Journals usually typeset the paper. To do this they start by importing everything into Word. Why - it is a standard. Absolute crap for typesetting, but that is the way it goes. Some journals are getting adept at using TeX and PDF.
In math and physics, all the real journals use TeX exclusively. This is because little to no typesetting is needed. A journal supplied macro/stylesheet is used that defines all of the allowable commands. You only have to put thought into getting the layout correct.
Ofcourse TeX is also used exclusively because nobody has time to click together 50 fucking equations using some god aweful gui.
For the record. Linux does it both ways; or at least it did. Everyone just seems to prefer one of those ways.
OSX does it both ways too. And because people have complained, they have/are changing the default from a file to a partition.
This is not a linuxcentric phenonina, it is just sensible.
But perhaps you are right. I am aware that the file is contiguous. But there would also have to be some magic that allows the kernel to completely ignore the filesystem when paging (without breaking anything). And the OS would have to be sensible enough to locate the file at the innermost portion of the harddrive.
I'll give you some objective differences. (I might be off on a few)
Does WNT run on over 15 architectures? It was intended to be crossplatform, but it never really did well with that. It's best other platform was the Alpha, and even that was utter shite. I don't think it ever supported 64bits.
Has WNT ever supported 64bits?
Does WNT support pluggable modules?
Does in run on mainframes?
Does anyone other than Microsoft and Microsoft funded persons cluster it? How well does it work with MOSIX?
Does it have a O(1) scheduler?
Does it have a fast Journaling file system that stores files in b-tree's? Does it support dozens of file systems?
Can I boot up, read my root/bin/lib partitiond to a ramdisk, and run completely from RAM?
Can I run with all of my partitions encrypted?
Did NSA make a security-enhanced version? What about when they left those registry keys?
Did they ever move the graphics out of kernel space? Ouch!
Does it have superfast zerocopy for NFS, webservers, and other net stuff?
Math exists without need for our physical laws. There are plenty of branches of mathematics with little to no relation to our real world.
The math that does apply to our real world in the form of physical laws, are only mathematical models. None of physics is exact, and it has never been.
For example: Pi cannot be found exactly in this universe. Pi does not exist in our world embeded in spheres. You cannot construct a perfect sphere in the real world because once you hit the plank scale you can no longer have perfect classical smoothness. Pi does not exist in our world embeded in oscillators, for much the same reason.
Pi only exists in math, in idea. The fact that we can use pi and math to model the universe to some degree of approximation is only a testiment to how ordered and logical the universe is.
So though you may have not realized it, the poster had an important point. Pi and math would be the same in all universes, because Pi doesn't exist in the universe any how, they only exist in our minds.
That is using the power series for the arctangent right. A much faster and better method is to use the rational approximation of arctangent, which yields the continued fraction for pi. Almost as easy, but much better.
There has been too much misinformation thrown out on both sides but you are dead wrong and the fact that you keep getting modded up shows that Slashdot is not the place for informed discussion.
1) Nothing in GR goes faster than light. Light travels a geodesic at maximal speed. You can spin around in your backyard until you throw up, but that will not move anything from point A to point B in less time than light can do it.
2) I really don't know what the hell you people are discussing about non-inertial reference frames, but the GR determines the metric and the curvature for our 4-dim Riemannian manifold , and therefore what frames are inertial and what frames are non-inertial are completely determined as well.
if you consider a frame of reference that is in rotation compared to a Galilean frame, then an object far enough away from the center of the rotation, and is stationary in the Galilean frame, will be superluminal in the rotating frame.
No, no, no, no. The universe is not rotating around you. Unlike SR, in GR you cannot pick any reference frame that you desire. If you could do this then you could totally fuck up the twin paradox.
Think about the twin paradox, and use your same reasoning. Let the astonaut twin be still and move the entire universe around him. By your reasoning, the astronaut twin would be old and the Earth twin would be young.
So you see, your concept of motion does not lead to results that coincide with reality and are therefore not useful.
Loop Gravity (this article) & Spin Networks (the easiest to quantize space time with)
Noncommutative Geometry (IMO the most promising)
Stochastic Gravity (the most humble)
And I am sure that I am forgeting more. But string theory gets the most attention and the most money. This is odd to me because string theory has/had some of the ugliest assumptions (particles are strings, supersymmetry) and introduces the most extra stuff (lots of extra Kaluza-Klein style dimensions, all kinds of extendend objects, excessive parameters).
If string theory ever reaches current holy grail, then I think it will end up being a completely different theory.
As capybara points out, all of the relativity stuff in the article is about special relativity (light cones, can't go faster than c, etc).
There are light cones in GR, they just get bent up a bit. You can't go faster than c in GR. GR encapsulates SR and, with no energy/momentum, reduces down to SR.
I would add a qualifier, that America seems to be the only 1st world country with any significant amount anti-science creationists. There are a handful of islamic countries like Turkey with a fair share.
In some sense, we are the stupidest of the smartest.
Yes, having many multiple terminal windows open is nice, but guess what! Because of X's idiotic single instance limitation, all of those terminals are children of the same proc. Crash one of them and all twenty of your tasks go down in a horrific pile of smoke. I'll take single threaded rxvts or xterms, thank you.
I've never used OSX, but I do have one thing to say.
Both GNOME and KDE are doing this. GNOME does it via gnome-name-service and KDE just uses tabs (which is somewhat useful at times). This practice saves memory if enough instances are run to
And I don't know how often terms crash on OSX, but they don't seem to do that on my box.
Your right, though it was not my misconception, but an error.
They get confused because they both annihilate the cross terms in probability calculations, one with respect to sums and one with respect to products.
P(A||B) = P(A) + P(B) : for A,B disjiont
In general: P(A||B) = P(A) + P(B) - P(A&B) P(A U B) = P(A) * P(B) : for A,B independent
In general: P(A&B) = P(A)*P(B|A) = P(B)*P(A|B)
I suppose if the language was more technical, there would be one adjective that would apply to both as commutivity and associativity applies to both multiplication and addition. At least it is not as bad as Topology terminology though.
"But if 1/10 of people who like book C like B, and D like B and E like B (etc) AND I like C,D, and E, then you have a 4/10 chance that I will like B. And that starts to be usefull."
Which is mostly false.
Imagine a scenario with more than 11 books C1, C2,... C11 such that 10% of the people who like book Ci also like book B. If one person liked all C books and you were to sum the probabilities as you did, then you would end up with that person being 110% likely to enjoy book B. And this is impossible as 0<=P<=1
The only time that you can add probabilities like that is when there is zero intersection between the groups - independence.
Re:It's a question of limits
on
Science Askew
·
· Score: 1
Wouldn't that be a supremum need not be the maximum. I mean, the Web having strictly nonnegative volume/measure/whatever forces it have an infimum and that has nothing to do with being finite/infinite.
Some arguments he asserts amount to 1=2 (not modulo 1) therefore GR is wrong. But if in fact 1=2, then math would be wrong - and thus there would be no point in math or any mathematical science like physics.
He basically keeps bringing up how some old dead guy calculated the scalar curvature in two different corrdinate systems and got two different answers. However, the curvature only depends on your metric, and Riemannian Geometry doesn't care what coordinate system you use as long as it maps the manifold to and from Euclidean space in a smooth manner.
When I talk about math as axiomatic and perfect, I am not yet applying it to anything. It in itself is pure.
Slashdot Mirror
The FCC I would assume.
Did you collect money and have to pay for repair and upgrades at the end of every level?
Wolfenstein was better than Wolfenstein3D in some regards.
If you were out of ammo, you could sneak up behind a Nazi, hold him up, take his ammo, and then put a bullet in his head.
Awesome.
Journals usually typeset the paper. To do this they start by importing everything into Word. Why - it is a standard. Absolute crap for typesetting, but that is the way it goes. Some journals are getting adept at using TeX and PDF.
In math and physics, all the real journals use TeX exclusively. This is because little to no typesetting is needed. A journal supplied macro/stylesheet is used that defines all of the allowable commands. You only have to put thought into getting the layout correct.
Ofcourse TeX is also used exclusively because nobody has time to click together 50 fucking equations using some god aweful gui.
They call it "snow".
I just wanted to rub it in.
You are taking all the fun out of being a loud mouthed zealot.
For the record. Linux does it both ways; or at least it did. Everyone just seems to prefer one of those ways.
OSX does it both ways too. And because people have complained, they have/are changing the default from a file to a partition.
This is not a linuxcentric phenonina, it is just sensible.
But perhaps you are right. I am aware that the file is contiguous. But there would also have to be some magic that allows the kernel to completely ignore the filesystem when paging (without breaking anything). And the OS would have to be sensible enough to locate the file at the innermost portion of the harddrive.
I suppose that it is possible.
Virtual memory, the memory hit when you run out of real memory, is slow
Maybe it wouldn't be so slow if it were a partition and not a file residing in a file system.
Shit that's just fucking lazy on the programmer's part.
Do they put their database in individual files too?
I'll give you some objective differences. (I might be off on a few)
Does WNT run on over 15 architectures? It was intended to be crossplatform, but it never really did well with that. It's best other platform was the Alpha, and even that was utter shite. I don't think it ever supported 64bits.
Has WNT ever supported 64bits?
Does WNT support pluggable modules?
Does in run on mainframes?
Does anyone other than Microsoft and Microsoft funded persons cluster it? How well does it work with MOSIX?
Does it have a O(1) scheduler?
Does it have a fast Journaling file system that stores files in b-tree's? Does it support dozens of file systems?
Can I boot up, read my root/bin/lib partitiond to a ramdisk, and run completely from RAM?
Can I run with all of my partitions encrypted?
Did NSA make a security-enhanced version? What about when they left those registry keys?
Did they ever move the graphics out of kernel space? Ouch!
Does it have superfast zerocopy for NFS, webservers, and other net stuff?
Will I ever stop?
OTOH I wonder how widespread Linux would be if he had used a BSD style license.
BSD was around at the time.
If you want the best fraction, then use some finite terms of the best rational approximation.
I believe the first few approximations go like
22/7 Archimedes
333/106
335/113 Tsu Chung-chi
103993/33102
Interesting enough Tsu called 22/7 the inaccurate value and his value the accurate value.
You talking is full of jibberish.
Here is the raw deal.
Math exists without need for our physical laws. There are plenty of branches of mathematics with little to no relation to our real world.
The math that does apply to our real world in the form of physical laws, are only mathematical models. None of physics is exact, and it has never been.
For example: Pi cannot be found exactly in this universe.
Pi does not exist in our world embeded in spheres. You cannot construct a perfect sphere in the real world because once you hit the plank scale you can no longer have perfect classical smoothness.
Pi does not exist in our world embeded in oscillators, for much the same reason.
Pi only exists in math, in idea. The fact that we can use pi and math to model the universe to some degree of approximation is only a testiment to how ordered and logical the universe is.
So though you may have not realized it, the poster had an important point. Pi and math would be the same in all universes, because Pi doesn't exist in the universe any how, they only exist in our minds.
That is using the power series for the arctangent right. A much faster and better method is to use the rational approximation of arctangent, which yields the continued fraction for pi. Almost as easy, but much better.
There has been too much misinformation thrown out on both sides but you are dead wrong and the fact that you keep getting modded up shows that Slashdot is not the place for informed discussion.
1) Nothing in GR goes faster than light. Light travels a geodesic at maximal speed. You can spin around in your backyard until you throw up, but that will not move anything from point A to point B in less time than light can do it.
2) I really don't know what the hell you people are discussing about non-inertial reference frames, but the GR determines the metric and the curvature for our 4-dim Riemannian manifold , and therefore what frames are inertial and what frames are non-inertial are completely determined as well.
if you consider a frame of reference that is in rotation compared to a Galilean frame, then an object far enough away from the center of the rotation, and is stationary in the Galilean frame, will be superluminal in the rotating frame.
No, no, no, no. The universe is not rotating around you. Unlike SR, in GR you cannot pick any reference frame that you desire. If you could do this then you could totally fuck up the twin paradox.
Think about the twin paradox, and use your same reasoning. Let the astonaut twin be still and move the entire universe around him. By your reasoning, the astronaut twin would be old and the Earth twin would be young.
So you see, your concept of motion does not lead to results that coincide with reality and are therefore not useful.
There are many contenders
Loop Gravity (this article) & Spin Networks (the easiest to quantize space time with)
Noncommutative Geometry (IMO the most promising)
Stochastic Gravity (the most humble)
And I am sure that I am forgeting more. But string theory gets the most attention and the most money. This is odd to me because string theory has/had some of the ugliest assumptions (particles are strings, supersymmetry) and introduces the most extra stuff (lots of extra Kaluza-Klein style dimensions, all kinds of extendend objects, excessive parameters).
If string theory ever reaches current holy grail, then I think it will end up being a completely different theory.
As capybara points out, all of the relativity stuff in the article is about special relativity (light cones, can't go faster than c, etc).
There are light cones in GR, they just get bent up a bit. You can't go faster than c in GR. GR encapsulates SR and, with no energy/momentum, reduces down to SR.
I would add a qualifier, that America seems to be the only 1st world country with any significant amount anti-science creationists. There are a handful of islamic countries like Turkey with a fair share.
In some sense, we are the stupidest of the smartest.
So that's how you issue the chmod a+w * command in Windows. You must have to be really super smart to use it.
Yes, having many multiple terminal windows open is nice, but guess what! Because of X's idiotic single instance limitation, all of those terminals are children of the same proc. Crash one of them and all twenty of your tasks go down in a horrific pile of smoke. I'll take single threaded rxvts or xterms, thank you.
I've never used OSX, but I do have one thing to say.
Both GNOME and KDE are doing this. GNOME does it via gnome-name-service and KDE just uses tabs (which is somewhat useful at times). This practice saves memory if enough instances are run to
And I don't know how often terms crash on OSX, but they don't seem to do that on my box.
Your right, though it was not my misconception, but an error.
They get confused because they both annihilate the cross terms in probability calculations, one with respect to sums and one with respect to products.
P(A||B) = P(A) + P(B) : for A,B disjiont
In general: P(A||B) = P(A) + P(B) - P(A&B)
P(A U B) = P(A) * P(B) : for A,B independent
In general: P(A&B) = P(A)*P(B|A) = P(B)*P(A|B)
I suppose if the language was more technical, there would be one adjective that would apply to both as commutivity and associativity applies to both multiplication and addition. At least it is not as bad as Topology terminology though.
"But if 1/10 of people who like book C like B, and D like B and E like B (etc) AND I like C,D, and E, then you have a 4/10 chance that I will like B. And that starts to be usefull."
... C11 such that 10% of the people who like book Ci also like book B. If one person liked all C books and you were to sum the probabilities as you did, then you would end up with that person being 110% likely to enjoy book B. And this is impossible as 0<=P<=1
Which is mostly false.
Imagine a scenario with more than 11 books C1, C2,
The only time that you can add probabilities like that is when there is zero intersection between the groups - independence.
Dang, lost my /. password.
Haha
Why is my user number so large?
Oh
Wouldn't that be a supremum need not be the maximum. I mean, the Web having strictly nonnegative volume/measure/whatever forces it have an infimum and that has nothing to do with being finite/infinite.
I mean jeez that's just soo obvious.
you could define 4 as prime if you were so inclined
But then the fundamental theorem of arithmetic would be false. I.E. 2^2 and 4 would be two _different_ factorizations that are equal.
I think that is different from saying that 1^k * Prod p^ki == Prod p^ki as 1^k doesn't ever give you any of the other prime numbers like 2,3,5...
Some arguments he asserts amount to 1=2 (not modulo 1) therefore GR is wrong. But if in fact 1=2, then math would be wrong - and thus there would be no point in math or any mathematical science like physics.
He basically keeps bringing up how some old dead guy calculated the scalar curvature in two different corrdinate systems and got two different answers. However, the curvature only depends on your metric, and Riemannian Geometry doesn't care what coordinate system you use as long as it maps the manifold to and from Euclidean space in a smooth manner.
When I talk about math as axiomatic and perfect, I am not yet applying it to anything. It in itself is pure.