I heard something about this on the radio last night (wasn't paying full attention).
But they were talking about quantum key exchange. Assuming that they're then using a standard symmetric key to encrypt the link it's still theoretically breakable, just the key exchange that isn't.
If they're quantum encrypting all the data then that's pretty astonishing - they were talking about video-conferencing so they need a reasonable bit rate and the fidelity rate has to be above 5/6[1] otherwise the link might be vulnerable to a quantum cloning attack.
[1] Assuming the best attack is a universal quantum cloning machine. The maximum theoretical fidelity isn't known for most non-universal quantum cloning machines (but is trivially known for some - e.g. 3/4 for a naive measure and retransmit). I don't know whether it's possible to prove that the 5/6 is a sufficient lower bound on the fidelity.
As well as being able to use both heads in playback mode you could use them in record/playback, playback/erase and erase/record.
My system started with both machines with erased disks. It then started recording on machine 1. Once the disk was full it continued recording on machine 2. Once the second disk was full it returned to machine 1.
Additionally, as recording started on disk2, erasing started on disk1. This cycle uses one head on each machine and can, in theory, continue indefinitely. In order to avoid any dropped frames, I actually started recording on machine 2 a few seconds before machine 1 disk was full so the system could only be run for a finite (but very long) time before needing to stop and erase the disks to start again.
The other head on each machine was then used for playback.
NO, your VHS/Betamax player did not have this first, unless it could record the show and play it back at the same time, allowing you to watch different segments of the show while it kept recording.
I wrote software to do this with a couple of Pioneer magneto-optical laser video disk recorders back in about 1996 (I can't remember the exact date but beyond any doubt it was before October 1998).
My implementation actually only allowed a delayed playback (of up to 30 minutes) but it would be trivial to have had the playback non-linear. Also, with the setup I was using there was a limit of playback only over the last 30 minutes of broadcast video.
It wasn't for home use, in 1996 prices the equipment was 50,000GBP per recorder. I also think the recording was analogue rather than digital but I'm not 100% certain on that.
Your ISP can easily and automatically intercept self-signed certificates and present their own certificate, which you will gladly accept. Now you think that you are less secure than a verified certificate but still "somewhat secure". Technically, you are however protected as much as plain unencrypted HTTP.
This is true. However, if someone with that level of access is that determined then they'll have other means including spy beams on windows, breaking in and installing keyloggers etc.
The most plausible threat is a large scale wiretapping with the data all stored. Only later will someone decide that a particular connection is "interesting" and go back and (try to) read that traffic.
If the government is that desperate to eavesdrop on me then I don't really care. It's going to cost them a lot of money and they're going to have to keep everything secret otherwise I'm going to realize that there has been eavesdropping.
If the government (or ISP) wants to trawl through everybody's data then I want it to be hard for them. In particular, when they lose a few tapes of the traffic though my ISP I'd like my stuff to stay private. It's the same with TLS on SMTP. I deliver direct to MX (where it's allowed) and negotiate TLS wherever possible. Almost everybody uses self signed (or at least "unknown" CA signed certs.) I lose nothing at all by allowing that and I'd certainly gain nothing by rejecting TLS whenever I couldn't recognise the CA and falling back to normal SMTP.
Accepting Scientific fact in the modern world usually requires an amout of thought and analysis. Accepting an ideology requires no thought at all. Humans are basically lazy.
It's worse than that. Science doesn't have an answer for most of the questions people are asking.
"Will MMR cause my child to become autistic?" Science says "probably not" with a very high confidence. But it will never say "no". Should I take the (tiny) risk and immunize my child? Yes - but if everybody else immunizes their children then possibly no because if everyone else is immunized then the probability of them getting exposed to the diseases is also tiny.
"Do cycle helmets help prevent head injury?" You can find (case controlled) studies that "prove" they have an efficacy of MORE than 100%. The favoured statistic of the pro-compulsion brigade is that they prevent 85% of head injuries - from a study that even the original authors now accept was flawed! However, helmet wearing appears to be positively correlated with head injury risk at the population level and almost everywhere that has implemented mandatory helmet laws has seen an increase in head injury risk. (There was one province in Canada that didn't see an increase in risk or a reduction in cycling post law - however the law was never enforced and helmet wearing rate didn't change). Nobody knows why whole population studies get the opposite result to case controlled studies. Risk compensation, self selection, rotational injury due to larger head and mass of helmet have all been proposed. (There was recently a very small study in the UK that found that drivers pass on average about 10cm closer to a helmeted cyclist than an unhelmeted one - but a long blond wig made them give more space to the cyclist - the study would need repeating on a much larger scale before any robust conclusions could be drawn)
Similar issues with HRT - case controlled studies showed positive benefits. Whole population studies showed it was harmful. And now there are yet more studies showing that it might be beneficial to some women.
It doesn't matter. At 6000K there just aren't enough photons with shorter wavelength than those of visible light to add appreciably to the total energy.
Nice graph here of the energy at different frequencies and temperatures.
The sun approximates to a black body (pretty well)
At around 6000K most of the energy is going to be visible and IR.
That will be above the atmosphere. I've no idea what proportion of each wavelength gets through the atmosphere but I know that UV is mostly blocked (and a good job too - that was one of the worries about ozone depletion)
The spectrum of the Sun's solar radiation is close to that of a black body with a temperature of about 5,800 K. About half that lies in the visible short-wave part of the electromagnetic spectrum and the other half mostly in the near-infrared part. Some also lies in the ultraviolet part of the spectrum.[1] When ultraviolet radiation is not absorbed by the atmosphere or other protective coating, it can cause a change in human skin pigmentation.
Go on then. Prove me wrong. Show how operator++(int) can be written/compiled without needing to copy the class. But remember your code has to be correct whether it is called as:
j=i++; or i++;
but you cannot tell which method will be used to call the function when you are writing/compiling it.
Bonus marks for solving the halting problem and being able to prove that operator++(int) does the same as operator++() in the case where they are called as i++ and ++i respectively.
I'm talking about C++ here. Using ++i in place of i++ is not a small efficiency saving when i is a large class type. In extreme cases it can be thousands to millions of times quicker to write ++i than i++. (Ideally classes would never get that big but, unfortunately, in legacy systems that have been enhanced to support new requirements over the years, classes can, and do, grow to enormous sizes)
In order to protect against things like this - including things like iterating through a loop and, later, having someone dramatically increase the size of the class you are incrementing, C++ programmers, by default, will use ++i instead of i++. When i is an integer it doesn't make any difference, but it makes sense to adopt a coding style that doesn't artificially hobble people who want to maintain and update the code in the future.
I find it astonishing that people on/. are arguing this point. I'm originally a C programmer, and still do about half my coding in C. When writing C++ I have to make a conscious effort to remember to write ++i because i++ trips off the fingers, but it's generally accepted that, in C++ you write ++i in place of i++ where it makes no difference to the behaviour.
Much like I find the discussion about i=i++; that started it all off incredible. It's undefined behaviour; there is no room for further discussion about what it means.
So, enlighten me. I've done a fair amount of work on compilers, although not C++ compilers.
But I'm at a loss to know how, when a compiler sees,
C operator++(int) { C tmp = *this; ++*this; return *this; }
It can optimize it to
func:
++*this; (machine code to do this)
return;
Just because in C you can see the whole context and can optimize i++ to be as fast as ++i doesn't mean that that optimization is possible in C++ in general. And hence why C++ programmers prefer ++i over i++ where it has no affect on the behaviour of the code.
And I'll bet you the amount of savings you get by switching to prefix instead of postfix increment is so miniscule as to render the entire discussion academic in most circumstances.
Think about it.
C& operator++();
As compared to
C operator++(int);
The postfix cannot return by reference so for postfix increment you have to copy the entire class.
And the typical way of implementing them:
C& operator++() { *this=*this+1; return *this; } C operator++(int) { C tmp = *this; ++*this; return tmp; }
The compiler may be able to optimize away the second copy for the postfix ++ but it's unlikely that the first copy can be optimized away; particularly as the compiler will have to be able to deduce that ++i and i++ will have the same effect. In the most general case this problem is undecidable.
The prefix case is trivially optimized by hand if it turns out that the copy cannot be optimized away by the compiler and it is a bottleneck. Of course, in many cases operator++() is written so that it doesn't call operator+() at all. That may be a case of premature optimization.
But it's not premature optimization. It's good practice, and will, for the most part, match the style of other C++ programmers working on the project. In C, good programmers will adopt the style of the project because it really doesn't make any difference with modern compilers.
The rationale for picking ++i over i++ when it makes no difference to how the code behaves is that i++ will almost always be slower than ++i. Additionally, sometimes changes elsewhere in the code can turn a fast, i++ into a slow i++ without affecting ++i. Good programmers don't leave unnecessary traps for their fellow programmers.
It doesn't matter if it's compiler-dependent or not. The correct answer to that question is: "This code is badly written. It never makes sense to write i = i++. You probably mean i++."
It's even worse than that. It's undefined behaviour. Anything could happen, including corrupting memory in such a way that the code fails catastrophically three days later and takes out the LSE for a day.
Bonus marks for pointing out that while C programmers usually write i++, C++ programmers will often write ++i. Although it doesn't matter here, in C++ operating on a class, the prefix ++ operator is usually faster than the postfix ++ operator due to it being hard for the compiler to optimize away the creation of a temporary in the postfix case.
What is mass? Why is there mass? Why are there three families of leptons? Why does time have a direction? Why does entropy increase? Why is energy conserved? etc, etc, etc.
It's the most basic things in physics where many of the most interesting questions are still to be answered.:-)
In some places there was main exchange -> fiber -> roadside box exchange -> copper -> consumer.
In the early days of ADSL it wasn't physically possible to fit ADSL equipment into the roadside box exchange. I think this might now have changed BICBW.
because it's not the ice-caps melting that will cause flooding but the expansion of the water from increased temperature of the oceans.
The ice caps are a drop in the ocean compared to that!
No. The Greenland Ice Sheet (GIS), West Antarctic Ice Sheet (WAIS) and East Antarctic Ice Sheet (EAIS) will, if they melt, cause an 80m+ rise in sea level. IIUC they're expected to melt in the order I've listed them.
However, in the next century it's not expected that there will be significant melting of any of these three and we're probably looking at less than a metre of sea level rise from ice melting (and less than a metre from water expansion as well).
Note that it is expected that sea level rise will continue for hundreds of years after temperatures have stabilized due to increased CO2 because the ice will take that long to melt.
We also know that water vapor soaks up 25 times as much heat as CO2, and that there's a lot more of it, especially over the oceans. Of course, the Global Warming Industry doesn't mention this, because it would make people wonder how much effect CO2 really has, except over cold deserts.
Of course they do. It's just that water vapour is a feedback not a forcing.
If we had a magic wand and could remove every molecule of water vapour from the air it would be back within weeks (IIRC more than 60% recovered in 14 days with 99% recovery in two months - you can download the models and try it yourself if you're interested)
If we had a magic wand and could remove every molecule of CO2 from the air we would freeze. Temperatures would start to drop. Water vapour would start to condense out. Temperatures would drop more in a vicious feedback.
Eventually, vulcanism would release CO2 back into the atmosphere. Over the course of a few tens or hundreds of millenia we'd start to warm back up. Eventually, with the correct orbital forcings and solar cycle we'd enter another temperate era. But it could be hundreds of millions of years before we get there.
And the same thing happens if you add CO2 to the atmosphere. It causes a small increase in temperature which causes an increase in water vapour which causes a futher increase in temperature until eventually we reach a new equilibrium.
Historically, climate change has occurred over centuries to millenia. Water vapour reaches equilibrium so quickly that it cannot possibly be a cause of the climate change although it does amplify it.
Climate change is now occuring over years to decades, hundreds of times faster than it ever has before. Even now, water vapour is staying in equilibrium.
So yes, climate scientists do take account of water vapour, it's just that it's not a forcing so is irrelevant to the initial state of their model. Even if they get the water vapour completely wrong at the start of the model run, it will correct itself so fast that it cannot have an effect on climate, only on weather, and climate scientists aren't interested in the weather.
2008 is going to be the coolest year of the 21st century so far. A combination of an el nina and low sunspot activity. But it's still on target to be one of the warmest years on record. It ought to have been one of the coldest years on record.
If your financials are so complex that you are unable to audit them for evidence of fraud then you need to hire professional help.
I don't know about this guy, but one of the things you do is pay your bank to manage your accounts. I'd be astonished if this guy wasn't paying large sums for the privilege of this bank giving away his money to fraudsters.
There's a big difference between "your records don't agree with mine" and "this transaction, right there, was not authorized by me and does not appear in any of my records". This fellow did the former but apparently never got as far as the latter.
But you're pretty stuck anyway. The problem is that these accounts change in value by thousands or tens of thousands per day. By the time you get to your bank to query it there are dozens more transactions. You say to your bank that you think there's 20000 missing according to the balance on the 200 page statement you received that month. By the time you get the statement though the post there's another 50000 gone in and 25000 gone out. So the bank says "no, there's another 25k in the account." So now you decide that it must have been a cheque that you'd paid in that had, for some reason, taken a long time to appear on your balance.
If these accounts include investments (I don't know) then the account could be changing by tens of thousands per minute. And to balance your statement you'll have to know every share price at the time the statement was printed. And then put that price into your calculations, remove any unpresented cheques etc.
The saying "If you can count your money, you don't have a billion dollars" really is true. And that's not counting down to the last penny. That's counting in millions.
Slashdot Mirror
I heard something about this on the radio last night (wasn't paying full attention).
But they were talking about quantum key exchange. Assuming that they're then using a standard symmetric key to encrypt the link it's still theoretically breakable, just the key exchange that isn't.
If they're quantum encrypting all the data then that's pretty astonishing - they were talking about video-conferencing so they need a reasonable bit rate and the fidelity rate has to be above 5/6[1] otherwise the link might be vulnerable to a quantum cloning attack.
[1] Assuming the best attack is a universal quantum cloning machine. The maximum theoretical fidelity isn't known for most non-universal quantum cloning machines (but is trivially known for some - e.g. 3/4 for a naive measure and retransmit). I don't know whether it's possible to prove that the 5/6 is a sufficient lower bound on the fidelity.
Tim.
Found a link to the device I was using:
http://www.laserdiscarchive.co.uk/laserdisc_archive/pioneer/pioneer_vdr-v1000/pioneer_vdr-v1000.htm
So it was half the price I remembered.
As well as being able to use both heads in playback mode you could use them in record/playback, playback/erase and erase/record.
My system started with both machines with erased disks. It then started recording on machine 1. Once the disk was full it continued recording on machine 2. Once the second disk was full it returned to machine 1.
Additionally, as recording started on disk2, erasing started on disk1. This cycle uses one head on each machine and can, in theory, continue indefinitely. In order to avoid any dropped frames, I actually started recording on machine 2 a few seconds before machine 1 disk was full so the system could only be run for a finite (but very long) time before needing to stop and erase the disks to start again.
The other head on each machine was then used for playback.
Tim.
NO, your VHS/Betamax player did not have this first, unless it could record the show and play it back at the same time, allowing you to watch different segments of the show while it kept recording.
I wrote software to do this with a couple of Pioneer magneto-optical laser video disk recorders back in about 1996 (I can't remember the exact date but beyond any doubt it was before October 1998).
My implementation actually only allowed a delayed playback (of up to 30 minutes) but it would be trivial to have had the playback non-linear. Also, with the setup I was using there was a limit of playback only over the last 30 minutes of broadcast video.
It wasn't for home use, in 1996 prices the equipment was 50,000GBP per recorder. I also think the recording was analogue rather than digital but I'm not 100% certain on that.
Tim.
Your ISP can easily and automatically intercept self-signed certificates and present their own certificate, which you will gladly accept. Now you think that you are less secure than a verified certificate but still "somewhat secure". Technically, you are however protected as much as plain unencrypted HTTP.
This is true. However, if someone with that level of access is that determined then they'll have other means including spy beams on windows, breaking in and installing keyloggers etc.
The most plausible threat is a large scale wiretapping with the data all stored. Only later will someone decide that a particular connection is "interesting" and go back and (try to) read that traffic.
If the government is that desperate to eavesdrop on me then I don't really care. It's going to cost them a lot of money and they're going to have to keep everything secret otherwise I'm going to realize that there has been eavesdropping.
If the government (or ISP) wants to trawl through everybody's data then I want it to be hard for them. In particular, when they lose a few tapes of the traffic though my ISP I'd like my stuff to stay private. It's the same with TLS on SMTP. I deliver direct to MX (where it's allowed) and negotiate TLS wherever possible. Almost everybody uses self signed (or at least "unknown" CA signed certs.) I lose nothing at all by allowing that and I'd certainly gain nothing by rejecting TLS whenever I couldn't recognise the CA and falling back to normal SMTP.
Tim.
Accepting Scientific fact in the modern world usually requires an amout of thought and analysis. Accepting an ideology requires no thought at all. Humans are basically lazy.
It's worse than that. Science doesn't have an answer for most of the questions people are asking.
"Will MMR cause my child to become autistic?" Science says "probably not" with a very high confidence. But it will never say "no". Should I take the (tiny) risk and immunize my child? Yes - but if everybody else immunizes their children then possibly no because if everyone else is immunized then the probability of them getting exposed to the diseases is also tiny.
"Do cycle helmets help prevent head injury?" You can find (case controlled) studies that "prove" they have an efficacy of MORE than 100%. The favoured statistic of the pro-compulsion brigade is that they prevent 85% of head injuries - from a study that even the original authors now accept was flawed! However, helmet wearing appears to be positively correlated with head injury risk at the population level and almost everywhere that has implemented mandatory helmet laws has seen an increase in head injury risk. (There was one province in Canada that didn't see an increase in risk or a reduction in cycling post law - however the law was never enforced and helmet wearing rate didn't change). Nobody knows why whole population studies get the opposite result to case controlled studies. Risk compensation, self selection, rotational injury due to larger head and mass of helmet have all been proposed. (There was recently a very small study in the UK that found that drivers pass on average about 10cm closer to a helmeted cyclist than an unhelmeted one - but a long blond wig made them give more space to the cyclist - the study would need repeating on a much larger scale before any robust conclusions could be drawn)
Similar issues with HRT - case controlled studies showed positive benefits. Whole population studies showed it was harmful. And now there are yet more studies showing that it might be beneficial to some women.
Tim.
It doesn't matter. At 6000K there just aren't enough photons with shorter wavelength than those of visible light to add appreciably to the total energy.
Nice graph here of the energy at different frequencies and temperatures.
http://en.wikipedia.org/wiki/Planck's_law
The 5500K curve is approximately the sun, the 3500K curve is approximately an incandescent lamp
Tim.
The sun approximates to a black body (pretty well)
At around 6000K most of the energy is going to be visible and IR.
That will be above the atmosphere. I've no idea what proportion of each wavelength gets through the atmosphere but I know that UV is mostly blocked (and a good job too - that was one of the worries about ozone depletion)
Tim.
p.s. Just found this: http://en.wikipedia.org/wiki/Solar_radiation
The spectrum of the Sun's solar radiation is close to that of a black body with a temperature of about 5,800 K. About half that lies in the visible short-wave part of the electromagnetic spectrum and the other half mostly in the near-infrared part. Some also lies in the ultraviolet part of the spectrum.[1] When ultraviolet radiation is not absorbed by the atmosphere or other protective coating, it can cause a change in human skin pigmentation.
Go on then. Prove me wrong. Show how operator++(int) can be written/compiled without needing to copy the class. But remember your code has to be correct whether it is called as:
j=i++;
or
i++;
but you cannot tell which method will be used to call the function when you are writing/compiling it.
Bonus marks for solving the halting problem and being able to prove that operator++(int) does the same as operator++() in the case where they are called as i++ and ++i respectively.
Tim.
I'm talking about C++ here. Using ++i in place of i++ is not a small efficiency saving when i is a large class type. In extreme cases it can be thousands to millions of times quicker to write ++i than i++. (Ideally classes would never get that big but, unfortunately, in legacy systems that have been enhanced to support new requirements over the years, classes can, and do, grow to enormous sizes)
In order to protect against things like this - including things like iterating through a loop and, later, having someone dramatically increase the size of the class you are incrementing, C++ programmers, by default, will use ++i instead of i++. When i is an integer it doesn't make any difference, but it makes sense to adopt a coding style that doesn't artificially hobble people who want to maintain and update the code in the future.
I find it astonishing that people on /. are arguing this point. I'm originally a C programmer, and still do about half my coding in C. When writing C++ I have to make a conscious effort to remember to write ++i because i++ trips off the fingers, but it's generally accepted that, in C++ you write ++i in place of i++ where it makes no difference to the behaviour.
Much like I find the discussion about i=i++; that started it all off incredible. It's undefined behaviour; there is no room for further discussion about what it means.
Tim.
Modifying the same value twice without an intervening sequence point.
Of course that return *this; should be return tmp; (although the compiler will be equally happy with either)
Tim.
So, enlighten me. I've done a fair amount of work on compilers, although not C++ compilers.
But I'm at a loss to know how, when a compiler sees,
C operator++(int)
{
C tmp = *this;
++*this;
return *this;
}
It can optimize it to
func:
++*this; (machine code to do this)
return;
Just because in C you can see the whole context and can optimize i++ to be as fast as ++i doesn't mean that that optimization is possible in C++ in general. And hence why C++ programmers prefer ++i over i++ where it has no affect on the behaviour of the code.
Tim.
And I'll bet you the amount of savings you get by switching to prefix instead of postfix increment is so miniscule as to render the entire discussion academic in most circumstances.
Think about it.
C& operator++();
As compared to
C operator++(int);
The postfix cannot return by reference so for postfix increment you have to copy the entire class.
And the typical way of implementing them:
C& operator++() { *this=*this+1; return *this; }
C operator++(int) { C tmp = *this; ++*this; return tmp; }
The compiler may be able to optimize away the second copy for the postfix ++ but it's unlikely that the first copy can be optimized away; particularly as the compiler will have to be able to deduce that ++i and i++ will have the same effect. In the most general case this problem is undecidable.
The prefix case is trivially optimized by hand if it turns out that the copy cannot be optimized away by the compiler and it is a bottleneck. Of course, in many cases operator++() is written so that it doesn't call operator+() at all. That may be a case of premature optimization.
Tim.
"Premature optimization is the root of all evil"
-Dan Knuth
Dan? :-)
Tim.
But it's not premature optimization. It's good practice, and will, for the most part, match the style of other C++ programmers working on the project. In C, good programmers will adopt the style of the project because it really doesn't make any difference with modern compilers.
The rationale for picking ++i over i++ when it makes no difference to how the code behaves is that i++ will almost always be slower than ++i. Additionally, sometimes changes elsewhere in the code can turn a fast, i++ into a slow i++ without affecting ++i. Good programmers don't leave unnecessary traps for their fellow programmers.
Tim.
I'll cross the sea to effect this marriage.
I'll cross the sea to affect this marriage.
Henry VI (part 3). Well one of them is anyway. They mean rather different things.
Tim.
Why not get the compiler to do the work for you:
$ gcc -o test -Wall -W -ansi -pedantic test.c
test.c: In function `main':
test.c:5: warning: operation on `i' may be undefined
$
Lo and behold. It's undefined behaviour.
$
i = 0
$
And look. I don't get the same results as you. (Probably depends on the precise version of gcc and also maybe the optimizations selected.)
Tim.
It doesn't matter if it's compiler-dependent or not. The correct answer to that question is: "This code is badly written. It never makes sense to write i = i++. You probably mean i++."
It's even worse than that. It's undefined behaviour. Anything could happen, including corrupting memory in such a way that the code fails catastrophically three days later and takes out the LSE for a day.
Bonus marks for pointing out that while C programmers usually write i++, C++ programmers will often write ++i. Although it doesn't matter here, in C++ operating on a class, the prefix ++ operator is usually faster than the postfix ++ operator due to it being hard for the compiler to optimize away the creation of a temporary in the postfix case.
Tim.
What is mass? Why is there mass? Why are there three families of leptons? Why does time have a direction? Why does entropy increase? Why is energy conserved? etc, etc, etc.
It's the most basic things in physics where many of the most interesting questions are still to be answered. :-)
Tim.
In some places there was main exchange -> fiber -> roadside box exchange -> copper -> consumer.
In the early days of ADSL it wasn't physically possible to fit ADSL equipment into the roadside box exchange. I think this might now have changed BICBW.
Tim.
because it's not the ice-caps melting that will cause flooding but the expansion of the water from increased temperature of the oceans.
The ice caps are a drop in the ocean compared to that!
No. The Greenland Ice Sheet (GIS), West Antarctic Ice Sheet (WAIS) and East Antarctic Ice Sheet (EAIS) will, if they melt, cause an 80m+ rise in sea level. IIUC they're expected to melt in the order I've listed them.
However, in the next century it's not expected that there will be significant melting of any of these three and we're probably looking at less than a metre of sea level rise from ice melting (and less than a metre from water expansion as well).
Note that it is expected that sea level rise will continue for hundreds of years after temperatures have stabilized due to increased CO2 because the ice will take that long to melt.
Tim.
We also know that water vapor soaks up 25 times as much heat as CO2, and that there's a lot more of it, especially over the oceans. Of course, the Global Warming Industry doesn't mention this, because it would make people wonder how much effect CO2 really has, except over cold deserts.
Of course they do. It's just that water vapour is a feedback not a forcing.
If we had a magic wand and could remove every molecule of water vapour from the air it would be back within weeks (IIRC more than 60% recovered in 14 days with 99% recovery in two months - you can download the models and try it yourself if you're interested)
If we had a magic wand and could remove every molecule of CO2 from the air we would freeze. Temperatures would start to drop. Water vapour would start to condense out. Temperatures would drop more in a vicious feedback.
Eventually, vulcanism would release CO2 back into the atmosphere. Over the course of a few tens or hundreds of millenia we'd start to warm back up. Eventually, with the correct orbital forcings and solar cycle we'd enter another temperate era. But it could be hundreds of millions of years before we get there.
And the same thing happens if you add CO2 to the atmosphere. It causes a small increase in temperature which causes an increase in water vapour which causes a futher increase in temperature until eventually we reach a new equilibrium.
Historically, climate change has occurred over centuries to millenia. Water vapour reaches equilibrium so quickly that it cannot possibly be a cause of the climate change although it does amplify it.
Climate change is now occuring over years to decades, hundreds of times faster than it ever has before. Even now, water vapour is staying in equilibrium.
So yes, climate scientists do take account of water vapour, it's just that it's not a forcing so is irrelevant to the initial state of their model. Even if they get the water vapour completely wrong at the start of the model run, it will correct itself so fast that it cannot have an effect on climate, only on weather, and climate scientists aren't interested in the weather.
Tim.
2008 is going to be the coolest year of the 21st century so far. A combination of an el nina and low sunspot activity. But it's still on target to be one of the warmest years on record. It ought to have been one of the coldest years on record.
Tim.
If your financials are so complex that you are unable to audit them for evidence of fraud then you need to hire professional help.
I don't know about this guy, but one of the things you do is pay your bank to manage your accounts. I'd be astonished if this guy wasn't paying large sums for the privilege of this bank giving away his money to fraudsters.
Tim.
But you're pretty stuck anyway. The problem is that these accounts change in value by thousands or tens of thousands per day. By the time you get to your bank to query it there are dozens more transactions. You say to your bank that you think there's 20000 missing according to the balance on the 200 page statement you received that month. By the time you get the statement though the post there's another 50000 gone in and 25000 gone out. So the bank says "no, there's another 25k in the account." So now you decide that it must have been a cheque that you'd paid in that had, for some reason, taken a long time to appear on your balance.
If these accounts include investments (I don't know) then the account could be changing by tens of thousands per minute. And to balance your statement you'll have to know every share price at the time the statement was printed. And then put that price into your calculations, remove any unpresented cheques etc.
The saying "If you can count your money, you don't have a billion dollars" really is true. And that's not counting down to the last penny. That's counting in millions.
Tim.