now a couple of million dollars on the table says Novell's lawyers think they can (otherwise it's just a bit circlejerk, and hey, that's likely too).
Or maybe Novell just wanted a couple hundred million George Washingtons.
Because one of four things I can think of is going on:
1. Novell just thought that the partnership would be good (remember, something like $300 mil) completely unrelated to IP issues 2. Novell's lawyers think there ARE IP issues with Linux, and they plan on dropping distribution of Linux when it hits the fan 3. Novell's lawyers think there are IP issues with Linux, and they are wagering that either the GPL will be held to be unenforcible, that they won't be exposed to large judgments, or that the Linux developers will be too poor/busy/not care enough to sue 4. Novell's lawyers can't read
Because if it turns out that there ARE IP issues with Linux, Novell can't distribute it either. Which means that they must cease or face a possible lawsuit from everyone who has code in the kernel.
Occam's razor, in my mind, is pretty clear that #1 seems the most probable.
But there are also a bunch of angry mobs that occur when their "greivance" is that their team didn't win the Superbowl. (Or heck, that their team did win the Superbowl.)
Not all mob scenes are due to an actual grievance. In fact, I have a hunch that most violent ones aren't...
Useful. But surely the == operator is less efficient.
True, but if you need it to be more efficient than you can define it yourself. (You might have to do != too; I don't know if != is defined in terms of the C# automatically gives you a += operator when you overload +...
That seems backwards from the way you'd do it in C++. One idiomatic way to implement + is copy one of the arguments, += the others to the copy, then return it. The reason you do it that way instead of the other way around is because it avoids the creation of temporaries.
Though maybe the GC of C# would make that much less of an issue...
Sounds like Lisp to me.... It's a mystery to me why anyone would voluntarily program in a language that makes the things that should be easy difficult, and makes things like segfaults, which should be pretty darn hard to do, easy enough to accomplish by accident. Yet so many people do it.
To at least a large extent, probably inertia. People know imperative languages, they know "traditional" ALGOL derivatives, so that's what they use, either because they don't see the grass on the other side is greener or because they decide that they don't have time to make it to the bridge and across it before running out of food, because while there is grass where you are, there's nothing until the other side of the bridge if you go for it. (Okay, that analogy got really unwieldy quickly, but I think there's something to it.)
For instance, the ML compiler is written in ML. Do I have any flipping clue at all how you would write a piece of software anywhere near as complex as the ML compiler with ML? Heck no! At least not well at all. If I were to attempt it, I'd end up with ! and ref everywhere because I don't know how to do it any better. I don't know how you would really take advantage of the fact that ML provides first-class functions or stuff like that. Sure, I can write little toy programs in ML, but nothing of that scale. Lisp is even worse; I don't have ANY clue how to go about it there. But could I write an ML compiler in C++? I think I could. It'd take me a while, and I couldn't guarantee performance, but I think I would wind up with something reasonable.
(BTW, in case you can't really tell, I'm sorta in the second group in the analogy. I think Lisp holds at least a lot of promise for helping development (though I'm not totally convinced), it's just that I don't have time to learn not only a new language but in some sense a new paradigm right now.)
C++ is like a sharp scalpel. Yes you can hurt yourself if you're unskilled, inexperienced or sloppy
"C++ gives you enough rope to shoot yourself in the foot"
Java and C# are like those scissors with rounded ends for kids. Totally inefficent but safe for beginners.
I'm not convinced of the "totally inefficient" bit. I think you'd be pressing it to do time-critical systems (indeed, current GC is more or less incompatible with realtime systems), OSs, etc., but I'm not convinced that they're not just fine for applications. This especially applies to C#, because C# GUIs are actually responsive. (Swing and to a lesser extent SWT lag a little.)
But there's a reason why surgeons don't use plastic scissors.
There's also a reason carpenters don't use scalpels. It's because different tools are good for different jobs.
Oh holy hell... stupid less than signs. You think the forum software could see that "hey, the user wrote < but didn't follow it with a legal html tag... he probably meant that < to be literal," but no, that would be too easy. Oh well, that's what the preview button's for. This is what I meant to say:
There somewhat is. There's a namespace inside std that includes templated functions for ==, !=, >, >=, and <=. (I forget what you have to do to get it; a quick Google search doesn't come up with it, and the book that I know talks about it is in my office.) If you define operator<, it will be able to infer any that you don't provide. (>= is !<; != is < || >; == is !< && !>; etc.)
It's not very fine-grained unfortunately; like you can't specify "I want to provide the other operators for this type but not this type" I think.
If you actually think that's true, read The Design and Evolution of C++. A very interesting account of the development of the language. If he forgets something, he can go back and read his book.
Only problem with it is it leaves off at 1993 or so, just as the ISO process was getting to its feet.
But the point is that Novell loses. Unless they're a lot dimmer over there than I realize, they're not going to deliberately do something that will give them as much or more of a problem than what they're supposedly being paid to solve.
And what happens if they did? Then the OO people just remove the offending code and we're no worse off than we would be if everyone just ignored Novell like people are saying. If there were any losses due to patent fights, it's probable that the OO people would be able to recover them from Novell, plus punitive damages.
Everybody who uses OO that does not come from Novell will be open to a lawsuit from MS.
And everyone who WROTE OO will SUE Novell for copyright infringement! If Novell can't or won't distribute the licenses for the patents that it uses (e.g. because they belong to MS), the GPL doesn't give them the right to distribute the thing at all!
This is what people seem to be overlooking. Novell can't introduce patented code into Linux or anything else because if they do THEY CAN'T DISTRIBUTE IT and open themselves up to lawsuits from every author who has code in whatever they are contaminating.
There's reason to keep an eye on Novell, but I think that people are blowing this deal WAY out of proportion.
I don't think that's true. If that dude can be clearly identified he has the right to that picture and you are not allowed to do anything with it without his consent.
For that, let's add a twist: what if you just shot a general picture of the entire field because there was a great sunset? Although you can still identify some people individually, do you still need a release? It's been successfully argued that if the point of the photo is not a specific person, but a broader scene, then the a release is not necessary. These cases usually involve innocuous items like postcards and other simple consumable products that are not tied to a promotion of an idea.
It's not as clear-cut as I thought, but there IS still a lot of room there.
Someone putting your picture in an exhibition without you having agreed to that is clearly unlawful.
Not in the US:
The main point to this section is "art," and in that context, the courts almost always rule in favor of no release..... Art exhibits--and indeed, the sales of photos as artwork--are exempt from requiring a release from subjects that happen to be portrayed. Courts have decided repeatedly on this matter, including those situations where other potential conflicts may be intertwined.
This one *IS* pretty clear-cut, but there are still occasionally exceptions.
It's the same with news reports
Now this is just ridiculous. Imagine if that were the case in the US... all the news outlets that wanted to publish pictures of Abramoff and would have had to get their permission? And if they refuse to give it they can't publish it? Surely this can't be the case in *any* free country. (Indeed, I would say that's a tautology, and that if you can't do that you *can't* be a free country.)
Can't really believe it to be much different in the US. But I may be wrong.
My impression is that Europe as a whole is somewhat more restricting of these things. It somewhat smells like artist's "moral rights" too, which we don't really have in the US.
Re:Outdoors = no reasonable expectation of privacy
on
Windows Live and Privacy
·
· Score: 2, Interesting
That being said, I think it's a gimmicky piece of crap, and honestly I can't foresee it being useful for anything Google Earth can't already do better.
Really? Can Google Earth show you a picture of the neighborhood you're thinking about moving into? Can it give you a picture at each of the intersections along the route it finds from A to B?
So one day you forget to do up your fly and go out to the shops. Some joker takes a picture of your pink boxers and it ends up on the front page of digg. Is this kind of thing fair?
Maybe not, but that doesn't mean it's wrong or illegal.
As long as you keep the photographs you made of people without asking their permission to do anything with them under your bed you're fine. Anything else, be it for someone's profit or not, would be unlawful.
I can't speak to other country's laws, but aparently YOU can't speak to the US's.
There's actually a relatively small number of things that you can't do with a photo. Here's the general rundown:
1. Are they recognizable by most people: if not, then you're pretty much in the clear (this doesn't mean do most people know you; but if you were to see the person after seeing the photograph would you be able to say "hey, that's the person from the photo")
2. Are they a focal point of the photograph, or just incidental? If they're just incidental, you're pretty much in the clear (e.g. if you take a picture of a beach, some random dude who happens to be in the photograph usually can't do anything)
3. Even if you don't pass the first two points, there's still a fairly wide variety of things that you can do with photos. For instance, if they are being used for news purposes, you can't do anything. If they're being used in an art exhibit, you probably can't do anything.
OpenBSD also has two other features that work well with this randomisation. They insert a random-sized gap between stack frames[1] so that the attacker doesn't know exactly where the return address will be, and a canary value that is written next to the return address, and checked for modification before it is returned.
How can OpenBSD do this? Isn't that in the purview of the program you're running and thus the compiler? Or are you saying that OpenBSD comes with GCC patches or something that causes it to do this?
Yes, because it's MS's responsibility to go in and correct third party software.
Sheesh, MS is in the right with this one, seriously. Other OSs have had address space randomization for a while now, including OpenBSD. If they hadn't introduced it, even if vulnerabilities were in 3rd party software Windows still takes the rap.
Is it too much for the/. groupthink for MS to be right at least once?
More likely, at installation they randomly rebase all the system DLLs so that your machine has a custom selection of default base addresses. You wouldn't get a random location each time a DLL is loaded but an attack would require information only available by running code on your machine
The DLLs are rebased at reboot. (Probably actually the first time they are loaded. DLLs specify in a section of the PE header if they should be randomized or not.)
First, one assumes that this randomizer is turned off during debugging
Why should we assume this?
Second, how random is the randomizer. Are there specific locations that are going to recur?
Yes. It's actually not very random at all. There is a small number of offsets that it chooses from. (Maybe as low as 16. I forget the exact number.)
Is it possible that even if an attack can't effect 100K machines, it might effect 5K?
Yep. But this can have a big effect if you're talking something like an internet worm. Dropping the number of machines that you can infect in any given round by 16 would *greatly* slow the spread of such worms, allowing more time for countermeasures to go up.
is the range small enough so that attacks can at least cause a machine to crash, if not execute arbitrary code
Yep. But you're gonna get that with any OS protection of buffer overflows I fear. The only reasonable thing to do if someone smashes the return address and the OS sees it is to kill the process.
(And actually, in that sense, the answer is no, because it would just kill the process.)
It shows that there is still some actual talent at MS
I'm pretty sure address space randomization has been implemented in other OSs for years; it's not an MS invention.
It goes a long way to eliminating what are called return-to-library (sometimes return-to-glibc) attacks. Wikipedia has an article, but I'm too lazy to link.
Basically there are three things that you can do if there is a buffer overflow exploit waiting (not all are always possible):
1. Overwrite other security-sensitive data, but let the control flow remain unaffected (read the paper "Non-Control-Data Attacks Are Realistic Threats" for a very interesting treatment of stuff along this line) 2. Write a bunch of data that are malicious machine instructions, writing until you overwrite the return address on the stack with the address of your code 3. Write a bunch of data into the buffer, overwriting the return address with existing code
It's #3 here that this is supposed to prevent. Right now, an attacker can sometimes say "if I write this data to the stack, and write the address of (say) exec, I can compromise the system." What address space randomization does is to make it so that the address of exec is changed. Previously, something like glibc would be linked in at the same address each run.
I would suspect that making a device which intereferes with first responders (or any gov't) frequency would count as harmful interference and be illegal under most coutnries laws.
Now now, if by "manually" you mean like actually lift it, the situation isn't quite that dire yet. There's still probably the option of pushing a little button inside the fire hall.
(Seriously, what did the GP think fire dept's did before wireless garage door openers were invented? Or for that matter when the power goes out? Do you think they have a backup generator running to open the doors?)
Is is just a coincidence that both Nvidia and ATI were each awarded Xbox contracts (Nvidia = Xbox, ATI = Xbox 360)? Perhaps there was some behind the scenes deals to thwart the development of FOSS graphics drivers.
Or maybe it's just as you say and those are the top two graphics companies and MS wanted the best.
It's like the "Never ascribe to malice, that which can be explained by incompetence" corallary to Occam's razor except more like "never ascribe to malice, that which can be explained by good business decisions."
Nothing new, other than maybe they are saying we can now confirm it was concrete with modern analysis techniques.
Which is PLENTY of reason for news, even if the theory was widely believed.
I mean, there's a theory that the Sphinx was built about 10,000 years earlier than was previously thought, by an entirely different civilization. It's not widely believed, but the guy does have some evidence.
As for the current theory, I doubt *IT* was widely believed either. I've watched a few shows on Egypt, and never heard of it before now.
Slashdot Mirror
now a couple of million dollars on the table says Novell's lawyers think they can (otherwise it's just a bit circlejerk, and hey, that's likely too).
Or maybe Novell just wanted a couple hundred million George Washingtons.
Because one of four things I can think of is going on:
1. Novell just thought that the partnership would be good (remember, something like $300 mil) completely unrelated to IP issues
2. Novell's lawyers think there ARE IP issues with Linux, and they plan on dropping distribution of Linux when it hits the fan
3. Novell's lawyers think there are IP issues with Linux, and they are wagering that either the GPL will be held to be unenforcible, that they won't be exposed to large judgments, or that the Linux developers will be too poor/busy/not care enough to sue
4. Novell's lawyers can't read
Because if it turns out that there ARE IP issues with Linux, Novell can't distribute it either. Which means that they must cease or face a possible lawsuit from everyone who has code in the kernel.
Occam's razor, in my mind, is pretty clear that #1 seems the most probable.
But there are also a bunch of angry mobs that occur when their "greivance" is that their team didn't win the Superbowl. (Or heck, that their team did win the Superbowl.)
Not all mob scenes are due to an actual grievance. In fact, I have a hunch that most violent ones aren't...
Useful. But surely the == operator is less efficient.
...
True, but if you need it to be more efficient than you can define it yourself. (You might have to do != too; I don't know if != is defined in terms of the C# automatically gives you a += operator when you overload +
That seems backwards from the way you'd do it in C++. One idiomatic way to implement + is copy one of the arguments, += the others to the copy, then return it. The reason you do it that way instead of the other way around is because it avoids the creation of temporaries.
Though maybe the GC of C# would make that much less of an issue...
Sounds like Lisp to me.... It's a mystery to me why anyone would voluntarily program in a language that makes the things that should be easy difficult, and makes things like segfaults, which should be pretty darn hard to do, easy enough to accomplish by accident. Yet so many people do it.
To at least a large extent, probably inertia. People know imperative languages, they know "traditional" ALGOL derivatives, so that's what they use, either because they don't see the grass on the other side is greener or because they decide that they don't have time to make it to the bridge and across it before running out of food, because while there is grass where you are, there's nothing until the other side of the bridge if you go for it. (Okay, that analogy got really unwieldy quickly, but I think there's something to it.)
For instance, the ML compiler is written in ML. Do I have any flipping clue at all how you would write a piece of software anywhere near as complex as the ML compiler with ML? Heck no! At least not well at all. If I were to attempt it, I'd end up with ! and ref everywhere because I don't know how to do it any better. I don't know how you would really take advantage of the fact that ML provides first-class functions or stuff like that. Sure, I can write little toy programs in ML, but nothing of that scale. Lisp is even worse; I don't have ANY clue how to go about it there. But could I write an ML compiler in C++? I think I could. It'd take me a while, and I couldn't guarantee performance, but I think I would wind up with something reasonable.
(BTW, in case you can't really tell, I'm sorta in the second group in the analogy. I think Lisp holds at least a lot of promise for helping development (though I'm not totally convinced), it's just that I don't have time to learn not only a new language but in some sense a new paradigm right now.)
C++ is like a sharp scalpel. Yes you can hurt yourself if you're unskilled, inexperienced or sloppy
"C++ gives you enough rope to shoot yourself in the foot"
Java and C# are like those scissors with rounded ends for kids. Totally inefficent but safe for beginners.
I'm not convinced of the "totally inefficient" bit. I think you'd be pressing it to do time-critical systems (indeed, current GC is more or less incompatible with realtime systems), OSs, etc., but I'm not convinced that they're not just fine for applications. This especially applies to C#, because C# GUIs are actually responsive. (Swing and to a lesser extent SWT lag a little.)
But there's a reason why surgeons don't use plastic scissors.
There's also a reason carpenters don't use scalpels. It's because different tools are good for different jobs.
Oh holy hell... stupid less than signs. You think the forum software could see that "hey, the user wrote < but didn't follow it with a legal html tag... he probably meant that < to be literal," but no, that would be too easy. Oh well, that's what the preview button's for. This is what I meant to say:
There somewhat is. There's a namespace inside std that includes templated functions for ==, !=, >, >=, and <=. (I forget what you have to do to get it; a quick Google search doesn't come up with it, and the book that I know talks about it is in my office.) If you define operator<, it will be able to infer any that you don't provide. (>= is !<; != is < || >; == is !< && !>; etc.)
It's not very fine-grained unfortunately; like you can't specify "I want to provide the other operators for this type but not this type" I think.
There somewhat is. There's a namespace inside std that includes templated functions for ==, !=, >, >=, and = is !; == is !; etc.)
It's not very fine-grained unfortunately; like you can't specify "I want to provide the other operators for this type but not this type" I think.
If you actually think that's true, read The Design and Evolution of C++. A very interesting account of the development of the language. If he forgets something, he can go back and read his book.
Only problem with it is it leaves off at 1993 or so, just as the ISO process was getting to its feet.
But the point is that Novell loses. Unless they're a lot dimmer over there than I realize, they're not going to deliberately do something that will give them as much or more of a problem than what they're supposedly being paid to solve.
And what happens if they did? Then the OO people just remove the offending code and we're no worse off than we would be if everyone just ignored Novell like people are saying. If there were any losses due to patent fights, it's probable that the OO people would be able to recover them from Novell, plus punitive damages.
Everybody who uses OO that does not come from Novell will be open to a lawsuit from MS.
And everyone who WROTE OO will SUE Novell for copyright infringement! If Novell can't or won't distribute the licenses for the patents that it uses (e.g. because they belong to MS), the GPL doesn't give them the right to distribute the thing at all!
This is what people seem to be overlooking. Novell can't introduce patented code into Linux or anything else because if they do THEY CAN'T DISTRIBUTE IT and open themselves up to lawsuits from every author who has code in whatever they are contaminating.
There's reason to keep an eye on Novell, but I think that people are blowing this deal WAY out of proportion.
It is true in the US. From here:
It's not as clear-cut as I thought, but there IS still a lot of room there.
Someone putting your picture in an exhibition without you having agreed to that is clearly unlawful.
Not in the US:
This one *IS* pretty clear-cut, but there are still occasionally exceptions.
It's the same with news reports
Now this is just ridiculous. Imagine if that were the case in the US... all the news outlets that wanted to publish pictures of Abramoff and would have had to get their permission? And if they refuse to give it they can't publish it? Surely this can't be the case in *any* free country. (Indeed, I would say that's a tautology, and that if you can't do that you *can't* be a free country.)
Can't really believe it to be much different in the US. But I may be wrong.
My impression is that Europe as a whole is somewhat more restricting of these things. It somewhat smells like artist's "moral rights" too, which we don't really have in the US.
That being said, I think it's a gimmicky piece of crap, and honestly I can't foresee it being useful for anything Google Earth can't already do better.
Really? Can Google Earth show you a picture of the neighborhood you're thinking about moving into? Can it give you a picture at each of the intersections along the route it finds from A to B?
So one day you forget to do up your fly and go out to the shops. Some joker takes a picture of your pink boxers and it ends up on the front page of digg. Is this kind of thing fair?
Maybe not, but that doesn't mean it's wrong or illegal.
As long as you keep the photographs you made of people without asking their permission to do anything with them under your bed you're fine. Anything else, be it for someone's profit or not, would be unlawful.
I can't speak to other country's laws, but aparently YOU can't speak to the US's.
There's actually a relatively small number of things that you can't do with a photo. Here's the general rundown:
1. Are they recognizable by most people: if not, then you're pretty much in the clear (this doesn't mean do most people know you; but if you were to see the person after seeing the photograph would you be able to say "hey, that's the person from the photo")
2. Are they a focal point of the photograph, or just incidental? If they're just incidental, you're pretty much in the clear (e.g. if you take a picture of a beach, some random dude who happens to be in the photograph usually can't do anything)
3. Even if you don't pass the first two points, there's still a fairly wide variety of things that you can do with photos. For instance, if they are being used for news purposes, you can't do anything. If they're being used in an art exhibit, you probably can't do anything.
Okay, that makes sense. Thanks for the answer.
OpenBSD also has two other features that work well with this randomisation. They insert a random-sized gap between stack frames[1] so that the attacker doesn't know exactly where the return address will be, and a canary value that is written next to the return address, and checked for modification before it is returned.
How can OpenBSD do this? Isn't that in the purview of the program you're running and thus the compiler? Or are you saying that OpenBSD comes with GCC patches or something that causes it to do this?
Yes, because it's MS's responsibility to go in and correct third party software.
/. groupthink for MS to be right at least once?
Sheesh, MS is in the right with this one, seriously. Other OSs have had address space randomization for a while now, including OpenBSD. If they hadn't introduced it, even if vulnerabilities were in 3rd party software Windows still takes the rap.
Is it too much for the
More likely, at installation they randomly rebase all the system DLLs so that your machine has a custom selection of default base addresses. You wouldn't get a random location each time a DLL is loaded but an attack would require information only available by running code on your machine
The DLLs are rebased at reboot. (Probably actually the first time they are loaded. DLLs specify in a section of the PE header if they should be randomized or not.)
First, one assumes that this randomizer is turned off during debugging
Why should we assume this?
Second, how random is the randomizer. Are there specific locations that are going to recur?
Yes. It's actually not very random at all. There is a small number of offsets that it chooses from. (Maybe as low as 16. I forget the exact number.)
Is it possible that even if an attack can't effect 100K machines, it might effect 5K?
Yep. But this can have a big effect if you're talking something like an internet worm. Dropping the number of machines that you can infect in any given round by 16 would *greatly* slow the spread of such worms, allowing more time for countermeasures to go up.
is the range small enough so that attacks can at least cause a machine to crash, if not execute arbitrary code
Yep. But you're gonna get that with any OS protection of buffer overflows I fear. The only reasonable thing to do if someone smashes the return address and the OS sees it is to kill the process.
(And actually, in that sense, the answer is no, because it would just kill the process.)
It shows that there is still some actual talent at MS
I'm pretty sure address space randomization has been implemented in other OSs for years; it's not an MS invention.
Write a bunch of data into the buffer, overwriting the return address with existing code
Sorry, this is a bit imprecise. I didn't say quite what I meant. This should say "overwriting the return address with the address of existing code"
It goes a long way to eliminating what are called return-to-library (sometimes return-to-glibc) attacks. Wikipedia has an article, but I'm too lazy to link.
Basically there are three things that you can do if there is a buffer overflow exploit waiting (not all are always possible):
1. Overwrite other security-sensitive data, but let the control flow remain unaffected (read the paper "Non-Control-Data Attacks Are Realistic Threats" for a very interesting treatment of stuff along this line)
2. Write a bunch of data that are malicious machine instructions, writing until you overwrite the return address on the stack with the address of your code
3. Write a bunch of data into the buffer, overwriting the return address with existing code
It's #3 here that this is supposed to prevent. Right now, an attacker can sometimes say "if I write this data to the stack, and write the address of (say) exec, I can compromise the system." What address space randomization does is to make it so that the address of exec is changed. Previously, something like glibc would be linked in at the same address each run.
I would suspect that making a device which intereferes with first responders (or any gov't) frequency would count as harmful interference and be illegal under most coutnries laws.
Who said they did?
"Interference" is not a reflexive relation. If you put a WAP next to six microwaves, you're going to have problems with your wi-fi network. But your WAP isn't going to overcook your food.
This story is about the USAF's use blocking garage door openers, not the other way around.
The questions is *when* did the US gov't acquire this frequency?
Another poster says 1934.
Now now, if by "manually" you mean like actually lift it, the situation isn't quite that dire yet. There's still probably the option of pushing a little button inside the fire hall.
(Seriously, what did the GP think fire dept's did before wireless garage door openers were invented? Or for that matter when the power goes out? Do you think they have a backup generator running to open the doors?)
Is is just a coincidence that both Nvidia and ATI were each awarded Xbox contracts (Nvidia = Xbox, ATI = Xbox 360)? Perhaps there was some behind the scenes deals to thwart the development of FOSS graphics drivers.
Or maybe it's just as you say and those are the top two graphics companies and MS wanted the best.
It's like the "Never ascribe to malice, that which can be explained by incompetence" corallary to Occam's razor except more like "never ascribe to malice, that which can be explained by good business decisions."
Nothing new, other than maybe they are saying we can now confirm it was concrete with modern analysis techniques.
Which is PLENTY of reason for news, even if the theory was widely believed.
I mean, there's a theory that the Sphinx was built about 10,000 years earlier than was previously thought, by an entirely different civilization. It's not widely believed, but the guy does have some evidence.
As for the current theory, I doubt *IT* was widely believed either. I've watched a few shows on Egypt, and never heard of it before now.