Since most of the GDP comes from the consumption of the middle class, any loss of jobs results in poor growth and diminished profits for the rich as well, unless the external market for goods expands.
Sorry, I'm throwing away all of my mod points just to respond to this nonsense.
This is **NOT** the RedHat model coming to Windows. Redhat's subscription is 100% voluntary. You can still get all of their software, including the Enterprise stuff, without spending a single cent in subscription. Redhat's software is GPL, it is guaranteed to remain Free forever.
With RedHat, you pay if you believe their service actually add value to your business. With Microsoft, if you *don't* pay, your business can't run. Period.
Well arXiv is just a repository of paper in near-draft status. That there are good papers in them is great, but almost incidental. These papers also get published somewhere else afterwards. Unless you are looking for stuff in your field and you go by word of mouth then there is no way of knowing if the paper you are reading holds any value. Trying to understand a paper not in your own field is usually a non-trivial undertaking, you'd like to know if it's been reviewed before you make the effort.
For example of non-trivial crap look up the numerous proofs of both P=NP and P!=NP in the math/computer science sections.
Yes the peer review process is only as good as the reviewers. I find that this is precisely what makes the difference between a good journal and a poorer one. With better journals, reviewers often make a more significant effort at providing insightful comments, or the editor will make a better effort at finding reviewers that do provide insightful comments. Sometimes a single comment can change the orientation of a paper around completely !
I'm more optimistic than you, I think IE7 will be more bloated, slower and even less reliable than IE6, which I try never to use except for windowsupdates.
Until Moore's law makes the integrated device that will be the mobile phone in a number of years capable of holding all of your music, until advances in HCI and lens technology makes the camera phone a workable alternative to respectable cameras, until battery techology improves to the point where you can have an autonomy of a week with all that stuff, people will still use a separate music player, a separate camera and a separate phone.
Sure, people will buy the integrated everything monster, because feature competition will very soon dictate that all "mobile phones" must have all of the above features plus more (java games!). In fact they will not buy them, they will rent them on monthly plans like they do now, and I believe not use them very much except for yacking away, just like today.
Will that put a dent into the iPod sales? sure, when everybody who wants one already has one.
Have you also noticed that phones compete on size? most people like a tiny phone. At the moment that means a tiny memory, a really tiny screen, a minuscule camera and a tiny battery with not much juice to power all of this. This is not going in the direction that Bill Gates wants, expecially with windows for mobiles.
The problem with arXiv.org is that the utter crap is right next to the brilliant, and there is far too much of the former. It makes it a complete waste of time to browse the archive. It is only useful as a more permanent repository.
Top scientists are usually editors of journals or series. They do their bit with regards to the peer review process. Young scientists can do most of the actual peer reviewing, this is not a problem as there are more of them, and it's not clear who is more afraid of novelty, whether it's old or young scientists.
Since the equilibrium has been disturbed we are in a time of change, and so lots of things are in a state of flux. I think journals will continue, they have the peer-review in place and that is the only thing that distinguishes science from crap. They will just become cheaper and more easily available, not the other way around.
It seems pretty obvious to me that Microsoft is indeed at the crossroad, but not at the crossroad to disappear.
They are simply too huge to go away. They would have to make mistake after mistake for the stock to drop and for confidence to evaporate.
Hopefully Microsoft will soon realize that the wild growth of the 90s is gone, that they have run out of IT sectors to simply phagocyte, that they now need to really innovate rather than copy and do good-enough work, that they need play nicer and start collaborating a bit more.
There was a survey yesterday that said that basically people who choose open-source do that to avoid vendor lock-in, not for the price.
Microsoft cannot lower prices and recapture lost market, this is a race to the bottom that they cannot win. What this survey says is that they also cannot embrace-and-extend standards they way they used to because the industry has wised up to this strategy.
They pretty much own the desktop market, but there is no growth there except the natural growth of the market itself. They cannot grow all that much on the server market because Windows is not enough of a jack-of-all-trade, doesn't run competitively on large machines, and that the cheap servers run on Linux/BSD.
They are stuck. Sure they can grow on consoles, in the living room and on mobiles, but there is more competition there, and the margins aren't as fat.
Microsoft will not go away, but I wish they would realize that, become less paranoid and start behaving like a better corporate citizen. A bit like IBM has become. Start with following and proposing standards that other people can interact with.
In the old days if you wanted to read a particular paper in a journal your library didn't carry you had to contact one of the authors and ask for a reprint of the article, which you would receive by snail-mail a few weeks later.
Now you just look it up on Google, most of the time it's there, or the author will send you a PDF a few hours later.
The main contribution of journals to research is no longer diffusion, now people usually don't go to the library to read a journal. They receive a summary of the month's issue by email and then go and consult it online. Clearly this could be replaced by informal web publication just as easily.
However the editorial board work is still essential. They make sure the peer-review process runs smoothly and that each paper looks nice in the end. This is not so easily replaced, even though the editors do a volunteer job.
What is definitely not clear is why journal should be allowed to charge scientist huge premiums for the privilege of having those same scientist work for them for free.
Over the next few years we should see the reactive journal boards realize this, and propose a very cheap online-only service. The IEEE is already thinking about this very hard. When others realize this works fine, the era of expensive printed journal will simply come to an end.
Next will be the issue of books. Scientists are already realizing that it is now extremely cheap to self-publish. Even a top-quality, 500 pages book costs less than $40 to print in small quantities. Yet publishing houses typically sell them $200 a piece or more. Then they go out of print but since the publisher has the copyright everybody is screwed.
For conferences, self-publishing is now more cost effective, and authors get to keep their copyright. Soon the era of expensive conference proceedings will also come to an end.
The last remaining bastion will be reference books or textbooks. These will remain in print for the next few years, because people appreciate having a nice book in hand rather than reading hundreds of pages online, but as the cost, speed and quality of desktop printers improve, we should see a new era of freely available, high-quality online textbooks. There are lots of them online already, ready for printing.
All of this will be good for science. No one will be able to claim in a paper they didn't know about so and so's work and don't have access to it. It will be increasingly easy to do dilettante science without the backing of a huge academic institution.
People will be able to follow a field of science extremely easily. Cross-fertilization will become the obvious way to make progress.
1- The original IE team was disbanded. The codebase has not been maintained seriously for *years* 2- Chances are this codebase is ugly and huge, and we all know that IE is very buggy and non-standard compliant. 3- Microsoft cannot allow IE to change every 3 months, and they cannot afford to start from scratch all over again. 4- Microsoft has a history of perverting standards to their own end. 5- Microsoft has lots of resources, but as we all know, putting more people on a late project makes it even later.
Microsoft has been resting on its laurels for a little too long.
Given all of the above, I'd say the next version of IE will simply be worse than the current one. For Microsoft, things are going to get worse before they get better, exactly how it went for Mozilla. Now however Mozilla has put its hard years behind, they are on a roll.
I wouldn't be surprised if the version of IE that will come with Longhorn will be so bad that people will switch en masse.
1- This is an online poll. About as useful as Slashdot's. 2- They had to remove Yoper Linux from first place to make it look as if they had remotely believable results. 3- They still compare results from year to year and try to infer conclusions.
Even with excellent methodology it is difficult to extract usable information from survey data (look at the vast majority of medical surveys. We still don't know for sure if mobile phones are dangerous and it took decades to persuade enough people than smoking was indeed deadly). In this case the results are useless and are a waste of time other than for entertainment value.
You are doing statistics on 3 events, essentially your margin of error is so huge that you can't infer any conclusion.
Both were risky endeavours ; however the parent's point is not completely moot. Apollo 1 was a new design and caused the death of 3 people on the ground. The rest of the missions went OK, even Apollo 13 who had massive systems failures, but enough redundancy built-in to make it back safely. Moreover you are not counting the Apollo predecessors which were somewhat similar in design, but simpler, and which didn't suffer any casualty.
In contrast, the SS exploded twice unexpectedly in the middle of the program. Apparently the SS is a complex system with design flaws no one really knows how to fix.
At any rate the SS is far deadlier than it was designed to be.
The terrorists have been hitting Western targets in their home countries for decades. It's actually much easier for them, nothing new here.
I doubt OBL and his friends had the resources to carry out two attacks of the magnitude of 9/11 anyway. Perhaps others have tried and failed, but somehow I believe that if a vast conspiracy of the same size as that attack had been foiled by US services everyone would have heard about it.
This is not to say that the US effort was for nothing, but rather that so far they have caught relatively small fry.
In general image understanding is equivalent to general AI. We won't get a CBIR system that works well before we get an AI that works and vice versa, because people expect to be able to match the *content* of the image they submit as template and not the general appearance of the image. The problem is then too unspecified.
Even in the restricted context of aeroplanes this is not a trivial problem. Someone in the list of replies submitted an image of a warthog (A-10) and got nonsensical results. Somehow the CBIR system would need to be able to infer a model of the A-10 from a given random 2D projection, and match it to the other 2D projections of the same A-10 model that they do have in the DB. This doesn't sound impossible but it is hard, and I suspect the Penn State people didn't do that. Instead they are probably matching on colour, texture, general appearance, etc.
This is not to say that CBIR is not a nice problem to apply new image processing/image analysis algorithms to, which are developed all the time.
Milk does help a little but not much. A baby's immunity mostly comes from the fact that they share their bloodstream with their mother's in the womb, and thus benefit from the mother's antibodies.
When babies are born their immune system has yet to produce antibodies. The mother's antibodies are slowly elimitated in the first 3-6 months of the life of the baby.
Then the young child starts becoming sick. Most new parents witness their baby's first cold, flu, ear infection etc at between 3 to 6 months of age.
Over the next year or so babies are often sick a lot, especially if they are in day care, and thus exposed to a lot of bugs. As they fall prey to a variety of illnesses they start developing their immune system, which doesn't fully mature until young adulthood.
A lot of criminals are either not very smart people (if you are smart there are much better ways to make a living) who think they will never get caught or desperate people who have essentially no choice. Hence a harsh punishment as deterrent simply doesn't work.
Witness the extremely tough sentences in south-east Asia for *carrying* drugs.
It doesn't require you to do any buy-in, just like proprietary and FOSS software can coexist on the same machine.
You can be a capitalist by day, as in be part of standard corporate America, and a commie by night, by doing some volunteer work for a local association.
Oil at $100 a barrel doesn't just mean that filling up your car is more expensive. Everything depends on oil, including growing and transporting your food.
The whole capitalist system is mostly a pyramid scheme that depends on growth. Expensive everything means less growth, more unemployment and potentially a nasty negative spiral when debt repaiments are not met, at the level of a whole country.
It can be very nasty. We will not run out of oil or out of oil replacements, this is not the problem.
Perhaps you are referring to a "crisis" as something as mild as what happened in the 70s when oil suddently became 3x as more expensive overnight. Oil was still available, it was just suddently not quite as cheap as before.
Now imagine that oil became suddenly 3x as expensive as it is now within the next 12 months. Would it be fun ? Is it impossible ?
This is not the worst that could happen. To bring us back to the middle ages would only take a small nuclear war. How does nuclear war between China and the US over oil sound to you ? Impossible again ? Within 20 years time China will be consuming as much oil as the US is consuming today.
One can only assume that China will have as aggressive a policy then as the US has now. I can't imagine things will go down smoothly.
Slashdot Mirror
That's in fact 10,000 days or 27 years.
According to lore, his best lovvy line in the entire trilogy, in Cloud City:
Leia: I love you
Han: I know.
was improvised on the set by Ford.
Since most of the GDP comes from the consumption of the middle class, any loss of jobs results in poor growth and diminished profits for the rich as well, unless the external market for goods expands.
Sorry, I'm throwing away all of my mod points just to respond to this nonsense.
This is **NOT** the RedHat model coming to Windows. Redhat's subscription is 100% voluntary. You can still get all of their software, including the Enterprise stuff, without spending a single cent in subscription. Redhat's software is GPL, it is guaranteed to remain Free forever.
With RedHat, you pay if you believe their service actually add value to your business. With Microsoft, if you *don't* pay, your business can't run. Period.
Well arXiv is just a repository of paper in near-draft status. That there are good papers in them is great, but almost incidental. These papers also get published somewhere else afterwards. Unless you are looking for stuff in your field and you go by word of mouth then there is no way of knowing if the paper you are reading holds any value. Trying to understand a paper not in your own field is usually a non-trivial undertaking, you'd like to know if it's been reviewed before you make the effort.
For example of non-trivial crap look up the numerous proofs of both P=NP and P!=NP in the math/computer science sections.
Yes the peer review process is only as good as the reviewers. I find that this is precisely what makes the difference between a good journal and a poorer one. With better journals, reviewers often make a more significant effort at providing insightful comments, or the editor will make a better effort at finding reviewers that do provide insightful comments. Sometimes a single comment can change the orientation of a paper around completely !
I'm more optimistic than you, I think IE7 will be more bloated, slower and even less reliable than IE6, which I try never to use except for windowsupdates.
Until Moore's law makes the integrated device that will be the mobile phone in a number of years capable of holding all of your music, until advances in HCI and lens technology makes the camera phone a workable alternative to respectable cameras, until battery techology improves to the point where you can have an autonomy of a week with all that stuff, people will still use a separate music player, a separate camera and a separate phone.
Sure, people will buy the integrated everything monster, because feature competition will very soon dictate that all "mobile phones" must have all of the above features plus more (java games!). In fact they will not buy them, they will rent them on monthly plans like they do now, and I believe not use them very much except for yacking away, just like today.
Will that put a dent into the iPod sales? sure, when everybody who wants one already has one.
Have you also noticed that phones compete on size? most people like a tiny phone. At the moment that means a tiny memory, a really tiny screen, a minuscule camera and a tiny battery with not much juice to power all of this. This is not going in the direction that Bill Gates wants, expecially with windows for mobiles.
The problem with arXiv.org is that the utter crap is right next to the brilliant, and there is far too much of the former. It makes it a complete waste of time to browse the archive. It is only useful as a more permanent repository.
Top scientists are usually editors of journals or series. They do their bit with regards to the peer review process. Young scientists can do most of the actual peer reviewing, this is not a problem as there are more of them, and it's not clear who is more afraid of novelty, whether it's old or young scientists.
Since the equilibrium has been disturbed we are in a time of change, and so lots of things are in a state of flux. I think journals will continue, they have the peer-review in place and that is the only thing that distinguishes science from crap. They will just become cheaper and more easily available, not the other way around.
It seems pretty obvious to me that Microsoft is indeed at the crossroad, but not at the crossroad to disappear.
They are simply too huge to go away. They would have to make mistake after mistake for the stock to drop and for confidence to evaporate.
Hopefully Microsoft will soon realize that the wild growth of the 90s is gone, that they have run out of IT sectors to simply phagocyte, that they now need to really innovate rather than copy and do good-enough work, that they need play nicer and start collaborating a bit more.
There was a survey yesterday that said that basically people who choose open-source do that to avoid vendor lock-in, not for the price.
Microsoft cannot lower prices and recapture lost market, this is a race to the bottom that they cannot win. What this survey says is that they also cannot embrace-and-extend standards they way they used to because the industry has wised up to this strategy.
They pretty much own the desktop market, but there is no growth there except the natural growth of the market itself. They cannot grow all that much on the server market because Windows is not enough of a jack-of-all-trade, doesn't run competitively on large machines, and that the cheap servers run on Linux/BSD.
They are stuck. Sure they can grow on consoles, in the living room and on mobiles, but there is more competition there, and the margins aren't as fat.
Microsoft will not go away, but I wish they would realize that, become less paranoid and start behaving like a better corporate citizen. A bit like IBM has become. Start with following and proposing standards that other people can interact with.
Disclaimer, I'm a researcher.
In the old days if you wanted to read a particular paper in a journal your library didn't carry you had to contact one of the authors and ask for a reprint of the article, which you would receive by snail-mail a few weeks later.
Now you just look it up on Google, most of the time it's there, or the author will send you a PDF a few hours later.
The main contribution of journals to research is no longer diffusion, now people usually don't go to the library to read a journal. They receive a summary of the month's issue by email and then go and consult it online. Clearly this could be replaced by informal web publication just as easily.
However the editorial board work is still essential. They make sure the peer-review process runs smoothly and that each paper looks nice in the end. This is not so easily replaced, even though the editors do a volunteer job.
What is definitely not clear is why journal should be allowed to charge scientist huge premiums for the privilege of having those same scientist work for them for free.
Over the next few years we should see the reactive journal boards realize this, and propose a very cheap online-only service. The IEEE is already thinking about this very hard. When others realize this works fine, the era of expensive printed journal will simply come to an end.
Next will be the issue of books. Scientists are already realizing that it is now extremely cheap to self-publish. Even a top-quality, 500 pages book costs less than $40 to print in small quantities. Yet publishing houses typically sell them $200 a piece or more. Then they go out of print but since the publisher has the copyright everybody is screwed.
For conferences, self-publishing is now more cost effective, and authors get to keep their copyright. Soon the era of expensive conference proceedings will also come to an end.
The last remaining bastion will be reference books or textbooks. These will remain in print for the next few years, because people appreciate having a nice book in hand rather than reading hundreds of pages online, but as the cost, speed and quality of desktop printers improve, we should see a new era of freely available, high-quality online textbooks. There are lots of them online already, ready for printing.
All of this will be good for science. No one will be able to claim in a paper they didn't know about so and so's work and don't have access to it. It will be increasingly easy to do dilettante science without the backing of a huge academic institution.
People will be able to follow a field of science extremely easily. Cross-fertilization will become the obvious way to make progress.
I can't wait, and I want to make that happen.
Think about the following:
1- The original IE team was disbanded. The codebase has not been maintained seriously for *years*
2- Chances are this codebase is ugly and huge, and we all know that IE is very buggy and non-standard compliant.
3- Microsoft cannot allow IE to change every 3 months, and they cannot afford to start from scratch all over again.
4- Microsoft has a history of perverting standards to their own end.
5- Microsoft has lots of resources, but as we all know, putting more people on a late project makes it even later.
Microsoft has been resting on its laurels for a little too long.
Given all of the above, I'd say the next version of IE will simply be worse than the current one. For Microsoft, things are going to get worse before they get better, exactly how it went for Mozilla. Now however Mozilla has put its hard years behind, they are on a roll.
I wouldn't be surprised if the version of IE that will come with Longhorn will be so bad that people will switch en masse.
1- This is an online poll. About as useful as Slashdot's.
2- They had to remove Yoper Linux from first place to make it look as if they had remotely believable results.
3- They still compare results from year to year and try to infer conclusions.
Even with excellent methodology it is difficult to extract usable information from survey data (look at the vast majority of medical surveys. We still don't know for sure if mobile phones are dangerous and it took decades to persuade enough people than smoking was indeed deadly). In this case the results are useless and are a waste of time other than for entertainment value.
You are doing statistics on 3 events, essentially your margin of error is so huge that you can't infer any conclusion.
Both were risky endeavours ; however the parent's point is not completely moot. Apollo 1 was a new design and caused the death of 3 people on the ground. The rest of the missions went OK, even Apollo 13 who had massive systems failures, but enough redundancy built-in to make it back safely. Moreover you are not counting the Apollo predecessors which were somewhat similar in design, but simpler, and which didn't suffer any casualty.
In contrast, the SS exploded twice unexpectedly in the middle of the program. Apparently the SS is a complex system with design flaws no one really knows how to fix.
At any rate the SS is far deadlier than it was designed to be.
The terrorists have been hitting Western targets in their home countries for decades. It's actually much easier for them, nothing new here.
I doubt OBL and his friends had the resources to carry out two attacks of the magnitude of 9/11 anyway. Perhaps others have tried and failed, but somehow I believe that if a vast conspiracy of the same size as that attack had been foiled by US services everyone would have heard about it.
This is not to say that the US effort was for nothing, but rather that so far they have caught relatively small fry.
In general image understanding is equivalent to general AI. We won't get a CBIR system that works well before we get an AI that works and vice versa, because people expect to be able to match the *content* of the image they submit as template and not the general appearance of the image. The problem is then too unspecified.
Even in the restricted context of aeroplanes this is not a trivial problem. Someone in the list of replies submitted an image of a warthog (A-10) and got nonsensical results. Somehow the CBIR system would need to be able to infer a model of the A-10 from a given random 2D projection, and match it to the other 2D projections of the same A-10 model that they do have in the DB. This doesn't sound impossible but it is hard, and I suspect the Penn State people didn't do that. Instead they are probably matching on colour, texture, general appearance, etc.
This is not to say that CBIR is not a nice problem to apply new image processing/image analysis algorithms to, which are developed all the time.
It took a while to surpass 2.95.3 in performance, I think 3.1.x ; however if you compile a lot of C++. The 3.x series was much more ISO-C++ compliant.
The 2.95.x series is still unmatched for speed of compilation I believe.
Milk does help a little but not much. A baby's immunity mostly comes from the fact that they share their bloodstream with their mother's in the womb, and thus benefit from the mother's antibodies.
When babies are born their immune system has yet to produce antibodies. The mother's antibodies are slowly elimitated in the first 3-6 months of the life of the baby.
Then the young child starts becoming sick. Most new parents witness their baby's first cold, flu, ear infection etc at between 3 to 6 months of age.
Over the next year or so babies are often sick a lot, especially if they are in day care, and thus exposed to a lot of bugs. As they fall prey to a variety of illnesses they start developing their immune system, which doesn't fully mature until young adulthood.
Absolutely correct, compare Saudi Arabia with Iran, although Saudi Arabia had their very first elections last week.
A lot of criminals are either not very smart people (if you are smart there are much better ways to make a living) who think they will never get caught or desperate people who have essentially no choice. Hence a harsh punishment as deterrent simply doesn't work.
Witness the extremely tough sentences in south-east Asia for *carrying* drugs.
It doesn't require you to do any buy-in, just like proprietary and FOSS software can coexist on the same machine.
You can be a capitalist by day, as in be part of standard corporate America, and a commie by night, by doing some volunteer work for a local association.
You sir deserve a lot of mod points.
With this kind of logic one would eventually decide to kill everybody else, and then one would have to live with oneself.
Oil at $100 a barrel doesn't just mean that filling up your car is more expensive. Everything depends on oil, including growing and transporting your food.
The whole capitalist system is mostly a pyramid scheme that depends on growth. Expensive everything means less growth, more unemployment and potentially a nasty negative spiral when debt repaiments are not met, at the level of a whole country.
It can be very nasty. We will not run out of oil or out of oil replacements, this is not the problem.
Perhaps you are referring to a "crisis" as something as mild as what happened in the 70s when oil suddently became 3x as more expensive overnight. Oil was still available, it was just suddently not quite as cheap as before.
Now imagine that oil became suddenly 3x as expensive as it is now within the next 12 months. Would it be fun ? Is it impossible ?
This is not the worst that could happen. To bring us back to the middle ages would only take a small nuclear war. How does nuclear war between China and the US over oil sound to you ? Impossible again ? Within 20 years time China will be consuming as much oil as the US is consuming today.
One can only assume that China will have as aggressive a policy then as the US has now. I can't imagine things will go down smoothly.
You do know doom3 runs natively on Linux, right? I believe Cedega may be able to run half-life.