name for that type as opposed to 'byte'.
Agreed. There is no Unicode value -1, so how does it make sense that you should be allowed to store that value in a type that represents a system where that value doesn't exist? I think it makes perfect sense, and I'm a pretty damn hardcore C fanboy. I never did like the unsigned-ness of char, I always thought there should have been a type 'byte' that represented an unsigned 8-bit value, and left char to deal with ASCII stuff.
Off topic, but is there any C that defines sizeof(char) != 1? I'd be interested to know.
There's this civil engineer who has a dream of a super-interstate-highway running from Los Angeles to Washington DC. After years of lobbying, he finally is awarded a contract to construct a perfectly-flat, perfectly-straight 16-lanes-each-direction highway from Los Angeles to Washington. So he employs all the best surveyors, to make sure the highway is a perfect straight-line from LA to DC and to make sure it stays perfectly flat. He subcontracts with only the best construction crews and uses only the best materials. He decides the existing tunnel-diggers and mountain removers just aren't up to the job, so he has bigger ones built, ones that can remove a half-mile swath of the Rockies in a week. They start in LA, and a few months later, they're overlooking the greater Washington metropolis.
Just one more hill to remove before they start work on the terminus and the merging into DC's outer belts. It's a small hill, really, but it's got to go. So they call in one of the smaller mountain-removers to remove it, and just as the machine's getting ready to erase the hill, a snake pops up out of the hill and screams "Wait!"
Obviously, this catches everyone by surprise, so they wait. The snake continues, "My name's Nate. Nate the Snake. You can't destroy this hill! You mustn't!"
"Why not?"
"Because there's a lever buried under this hill, and it's attached to a doomsday device. If the lever is tripped, it'll blow up the entire Earth!"
The engineer consults with his team. "What do you think?" "It's a talking snake." "Yeah, but do we believe him? Do we go around the hill, or do we plow it over?" "Do we believe him!? It's a talking snake! Who'd believe us?" In the end, they decide to err on the side of caution and build the highway around the hill. So, when they're finished, they've got a highway running from LA to DC that's perfectly-flat and perfectly-straight, except for this minor detour around a hill.
So they've got the ribbon-cutting ceremony, and as a perq for designing and building the highway, the engineer gets to be the very first person to use it. So he hops into his Lambourghini in LA at dawn and floors the pedal. 30, 40, 50mph. Shifts into 2d gear. 70, 90, 110mph. He keeps accelerating until the car just can's go any faster.
Shortly before sunset, he's approaching DC, and he remembers the hill. So he slows down to around 225mph to negotiate the slight turn. And he sees Nate crossing the road! He can't squish a talking snake! Especially one that kept them from blowing up the Earth. So he swerves to avoid hitting Nate, and plows into the hill at over 200mph. He trips the lever, and the Earth blows up.
The moral of the story...
Better Nate than lever.
Please, if you know what the halting problem is, you should know that you can set up a threshold of iterations before you determine the data is too deeply linked to be worth mining. Fractal image generation is similar, you choose an arbitrary number of iterations (usually related to the number of colors you can display) before giving up. Regardless, if I wrote a robot, I'd only go one request deep, since you can't get there or any further via a hard link.
Why not? It's not hard to parse out all paths the javascript can take from onClick events and follow the request object. It just means the robots will need to implement javascript if they want to get behind AJAX.
Web crawlers will eventually be released that figure out how to parse the javascript for XMLHttpRequest objects and examine the returned document. Besides, using AJAX for navigation is retarded. Using it for updating a shopping cart, on the other hand, is not. If you use AJAX to direct users around your site and break the back button, god help you.