Maybe I'm misunderstanding the question, however it does prove the Universe is at least as old as 5000 years because the light took 5000 years to reach the Earth. It must have had a universe to travel into.
Unless somehow God created all in one go a very complex universe that looked in every respect some 13 billion years old at least, just to fool the scientists that would come some large number of generations down the track. That's going to a whole lot of bother though to cover His tracks.
For a God Almighty nothing is impossible of course.
Actually it is pretty unusual these days when soldiers kill each other, they prefer to kill civilians instead:
Civilian casualties were 20 percent in World War I, 50 percent in World War II, 80 percent in the Vietnam War, and estimated at about 90 percent in today's wars. (J.J.Pettman, _Working Women: A feminist international politics_, London, Routledge, 1996, p.89).
Why would the next iteration of mass storage media be DRM-free ? content companies still have incentive to protect their properties, and the IP tax is working very well, thank you.
With the current system even when production goes offshore the largest proportion of the benefits still goes to the inventors of the technology.
MS has a vision (windows everywhere) and a business plan (own the O/S everywhere from cell phones to supercomputers, crush everybody that disagrees, leverage O/S dominance to applications).
You don't see MS diversifying into airlines, insurance, manufacturing, automobiles and whatnot.
The question is: what is Google up to? Are they on a collision course with Microsoft in the short term ?
In the last few months they have been putting out products for the windows platform only: hard disk search tools, image indexing tools, etc.
To me this means that Google is getting less interested in the search business. Are they going to turn into some run-of-the-mill software house for the windows platform only or are they up to something else?
MacOS/X doesn't blow Linux out of the water uniformly, like the statisticians say.
You can't run MacOS/X on commodity hardware, and it doesn't run 64-bit code right now, much less on commodity 64-bit hardware like the Athlon-64. Linux has been running 64-bits for years on most plaforms that support it, including PPC !
Commodity hardware also means a much wider choice of component. While my iBook is nice, I find it a little slow. A 2GHz CPU in it would be nice, but won't happen for a long time. G5 laptops are not even on the horizon, when I would have been able to run Linux on portable Athlon64 laptops for some time now.
While MacOS/X is much easier to learn, much more polished at the user level, there is very little that it can do that Linux can't. At the expert user, server maintainer or developer level Linux is much more configurable and flexible. Why do I have to click to focus, to give a trivial example.
While overall MacOS/X is better supported by manufacturer than Linux, this is not uniformly true. For example AFAIK java 1.5 doesn't run yet on Macs, but I may be wrong. Doom3 is another example.
Finally Linux is 100% Free. This means that with enough effort and motivation Linux can be made to do just about anything computable, whereas with OS/X one is at the mercy of a commercial entity. So far it has been pretty benign but things could change.
Should the dollar continue to slide, it should have negative effects on the price of crude oil.
Expensive crude oil would make it hard for the American economy to continue to improve. American manufactured good would become more expensive to produce.
In the early 70s the dollar was very low, and it wasn't until the end of the decenny when the dollar had more than doubled in value that the American economy started the period of growth that continued until the end of the century.
The expensive dollar made the crude oil cheaper for America at a time when its price was naturally climbing due to OPEC's political stance, and much more expensive in Europe.
Wasn't making money off Iraq the one point of the war? Iraq was the second-largest oil producer in the world and sits over one of the largest known oil reserve.
The largest one (Saudi Arabia) is already under strong American control. Iraq was in danger of becoming a centre of profit for European oil companies, but now thanks to the war it will again be under American control.
If you don't believe that these were the reasons for the war then you need to reevaluate how and why politicians make decisions.
Because it is extremely inefficient to simulate the quantum world, as everything happens in parallel.
In effect you go back to square one. To simulate N qbits roughly your quantum computer simulator must have the capacity to completely explore 2^N states. It quickly becomes unmanageable, and you revert to the original problem.
Equivalently you can say that if you have the traditional computing power to solve the problems that a given quantum computer would be able to solve easily, then you approximately have the capacity to simulate this quantum computer (give or take an order of magnitude perhaps).
Your approach wouldn't work. Perhaps a given fast computer would, say, be able to simulate 7 qbits. Then 2 such computers would only be able to simulate 8 qbits, not 14 ; a thousand such computers would only be able to simulate 17 qbits, and so on.
BTW, some people say that the reason why we haven't been able to produce a strong AI yet is that some quantum effects happen in the brain. Roger Penrose in particular is a big proponent of this idea.
On the other hand some AI people say that the only reasons why we haven't got strong AI yet is (a) we don't really know how the brain works yet, and (b) we need more computing capacity. More research is needed for (a), and Penrose would agree with this, but eventually we'll have all the computing capacity we need.
However if thought is based on quantum processes then we might require quantum computers to simulate it, who knows. This could mean that strong AI is some years away.
Yes you do have to choose between career and family, to some extent, and the choice is much more acute for women. It is not all that bad, both alternatives have good and bad sides. Having kids is its own reward!
Unfortunately it is believed that research capability tapers off with age, and it is true to some degree. I assume this is the reason why universities are so hard on their junior staff. They pressure them to produce results at an early age because this is when they are the most productive.
Speaking as a researcher myself with kids, having children does put an enormous damper on the time you have to yourself and therefore your job. A research job is not 9-5, it 100% of the time or pretty close to it. Things can change if you can count on your husband or your extended family to do a great deal. You can then still put the extra mile from time to time, but not on a regular basis (or expect the regularity not to happen!). Make sure they are on board.
Starting having children at 35 is not great BTW. You tire more easily than at 25 or 30.
All the best with your choices, but remain confident, having kids is only really hard during their first year, it does ease off a bit after that. No matter what, if you are a good researcher a couple of kids will not prevent you from getting a great career.
It is interesting to see that people don't believe they are part of a crowd with relatively predictable behaviours.
The fact is that relying on the behaviours of individuals has been found by businesses to be less efficient than to rely on statistics.
Like it or not, as a younger driver you are much more likely to be involved in an accident than when you are older. It makes more sense for insurance companies to make younger driver pay more and older established driver less.
Eventually younger drivers get experience and pay less. The good thing is that you are likely to be an established driver much longer than a learner.
If it made any business sense to do what you proposed you'd see a company proposing the conditions that you want. What would happen is that this company would get all the high-risk younger drivers and none of the established, low risk ones, who would go to the competition practicing discrimination.
Yes, for example why is it that if you want to edit a cell in Excel, you need to click on it, then move your cursor to the editor field in the toolbar?
This is terribly inefficient, yet Gnumeric and OOo do the same thing.
No other word processor, perhaps, but there are excellent open-source solutions for writing material.
I'm not a textbook author but I'm a researcher, I write stuff every day as well, from course notes to articles via presentations.
Honest question: What do you think your chances are to be able to re-read and re-edit your current work in 15 years time?
AFAIK no current version of Word can correctly read 1990 word-4.0 files. I've had that problem not long ago.
I a previous job I was handling word files regularly. It was pretty common to see one colleague print the very same version of the file you have, with the same version of the software on the same printer and have it still look different. Sometimes dramatically so, like uncorrectly labelled lists.
My experience with revision notes in Word (2001, not 2003) is that they don't compare with a revision control system. Word revisions do work between 2 people but not more, and it quickly becomes very hard to track who's done what in heavily edited areas. In case of disagreement they can become a liability. Hopefully the 2003 version are better.
OTOH since I moved from Word to TeX 14 years ago, I've had zero problem of that nature since it only uses ASCII (or Unicode these days). Revisions between multiple people are handled with CVS (or subversion these days), so you know exactly who did what when and why (thanks to comments), and you can revert easily.
I've edited two books of proceedings. The first was done using only TeX, and it was essentially a matter of deciding how the book would look in the end. It was possible to produce a single PDF document of the whole book with a consistent look without too many problems. We got a refund from the publisher because they assumed that we would deliver them a PDF document with too many mistakes in them that they would spend days to correct, but it simply didn't happen. It was basically right the first time.
For the second we accepted Word documents, and it was a nightmare to make all of them look even remotely alike, and we didn't succeed in producing a single PDF of all of them. It remained a collection of disparate individual articles, and we did spend about a week with the publisher fixing all manners of mistakes (like missing figures, page numbers, etc...)
GNUstep has made huge strides in the last couple of years but it is not what I would call a bulletproof development environment, it has compatibility problems with what Interface Builder generates on the Macs (and old NeXTStep workstations) and it only implements a subset of the Mac Toolkit anyway.
On the other hand Sun has access to the old OpenStep implementation since they co-developed it with NeXT.. IIRC it is more complete and more stable than GNUStep, although it relies on PS rather than PDF for drawing. It would be interesting if it could be open-sourced and like you suggest made the basis of their suite.
Unfortunately I don't think this will happen. Sun has a huge vested interest in X.
Most of the time in compiled applications you have instructions and data in your cache, and not so many pointers.
Data or instruction size don't change in a 64-bit machine, however your capacity to manipulate this data improves because you get access to 64-bit registers. On some machines you even get many more registers, which can help (it certainly does on AMD's x86_64)
I have found that without exceptions so far recompiling an application for 64-bit processors has improved performance on all the platforms I've tried, including SGI.
Now if you run a lot of interpreters or virtual machines or if you run C++ code with lots of virtual methods, then indeed your pointer count is high in the cache and performance may decrease, but this is not a foregone conclusion.
Semantics, vocabulary. Replace fact with observation.
Observable events are the foundation of science, by definition without observation one cannot have science. BTW this is why mathematics are not a science. These collection of observations are popularly called facts. A fact is something that has indeed been observed. A fact and an observation are one and the same.
By extention facts are loosely associated with extremely probable events like you describe, as in "it's a fact that the sun rises every morning", but this is not the definition that I'm using. For me it is a fact that the sun did rise yesterday.
When an apple becomes ripe enough it has been observed that it falls from the tree in an extremely deterministic fashion (down). This is not theory, this is really what has happened (as far as reality exists).
While quantum mechanic theory (for example) would predict that in extremely infrequent occurences the apple would in fact fall sideways due to tunnelling effects, in reality this has never been observed, and so it has never been a fact.
However it is a fact that electrons do tunnel through junctions in semiconductors due to the same effect. Am I repeating myself too much here?
Actually you should check the Apple store prices. Apple makes the *cheapest* laptops and desktops these days. Cheaper than DELL.
I know because only a few months ago I was on the lookout for the cheapest laptop that I could possibly buy with very low level requirements. I only wanted it with worldwide warranty (this is for actually travelling with it) reasonably light (not 3kg) and with a reasonable battery life (not 90 minutes). To my utter amazement Apple had the best offer by far.
Slashdot Mirror
You'll find that in many wars and disasters people hang on to what they have because this is all they have and they think they can somehow defend it.
Talk to firemen. It's pretty hard to make someone leave their house even in the middle of a raging inferno.
Don't be too sure about what your own reaction would be in a similar situation. These are not situations where logic and rationality prevail.
Under MacOS/X there is a settings where the keyboard is Dvorak for normal typing but Qwerty as soon as the "apple" key is depressed for shortcuts.
Therefore you get all the advantages of Dvorak (or supposed advantages) and your shortcuts are in the same place as ever.
Maybe I'm misunderstanding the question, however it does prove the Universe is at least as old as 5000 years because the light took 5000 years to reach the Earth. It must have had a universe to travel into.
Unless somehow God created all in one go a very complex universe that looked in every respect some 13 billion years old at least, just to fool the scientists that would come some large number of generations down the track. That's going to a whole lot of bother though to cover His tracks.
For a God Almighty nothing is impossible of course.
Actually it is pretty unusual these days when soldiers kill each other, they prefer to kill civilians instead:
Civilian casualties were 20 percent in World War I, 50 percent in World War II, 80 percent in the Vietnam War, and estimated at about 90 percent in today's wars. (J.J.Pettman, _Working Women: A feminist international politics_, London, Routledge, 1996, p.89).
Why would the next iteration of mass storage media be DRM-free ? content companies still have incentive to protect their properties, and the IP tax is working very well, thank you.
With the current system even when production goes offshore the largest proportion of the benefits still goes to the inventors of the technology.
Replace endgame with "vision" or business plan.
MS has a vision (windows everywhere) and a business plan (own the O/S everywhere from cell phones to supercomputers, crush everybody that disagrees, leverage O/S dominance to applications).
You don't see MS diversifying into airlines, insurance, manufacturing, automobiles and whatnot.
The question is: what is Google up to? Are they on a collision course with Microsoft in the short term ?
In the last few months they have been putting out products for the windows platform only: hard disk search tools, image indexing tools, etc.
To me this means that Google is getting less interested in the search business. Are they going to turn into some run-of-the-mill software house for the windows platform only or are they up to something else?
> Name a feature in XP Pro that a typical home user would need
Immunity to spyware, thank you (works in Linux, so far).
Disclaimer, I'm typing this on my beloved iBook.
MacOS/X doesn't blow Linux out of the water uniformly, like the statisticians say.
You can't run MacOS/X on commodity hardware, and it doesn't run 64-bit code right now, much less on commodity 64-bit hardware like the Athlon-64. Linux has been running 64-bits for years on most plaforms that support it, including PPC !
Commodity hardware also means a much wider choice of component. While my iBook is nice, I find it a little slow. A 2GHz CPU in it would be nice, but won't happen for a long time. G5 laptops are not even on the horizon, when I would have been able to run Linux on portable Athlon64 laptops for some time now.
While MacOS/X is much easier to learn, much more polished at the user level, there is very little that it can do that Linux can't. At the expert user, server maintainer or developer level Linux is much more configurable and flexible. Why do I have to click to focus, to give a trivial example.
While overall MacOS/X is better supported by manufacturer than Linux, this is not uniformly true. For example AFAIK java 1.5 doesn't run yet on Macs, but I may be wrong. Doom3 is another example.
Finally Linux is 100% Free. This means that with enough effort and motivation Linux can be made to do just about anything computable, whereas with OS/X one is at the mercy of a commercial entity. So far it has been pretty benign but things could change.
What I say for Linux holds for the BSDs too.
Born in Toulouse?
Should the dollar continue to slide, it should have negative effects on the price of crude oil.
Expensive crude oil would make it hard for the American economy to continue to improve. American manufactured good would become more expensive to produce.
In the early 70s the dollar was very low, and it wasn't until the end of the decenny when the dollar had more than doubled in value that the American economy started the period of growth that continued until the end of the century.
The expensive dollar made the crude oil cheaper for America at a time when its price was naturally climbing due to OPEC's political stance, and much more expensive in Europe.
With a cool head it is difficult to argue against the fact that the tax cuts helped fuel the trade deficit, at least in the short term.
Wasn't making money off Iraq the one point of the war? Iraq was the second-largest oil producer in the world and sits over one of the largest known oil reserve.
The largest one (Saudi Arabia) is already under strong American control. Iraq was in danger of becoming a centre of profit for European oil companies, but now thanks to the war it will again be under American control.
If you don't believe that these were the reasons for the war then you need to reevaluate how and why politicians make decisions.
Because it is extremely inefficient to simulate the quantum world, as everything happens in parallel.
In effect you go back to square one. To simulate N qbits roughly your quantum computer simulator must have the capacity to completely explore 2^N states. It quickly becomes unmanageable, and you revert to the original problem.
Equivalently you can say that if you have the traditional computing power to solve the problems that a given quantum computer would be able to solve easily, then you approximately have the capacity to simulate this quantum computer (give or take an order of magnitude perhaps).
Your approach wouldn't work. Perhaps a given fast computer would, say, be able to simulate 7 qbits. Then 2 such computers would only be able to simulate 8 qbits, not 14 ; a thousand such computers would only be able to simulate 17 qbits, and so on.
BTW, some people say that the reason why we haven't been able to produce a strong AI yet is that some quantum effects happen in the brain. Roger Penrose in particular is a big proponent of this idea.
On the other hand some AI people say that the only reasons why we haven't got strong AI yet is (a) we don't really know how the brain works yet, and (b) we need more computing capacity. More research is needed for (a), and Penrose would agree with this, but eventually we'll have all the computing capacity we need.
However if thought is based on quantum processes then we might require quantum computers to simulate it, who knows. This could mean that strong AI is some years away.
Hi,
Yes you do have to choose between career and family, to some extent, and the choice is much more acute for women. It is not all that bad, both alternatives have good and bad sides. Having kids is its own reward!
Unfortunately it is believed that research capability tapers off with age, and it is true to some degree. I assume this is the reason why universities are so hard on their junior staff. They pressure them to produce results at an early age because this is when they are the most productive.
Speaking as a researcher myself with kids, having children does put an enormous damper on the time you have to yourself and therefore your job. A research job is not 9-5, it 100% of the time or pretty close to it. Things can change if you can count on your husband or your extended family to do a great deal. You can then still put the extra mile from time to time, but not on a regular basis (or expect the regularity not to happen!). Make sure they are on board.
Starting having children at 35 is not great BTW. You tire more easily than at 25 or 30.
All the best with your choices, but remain confident, having kids is only really hard during their first year, it does ease off a bit after that. No matter what, if you are a good researcher a couple of kids will not prevent you from getting a great career.
It is interesting to see that people don't believe they are part of a crowd with relatively predictable behaviours.
The fact is that relying on the behaviours of individuals has been found by businesses to be less efficient than to rely on statistics.
Like it or not, as a younger driver you are much more likely to be involved in an accident than when you are older. It makes more sense for insurance companies to make younger driver pay more and older established driver less.
Eventually younger drivers get experience and pay less. The good thing is that you are likely to be an established driver much longer than a learner.
If it made any business sense to do what you proposed you'd see a company proposing the conditions that you want. What would happen is that this company would get all the high-risk younger drivers and none of the established, low risk ones, who would go to the competition practicing discrimination.
In rock climbing muscle mass is not an asset. It is far better to be weaker and lighter than big and strong.
- use your legs
- use your eyes
- lose weight.
Until the crash, Concord ran at an operational profit, meaning it was better to let it fly than not, but it never repaid the development costs.
So overall the project ended up posting a significant loss, meaning that financially it would have been better not to start.
Now prestige, cooperation, R&D benefits and so on can't really be measured purely on financial terms.
Yes, for example why is it that if you want to edit a cell in Excel, you need to click on it, then move your cursor to the editor field in the toolbar?
This is terribly inefficient, yet Gnumeric and OOo do the same thing.
Hi,
No other word processor, perhaps, but there are excellent open-source solutions for writing material.
I'm not a textbook author but I'm a researcher, I write stuff every day as well, from course notes to articles via presentations.
Honest question: What do you think your chances are to be able to re-read and re-edit your current work in 15 years time?
AFAIK no current version of Word can correctly read 1990 word-4.0 files. I've had that problem not long ago.
I a previous job I was handling word files regularly. It was pretty common to see one colleague print the very same version of the file you have, with the same version of the software on the same printer and have it still look different. Sometimes dramatically so, like uncorrectly labelled lists.
My experience with revision notes in Word (2001, not 2003) is that they don't compare with a revision control system. Word revisions do work between 2 people but not more, and it quickly becomes very hard to track who's done what in heavily edited areas. In case of disagreement they can become a liability. Hopefully the 2003 version are better.
OTOH since I moved from Word to TeX 14 years ago, I've had zero problem of that nature since it only uses ASCII (or Unicode these days). Revisions between multiple people are handled with CVS (or subversion these days), so you know exactly who did what when and why (thanks to comments), and you can revert easily.
I've edited two books of proceedings. The first was done using only TeX, and it was essentially a matter of deciding how the book would look in the end. It was possible to produce a single PDF document of the whole book with a consistent look without too many problems. We got a refund from the publisher because they assumed that we would deliver them a PDF document with too many mistakes in them that they would spend days to correct, but it simply didn't happen. It was basically right the first time.
For the second we accepted Word documents, and it was a nightmare to make all of them look even remotely alike, and we didn't succeed in producing a single PDF of all of them. It remained a collection of disparate individual articles, and we did spend about a week with the publisher fixing all manners of mistakes (like missing figures, page numbers, etc...)
To me Word has a long way to go still.
GNUstep has made huge strides in the last couple of years but it is not what I would call a bulletproof development environment, it has compatibility problems with what Interface Builder generates on the Macs (and old NeXTStep workstations) and it only implements a subset of the Mac Toolkit anyway.
On the other hand Sun has access to the old OpenStep implementation since they co-developed it with NeXT.. IIRC it is more complete and more stable than GNUStep, although it relies on PS rather than PDF for drawing. It would be interesting if it could be open-sourced and like you suggest made the basis of their suite.
Unfortunately I don't think this will happen. Sun has a huge vested interest in X.
$36 billion *sales*. RTFA. Not a bad number mind you.
actually 33 bits.
Most of the time in compiled applications you have instructions and data in your cache, and not so many pointers.
Data or instruction size don't change in a 64-bit machine, however your capacity to manipulate this data improves because you get access to 64-bit registers. On some machines you even get many more registers, which can help (it certainly does on AMD's x86_64)
I have found that without exceptions so far recompiling an application for 64-bit processors has improved performance on all the platforms I've tried, including SGI.
Now if you run a lot of interpreters or virtual machines or if you run C++ code with lots of virtual methods, then indeed your pointer count is high in the cache and performance may decrease, but this is not a foregone conclusion.
Semantics, vocabulary. Replace fact with observation.
Observable events are the foundation of science, by definition without observation one cannot have science. BTW this is why mathematics are not a science. These collection of observations are popularly called facts. A fact is something that has indeed been observed. A fact and an observation are one and the same.
By extention facts are loosely associated with extremely probable events like you describe, as in "it's a fact that the sun rises every morning", but this is not the definition that I'm using. For me it is a fact that the sun did rise yesterday.
When an apple becomes ripe enough it has been observed that it falls from the tree in an extremely deterministic fashion (down). This is not theory, this is really what has happened (as far as reality exists).
While quantum mechanic theory (for example) would predict that in extremely infrequent occurences the apple would in fact fall sideways due to tunnelling effects, in reality this has never been observed, and so it has never been a fact.
However it is a fact that electrons do tunnel through junctions in semiconductors due to the same effect. Am I repeating myself too much here?
Russell did not resolve the issues. He tackled them and so did many before him and after, see Plato,Descartes, Kant, Moore, etc.
Actually you should check the Apple store prices. Apple makes the *cheapest* laptops and desktops these days. Cheaper than DELL.
I know because only a few months ago I was on the lookout for the cheapest laptop that I could possibly buy with very low level requirements. I only wanted it with worldwide warranty (this is for actually travelling with it) reasonably light (not 3kg) and with a reasonable battery life (not 90 minutes). To my utter amazement Apple had the best offer by far.
As you may know Apple now sells a $450 desktop.