Speaking of moving abroad, there are good incentives for it now.
For example, doing science in American university is hard.
Not because science is hard, but because the university expects you to spend all your work time on teaching students (which, with the low level of course materials nowadays, has usually little to do with ones research), sitting in commissions, etc.
The research is done mostly in spare time and by graduate students. Also, in some fields, postdocs can do research for 1-2 years, before they have to dedicate all their time to looking for a new job.
In contrast, in Canada for example, the salaries are smaller than in US and they have higher taxes, however, there are more grants and the required teaching is much less. Which attracts people who are more interested in science than making money.
From a point of view of recent PhDs, working in
US means spending a lot of time on things not relevant to ones research with the prospect of getting tenure being very uncertain, while accepting a position in Canada implies having time to do research, grant for travel and social benifits (like good schools for children).
I think you missed the point of the original article.
Pamela's complaint is not that Sun does not follow GPL in a legal (i.e. technical) sense, but rather Sun does not follow the spirit of GPL.
Read the other quotes - they make abundantly clear that Sun views GPL software as something to be exploited and that Sun wants to be in position of being the only supplier, very much like Microsoft is.
This is sure a nice strategy for a money-making company, but it does not play nice with customers or developers.
They mean in "about the right amounts predicted by the theory".
Keep in mind is that most of the problems in nuclear physics cannot be solved exactly and numeric solutions (i.e. brute force simulation) are considered good when the answers they give are of the same order of magnitude as experiment.
Note that you can still have precise measurement when the theory does not exist.
Also note, that repeating the measurement costs money and time and results in better precision.
I guess they stopped at the point when they were sure that results warrant claim that some nuclear process is going on.
What I am saying is that you have flexibility to define what your system is. And this is used to make the laws of thermodynamics balance.
Now the question is what do you mean when you say "system of life" ?
I don't know of any precise definition. The best one can try to do is to include the whole universe with the presumption that you are not missing anything. But then your total energy is infinite.
Also, I do not care whether life is outputting more energy than it receives. What I care about whether it can increase supply of energy avaiable to it!
I.e. I care about whether 10 years from now I would be able to do more, perhaps, by doing smarter things or, perhaps, by simply having access to more energy.
This was the whole point of my comments: the original article mentioned that if programmers are not compensated outright than 1st law of thermodynamics somehow implies they must be losing something.
I made the point that application of this law has a problem. Note that I said "Sentience and its products are manifestly quantum phenomena and so do NOT have to obey such law."
As another poster mentioned information is not described well by thermodynamics (but you can actually apply stochastical mechanics to it).
However, even if you look at the conventional energy available, you still have positive surplus in most cases. This is usually called "progress" - doing things smarter or simply on a larger scale.
From my point of view E=mcc is *NOT* the limit of the total energy, but rather a limit on the rate with which you can increase your output. Yes, a given kilogram of matter will not yield more than a certain amount of energy, but you can always get another kilogram and so on. On the other hand
you cannot reach more matter faster than (ct) cubed and so your increase in the output is limited.
I also feel there is a misunderstanding about application of physical laws. You can apply them
in a situation they were not designed for and they can break then. Finding out why they break and whether it is possible to make them work, perhaps with some modifications, is the business of Physics.
The very reason you do not expect the law to break in an ordinary situation is because it was extensively tested and any observed breakage likely indicates a flaw in the experiment.
(Or, if you are lucky, a new phenomenon).
Lastly, thermodynamics most definitely applies to open systems. See, eg, Hyperphysics webpages
P.S. Looks like I used up the word "point" too much.. Sorry - its late night and I have more problems to solve..
These reactors produce energy by accelerating nuclear fission of atoms. This increases the amount of energy released in a given time interval.
What is specific to breeder reactors is that as a byproduct of their process they produce new fuel.
So, if you look at the Earth as a whole and subtract energy delivered by the Sun, as well as the energy produced by nuclear fission within Earth core you will see a small (compared to the components we just subtracted) excess.
Unlike regular fission reactors, you cannot say that breeder reactors use up a limited natural resource.
More abstractly: the first law of thermodynamics will (formally) fail if you try to apply it to a system with access to an unlimited supply of energy, at least compared to the lifetime of the system.
I.e. the first law of thermodynamics states that
change in the internal energy is equal to heat added to the system minus work done by the system.
This can break down in a situation where internal supply of energy is unlimited (i.e. breeder reactors, nuclear fusion, I would like to include
creativity in this list) so that the change in internal energy is not well defined.
Of course, when you do analysis of such a system you would still like to be able to use thermodynamics, so you need to patch up first law somehow.
One way to accomplish this is to isolate the source of unlimited energy and characterize it.
Then the rest of the system will obey the 1st law of thermodynamics. For example, for breeder reactors you can learn how much energy a given atom can produce and how easy it is to convert normally stable atoms into fissionable ones. This results in relatively simple equations that describe influx of energy into the reactor due to fission. Of course, doing the same thing with creativity is a lot more difficult.
The usual examples involve small-scale systems and open systems where influx of external energy is possible.
Imagine, for a moment, a pyramid scheme whereby
a person borrows $5 dollars from two of his friends, but when asked for same says: I won't give anything to you cause I just borrowed them and I need them, however you can do the same.
So they do the same and so on.
Of course, you will now note that the number of people is finite and this is going to stop when a person realizes that all his friends already borrowed money and are not giving more than once !
Now imagine the very same pyramid scheme, but assign a very large time for each transaction - say 35 years. Then, by the time you reach 10th layer of people in the pyramid scheme there would be 10 generations of new-born people !
If the population increases it can well be that the process can continue to infinity - or at least long enough until we reach a stage where our current knowledge loses predictive power.
(Btw, this is how Social Security works - the taxes are not put into a bank account, but paid out to the people who need Social security now, with the idea that future generations will pay people who paid Social security today.)
The same can be done with energy - you can have a mathematical model that slightly disbalances the equations so that bits of energy are brought closer to the origin. If the total energy of the entire system is infinite it can very well happen that in any bounded domain the energy increases because of the influx of it from outside and the influx does not stop.
A usual way to deal with such situation is to simply account for the extra energy by declaring the bounded domain an "open" system and have some sort of equation for the influx of energy.
This would be the approach, for example, to describe an electrical motor - electrical energy comes in and the motor produces mechanical energy.
This example is actually quite good, since you would not want to go down the chain to find "who produced the electrical energy". Because that can turn out to be hydro-plant, which is powered by river which is powered by Earth, which is powered by the Sun - and your motor will need to be rebuilt many times before the Sun will run out of fuel. (And then again, who knows, maybe there would be a way to refuel it as well).
Let me reply to them without quoting as it would take much space otherwise.
1. First of all, I want to describe more carefully
what I mean when I say "choice" as long as this
discussion is concerned.
It is most definitely NOTif(a==1)b=0;else b=2;. This I would call
a "decision".
Thus a decision does not have to be based on choice.
What is choice then ? One might imagine a situation when two possibilities exist, with no reason to always choose one for another. For example, suppose you are asked to pick a color - red or blue. So this can be called a choice.
Now a more interesting, and tricky, situation is when there is a preference - but you cannot determine it well. Then you are still making a choice.
Still, this is unsatisfactory, as there are decisions that would satisfy the same description.
So I make yet another attempt - a choice is something that reflects your identity. Something unique to yourself. As long as we are quoting quantum mechanics, I would say that the choice is related to identity in the same way as gauge fields are related to fermions.
2. Now let's tackle the question of determinism.
To discuss this I would need the notion of "transducer".
In engineering literature a transducer is a device that converts
something to something - for example air pressure into voltage.
Any real-world transducer is somewhat imprecise, insensitive and noisy - the output will vary slightly even in identical situations.
A deterministic machine is an idealized transducer - one that has no noise whatsoever, albeit the dependence of the output on the input can be very complicated.
3.Now I want to claim that the closer a given transducer is to an ideal one the harder it is to observe if one has only the access to inputs and outputs.
Indeed, in order to conclude that there is a transducer between input and output they must be
identified. But, if the relationship is perfect and has no noise, one can always suspect that there is no transducer and we are simply dealing with some sort of mathematical law.
In fact, this situation happens - with physical laws. Consider Maxwell equations - they are true beyond measurement errors on a wide range of scales. If we were restricted to those scales we can well conclude that the quantities they connect are the same !
For example, we would dispense with the concept of a medium that carries E-M waves (ether) and simply state that
E-M waves propagate in empty space (vacuum). Of course, if one reaches the level of quantum mechanics then one finds out that vacuum is not really empty, albeit what it actually is is still under study.
4.
Thus, a decision that is very close to being deterministic (from the subjects point of view) will have the property that the subject is not aware of it. A good example is a reflex - if a doctor hits you on the knee you jerk back. This "decision" is not perfect - as you can try to control it somewhat.
The if-then situation, even when executed in your head is more complicated and I will only mention that it is still possible to work out which parts of this process are almost deterministic (and invisible without much effort)
and which are based on choice). One note though: it is not enough to consider just the if-then alone but also the context surrounding it.
5.Now to self-awareness - from the point of view of exposition above this becomes a tautology - either a process is close to deterministic (and very hard to observe) or it is
based on choice and you are expressing part of your identity to make it.
A subject that does not observe its own decision making is not self-aware (as there is no awareness to speak of) and the subject that does observe it makes a choice.
It is curious to note that the measurement limit
in these matters is the fact that we cannot observe other peoples decision ma
First of all, let me point out that I feel I am qualified to talk about such matters.
Secondly, I would be the first to agree with you
that an argument of such nature will unlikely be completely rigorous when formulated in as few words as my original post has contained and is best to be considered as intuition.
Therefore, while I applaud your critical thinking, I feel that you could make a better argument by considering whether there an easy
answer to the deficiencies you found, because if there is (which is my point of view right now) your argument loses its strength. By the way, I am curious - why do you think it is a negative sum ?
Lastly, you have raise a very interesting point about open systems which I would like to comment upon.
I would like to point out that, according to most modern theories, the universe we are in is larger then region we can observe (due to the fact that speed of light is finite). So it is reasonable to consider universe as an open system.
Earth is not a closed system due to the influx of energy from Sun.
The Solar system can be considered closed for many purposes - as the influence of our galaxy and other objects is, perhaps, quite small and spans large timescales.
However, there is considerable influence that astronomical phenomena exert on scientific progress, in no small part because some natural phenomena simply can not be duplicated in laboratory, no matter how large. So the Solar system is not closed from this point of view.
The very inevitable bleakness of such conclusion suggests there is a flaw in the argument (as most science is usually neutral whenever emotions are concerned).
You contradict yourself. Either science is neutral "whenver emotions are concerned" or "inevitable bleakness" of a given POV somehow invalidates an argument. You may argue all you like that the First Law does not apply to software, but this is a pretty poor way to do that.
There is no contradiction. The argument is that usually a scientific conclusion will usually not provide conclusions that are either bleak or bright, but will usually have two sides.
The fact that the argument does pick a particular side is an intuition that something might be amiss.
Well, you see, the classical phenomena implies determinism and thus pretty much excludes a concept of choice from the theory.
Now, this is fine as long as you are modeling an external system - the same behavior can often be produced with and without choice. (Good example:
Big Blue versus Kasparov).
However, sentient beings have the property that they are self-aware - i.e. they observe themselves making a choice.
This is important, as while you can always doubt
whether someone else is sentient, it is unlikely
you will ever doubt that you are.
The only part of physics where we encounter choice nowadays is quantum mechanics, so I used
"quantum phenomena" to label the process.
Now, it is true that one could argue that, perhaps,
sentience is something beyond quantum mechanics and requires a different theory. However, this point is moot as quantum mechanics is not really complete yet and, in particular, is not a self-contained theory (usual expositions draw on
classical phenomena to construct models and the attempt to construct a theory from first principles are not mathematically rigorous).
It is curious to ponder the implications of this argument. In particular, suppose we assume that
brains are quantum computers - what does this imply about new ways we can use them ?
"As for Richard Stallman's "Free but shackled: The Java trap," it's hard to know where to begin. He has his own rather peculiar definition of "free" that I think violates the First Law of Thermodynamics (energy is conserved)"
Such invocation of energy conservation law is, unfortunately, a common fallacy.
Before explaining it away, let me formulate a counter-example:
This law implies that everything is a zero-sum game. If it was literally true (and not only for
select systems) it would imply that life
is a zero-sum game - whatever you gain must be torn from something else and in the end everything turns to dust. The very inevitable
bleakness of such conclusion suggests there is a
flaw in the argument (as most science is usually
neutral whenever emotions are concerned).
In fact, there are at least two flaws:
The energy is NOT conserved in open systems (which are most systems we ever deal with, closed system is an approximation)
The energy is conserved on average,
and can be "borrowed from vacuum" whenever quantum effects show up.
Sentience and its products are manifestly quantum phenomena and so do NOT have to obey such law. This is actually good news: it means there is a way to contribute to Open Source, without
sacrificing anything and without requiring
immediate payback.
As I understand it, Cray (company) has always been focusing on "capability" solutions, that is
a situation where a buyer wants the ability to perform certain task much more than they want to save money. (A good example would be NSA wanting specialized hardware for cracking).
Thus, they usually sell few systems (compared to
the multitude of Linux clusters being built) and they can afford to develop specialized hardware
to boost performance in a particular area.
This almost automatically implies being too expensive for majority of places that use Linux clusters - whether a lot of performance is required or not.
I once found on Canon's website a manual on how to clean an digital SLR sensor.
Basically a digital SLR is a body+electronics+sensor and you can screw on your own lens. Note that while professional camera bodies are expensive, good lenses are also very expensive, so it makes sense to keep your lenses when you change body to a new camera. And, of course, if you are going to change the sensor you need to change the electronics too - which leaves the metal case which is not that expensive (look at regular film SLRs - they go for around $150-300 nowadays).
Now back to cleaning: the manual said something like this:
do not clean if you don't need it.
purchase a special brush from us
carefully unscrew the lens.
carefully unpack the brush
swipe with the brush ONCE
throw the brush away.
do not repeat often
What happens is that the CMOS sensor is bare and anything harsher will likely damage it. So you really want it sealed - which implies having something transparent (like a lens;) in front of it that you can't remove.
Good lens are expensive so you can just as well make it a part of the lens system to reduce the price.
ATI does provide specifications - under NDA, but it allows publishing open source code.
The problem with current hardware is that it became very complex, so one needs much more time per single session of coding to achieve something useful. Long stretches of spare time are hard to come by.
Also, I want to correct you - it was Weather Channel that funded development of 3d DRI driver for radeon 8500 and 9200 cards, not ATI.
(see Tungsten Graphics)
Also ATI does provide sample cards to developers
which is big help (hard to find bugs in drivers without testing on real hardware).
Actually it is even more complicated because the meaning of "best estimate" depends strongly on the problem.
You see estimating probability from prior events always results in some error - you may overestimate it or underestimate it.
If you are relatively insensitive to the direction of the error than it is usually best to use maximum likelihood estimator - which is indeed the prior rate. This can be used, for example, to analyze the distribution of craters on Mars.
However, in our case (Earth) underestimating probability is much worse than overestimating it. A typical way to deal with this is to construct a risk function that depends on existing data and your future probability estimate and then try to minimize it.
As you can expect in our case this results in the probability estimate being larger than the observed rate, so it is not impossible that the estimate is 50% in next 10 years, even though no large asteroids or comets collided with Earth in the previous 95 years (remember Tunguska ?).
Logitech used to make USB Trackballs that had red leds shining through.. Very cool - pity they recently stopped making those in favor of a cheaper model.
Actually there is a good reason to send these notification e-mails: it could be that a legitimate attachment was misidentified as a virus.
In this case I want the sender to receive automated reply so they know that something is amiss.
And, of course, people who send.doc files as attachments are more likely to assume that all e-mail they send out is eventually read and that none of it could be filtered out.
Thus, the notification e-mails serve a purpose, they are pretty easy to filter out and, if done properly, they are a lot shorter than the virus e-mail and add negligible load.
This is actually how it was and still is. People with gold medals or scientific competition accomplishments can get in easier - or even bypass the exams completely. Each university conducts its own exams, so you rarely can apply to more than two.
So there are ranks of people in the admissions process:
Those who are admitted right after applying, with no exams
Those who have to take less exams (often just one) and if they get an A are admitted right away.
Those who have to take all the exams.
The last two are sorted by how well they did.
And, oh yeah, nowadays they have a new category:
those who pay with hard currency.
Slashdot Mirror
For example, doing science in American university is hard.
Not because science is hard, but because the university expects you to spend all your work time on teaching students (which, with the low level of course materials nowadays, has usually little to do with ones research), sitting in commissions, etc.
The research is done mostly in spare time and by graduate students. Also, in some fields, postdocs can do research for 1-2 years, before they have to dedicate all their time to looking for a new job.
In contrast, in Canada for example, the salaries are smaller than in US and they have higher taxes, however, there are more grants and the required teaching is much less. Which attracts people who are more interested in science than making money.
From a point of view of recent PhDs, working in US means spending a lot of time on things not relevant to ones research with the prospect of getting tenure being very uncertain, while accepting a position in Canada implies having time to do research, grant for travel and social benifits (like good schools for children).
Pamela's complaint is not that Sun does not follow GPL in a legal (i.e. technical) sense, but rather Sun does not follow the spirit of GPL.
Read the other quotes - they make abundantly clear that Sun views GPL software as something to be exploited and that Sun wants to be in position of being the only supplier, very much like Microsoft is.
This is sure a nice strategy for a money-making company, but it does not play nice with customers or developers.
Keep in mind is that most of the problems in nuclear physics cannot be solved exactly and numeric solutions (i.e. brute force simulation) are considered good when the answers they give are of the same order of magnitude as experiment.
Note that you can still have precise measurement when the theory does not exist.
Also note, that repeating the measurement costs money and time and results in better precision. I guess they stopped at the point when they were sure that results warrant claim that some nuclear process is going on.
What I am saying is that you have flexibility to define what your system is. And this is used to make the laws of thermodynamics balance.
Now the question is what do you mean when you say "system of life" ?
I don't know of any precise definition. The best one can try to do is to include the whole universe with the presumption that you are not missing anything. But then your total energy is infinite.
Also, I do not care whether life is outputting more energy than it receives. What I care about whether it can increase supply of energy avaiable to it!
I.e. I care about whether 10 years from now I would be able to do more, perhaps, by doing smarter things or, perhaps, by simply having access to more energy.
This was the whole point of my comments: the original article mentioned that if programmers are not compensated outright than 1st law of thermodynamics somehow implies they must be losing something.
I made the point that application of this law has a problem. Note that I said "Sentience and its products are manifestly quantum phenomena and so do NOT have to obey such law."
As another poster mentioned information is not described well by thermodynamics (but you can actually apply stochastical mechanics to it). However, even if you look at the conventional energy available, you still have positive surplus in most cases. This is usually called "progress" - doing things smarter or simply on a larger scale.
From my point of view E=mcc is *NOT* the limit of the total energy, but rather a limit on the rate with which you can increase your output. Yes, a given kilogram of matter will not yield more than a certain amount of energy, but you can always get another kilogram and so on. On the other hand you cannot reach more matter faster than (ct) cubed and so your increase in the output is limited.
I also feel there is a misunderstanding about application of physical laws. You can apply them in a situation they were not designed for and they can break then. Finding out why they break and whether it is possible to make them work, perhaps with some modifications, is the business of Physics.
The very reason you do not expect the law to break in an ordinary situation is because it was extensively tested and any observed breakage likely indicates a flaw in the experiment. (Or, if you are lucky, a new phenomenon).
Lastly, thermodynamics most definitely applies to open systems. See, eg, Hyperphysics webpages
P.S. Looks like I used up the word "point" too much.. Sorry - its late night and I have more problems to solve..
Yes, you are right, I did not focus on life specifically. Let me give one more example.
Consider nuclear fission reactors. Specifically, breeder reactors.
These reactors produce energy by accelerating nuclear fission of atoms. This increases the amount of energy released in a given time interval.
What is specific to breeder reactors is that as a byproduct of their process they produce new fuel. So, if you look at the Earth as a whole and subtract energy delivered by the Sun, as well as the energy produced by nuclear fission within Earth core you will see a small (compared to the components we just subtracted) excess.
Unlike regular fission reactors, you cannot say that breeder reactors use up a limited natural resource.
More abstractly: the first law of thermodynamics will (formally) fail if you try to apply it to a system with access to an unlimited supply of energy, at least compared to the lifetime of the system.
I.e. the first law of thermodynamics states that change in the internal energy is equal to heat added to the system minus work done by the system.
This can break down in a situation where internal supply of energy is unlimited (i.e. breeder reactors, nuclear fusion, I would like to include creativity in this list) so that the change in internal energy is not well defined.
Of course, when you do analysis of such a system you would still like to be able to use thermodynamics, so you need to patch up first law somehow.
One way to accomplish this is to isolate the source of unlimited energy and characterize it. Then the rest of the system will obey the 1st law of thermodynamics. For example, for breeder reactors you can learn how much energy a given atom can produce and how easy it is to convert normally stable atoms into fissionable ones. This results in relatively simple equations that describe influx of energy into the reactor due to fission. Of course, doing the same thing with creativity is a lot more difficult.
Imagine, for a moment, a pyramid scheme whereby a person borrows $5 dollars from two of his friends, but when asked for same says: I won't give anything to you cause I just borrowed them and I need them, however you can do the same.
So they do the same and so on.
Of course, you will now note that the number of people is finite and this is going to stop when a person realizes that all his friends already borrowed money and are not giving more than once !
Now imagine the very same pyramid scheme, but assign a very large time for each transaction - say 35 years. Then, by the time you reach 10th layer of people in the pyramid scheme there would be 10 generations of new-born people !
If the population increases it can well be that the process can continue to infinity - or at least long enough until we reach a stage where our current knowledge loses predictive power.
(Btw, this is how Social Security works - the taxes are not put into a bank account, but paid out to the people who need Social security now, with the idea that future generations will pay people who paid Social security today.)
The same can be done with energy - you can have a mathematical model that slightly disbalances the equations so that bits of energy are brought closer to the origin. If the total energy of the entire system is infinite it can very well happen that in any bounded domain the energy increases because of the influx of it from outside and the influx does not stop.
A usual way to deal with such situation is to simply account for the extra energy by declaring the bounded domain an "open" system and have some sort of equation for the influx of energy.
This would be the approach, for example, to describe an electrical motor - electrical energy comes in and the motor produces mechanical energy.
This example is actually quite good, since you would not want to go down the chain to find "who produced the electrical energy". Because that can turn out to be hydro-plant, which is powered by river which is powered by Earth, which is powered by the Sun - and your motor will need to be rebuilt many times before the Sun will run out of fuel. (And then again, who knows, maybe there would be a way to refuel it as well).
Let me reply to them without quoting as it would take much space otherwise.
1. First of all, I want to describe more carefully what I mean when I say "choice" as long as this discussion is concerned. It is most definitely NOT if(a==1)b=0;else b=2;. This I would call a "decision".
Thus a decision does not have to be based on choice.
What is choice then ? One might imagine a situation when two possibilities exist, with no reason to always choose one for another. For example, suppose you are asked to pick a color - red or blue. So this can be called a choice.
Now a more interesting, and tricky, situation is when there is a preference - but you cannot determine it well. Then you are still making a choice.
Still, this is unsatisfactory, as there are decisions that would satisfy the same description.
So I make yet another attempt - a choice is something that reflects your identity. Something unique to yourself. As long as we are quoting quantum mechanics, I would say that the choice is related to identity in the same way as gauge fields are related to fermions.
2. Now let's tackle the question of determinism.
To discuss this I would need the notion of "transducer".
In engineering literature a transducer is a device that converts something to something - for example air pressure into voltage.
Any real-world transducer is somewhat imprecise, insensitive and noisy - the output will vary slightly even in identical situations.
A deterministic machine is an idealized transducer - one that has no noise whatsoever, albeit the dependence of the output on the input can be very complicated.
3.Now I want to claim that the closer a given transducer is to an ideal one the harder it is to observe if one has only the access to inputs and outputs.
Indeed, in order to conclude that there is a transducer between input and output they must be identified. But, if the relationship is perfect and has no noise, one can always suspect that there is no transducer and we are simply dealing with some sort of mathematical law.
In fact, this situation happens - with physical laws. Consider Maxwell equations - they are true beyond measurement errors on a wide range of scales. If we were restricted to those scales we can well conclude that the quantities they connect are the same !
For example, we would dispense with the concept of a medium that carries E-M waves (ether) and simply state that E-M waves propagate in empty space (vacuum). Of course, if one reaches the level of quantum mechanics then one finds out that vacuum is not really empty, albeit what it actually is is still under study.
4. Thus, a decision that is very close to being deterministic (from the subjects point of view) will have the property that the subject is not aware of it. A good example is a reflex - if a doctor hits you on the knee you jerk back. This "decision" is not perfect - as you can try to control it somewhat.
The if-then situation, even when executed in your head is more complicated and I will only mention that it is still possible to work out which parts of this process are almost deterministic (and invisible without much effort) and which are based on choice). One note though: it is not enough to consider just the if-then alone but also the context surrounding it.
5.Now to self-awareness - from the point of view of exposition above this becomes a tautology - either a process is close to deterministic (and very hard to observe) or it is based on choice and you are expressing part of your identity to make it.
A subject that does not observe its own decision making is not self-aware (as there is no awareness to speak of) and the subject that does observe it makes a choice.
It is curious to note that the measurement limit in these matters is the fact that we cannot observe other peoples decision ma
Secondly, I would be the first to agree with you that an argument of such nature will unlikely be completely rigorous when formulated in as few words as my original post has contained and is best to be considered as intuition.
Therefore, while I applaud your critical thinking, I feel that you could make a better argument by considering whether there an easy answer to the deficiencies you found, because if there is (which is my point of view right now) your argument loses its strength. By the way, I am curious - why do you think it is a negative sum ?
Lastly, you have raise a very interesting point about open systems which I would like to comment upon.
I would like to point out that, according to most modern theories, the universe we are in is larger then region we can observe (due to the fact that speed of light is finite). So it is reasonable to consider universe as an open system.
Earth is not a closed system due to the influx of energy from Sun.
The Solar system can be considered closed for many purposes - as the influence of our galaxy and other objects is, perhaps, quite small and spans large timescales.
However, there is considerable influence that astronomical phenomena exert on scientific progress, in no small part because some natural phenomena simply can not be duplicated in laboratory, no matter how large. So the Solar system is not closed from this point of view.
You contradict yourself. Either science is neutral "whenver emotions are concerned" or "inevitable bleakness" of a given POV somehow invalidates an argument. You may argue all you like that the First Law does not apply to software, but this is a pretty poor way to do that.
There is no contradiction. The argument is that usually a scientific conclusion will usually not provide conclusions that are either bleak or bright, but will usually have two sides.
The fact that the argument does pick a particular side is an intuition that something might be amiss.
Now, this is fine as long as you are modeling an external system - the same behavior can often be produced with and without choice. (Good example: Big Blue versus Kasparov).
However, sentient beings have the property that they are self-aware - i.e. they observe themselves making a choice.
This is important, as while you can always doubt whether someone else is sentient, it is unlikely you will ever doubt that you are.
The only part of physics where we encounter choice nowadays is quantum mechanics, so I used "quantum phenomena" to label the process.
Now, it is true that one could argue that, perhaps, sentience is something beyond quantum mechanics and requires a different theory. However, this point is moot as quantum mechanics is not really complete yet and, in particular, is not a self-contained theory (usual expositions draw on classical phenomena to construct models and the attempt to construct a theory from first principles are not mathematically rigorous).
It is curious to ponder the implications of this argument. In particular, suppose we assume that brains are quantum computers - what does this imply about new ways we can use them ?
Such invocation of energy conservation law is, unfortunately, a common fallacy.
Before explaining it away, let me formulate a counter-example:
This law implies that everything is a zero-sum game. If it was literally true (and not only for select systems) it would imply that life is a zero-sum game - whatever you gain must be torn from something else and in the end everything turns to dust. The very inevitable bleakness of such conclusion suggests there is a flaw in the argument (as most science is usually neutral whenever emotions are concerned).
In fact, there are at least two flaws:
- The energy is NOT conserved in open systems (which are most systems we ever deal with, closed system is an approximation)
- The energy is conserved on average,
and can be "borrowed from vacuum" whenever quantum effects show up.
Sentience and its products are manifestly quantum phenomena and so do NOT have to obey such law. This is actually good news: it means there is a way to contribute to Open Source, without sacrificing anything and without requiring immediate payback.Thus, they usually sell few systems (compared to the multitude of Linux clusters being built) and they can afford to develop specialized hardware to boost performance in a particular area.
This almost automatically implies being too expensive for majority of places that use Linux clusters - whether a lot of performance is required or not.
I once found on Canon's website a manual on how to clean an digital SLR sensor.
Basically a digital SLR is a body+electronics+sensor and you can screw on your own lens. Note that while professional camera bodies are expensive, good lenses are also very expensive, so it makes sense to keep your lenses when you change body to a new camera. And, of course, if you are going to change the sensor you need to change the electronics too - which leaves the metal case which is not that expensive (look at regular film SLRs - they go for around $150-300 nowadays).
Now back to cleaning: the manual said something like this:
What happens is that the CMOS sensor is bare and anything harsher will likely damage it. So you really want it sealed - which implies having something transparent (like a lens ;) in front of it that you can't remove.
Good lens are expensive so you can just as well make it a part of the lens system to reduce the price.
The problem with current hardware is that it became very complex, so one needs much more time per single session of coding to achieve something useful. Long stretches of spare time are hard to come by.
Also, I want to correct you - it was Weather Channel that funded development of 3d DRI driver for radeon 8500 and 9200 cards, not ATI. (see Tungsten Graphics)
Also ATI does provide sample cards to developers which is big help (hard to find bugs in drivers without testing on real hardware).
You see estimating probability from prior events always results in some error - you may overestimate it or underestimate it.
If you are relatively insensitive to the direction of the error than it is usually best to use maximum likelihood estimator - which is indeed the prior rate. This can be used, for example, to analyze the distribution of craters on Mars.
However, in our case (Earth) underestimating probability is much worse than overestimating it. A typical way to deal with this is to construct a risk function that depends on existing data and your future probability estimate and then try to minimize it.
As you can expect in our case this results in the probability estimate being larger than the observed rate, so it is not impossible that the estimate is 50% in next 10 years, even though no large asteroids or comets collided with Earth in the previous 95 years (remember Tunguska ?).
It is also usually the case in such projects that the core team is defined as people that actively work on a project - and not by a list.
Few people can work full-time on a project and so the activity goes up and down for each individual developer.
The XFree86 core team was actually something you could be "a member in", with special privileges
(like direct CVS access and special mailing list).
Correction:
AIW 9700 Pro is not supported yet by GATOS drivers.. work is just starting.
Small correction:
ordinary matter also consists of leptons (i.e. electrons) which are similar to quarks, but have integer charge.
Logitech used to make USB Trackballs that had red leds shining through.. Very cool - pity they recently stopped making those in favor of a cheaper model.
In this case I want the sender to receive automated reply so they know that something is amiss.
And, of course, people who send
Thus, the notification e-mails serve a purpose, they are pretty easy to filter out and, if done properly, they are a lot shorter than the virus e-mail and add negligible load.
.. to have been reading a lot of Nigerian spam lately.
This is actually how it was and still is. People with gold medals or scientific competition accomplishments can get in easier - or even bypass the exams completely. Each university conducts its own exams, so you rarely can apply to more than two.
So there are ranks of people in the admissions process:
- Those who are admitted right after applying, with no exams
- Those who have to take less exams (often just one) and if they get an A are admitted right away.
- Those who have to take all the exams.
The last two are sorted by how well they did.And, oh yeah, nowadays they have a new category: those who pay with hard currency.
We already know how to create the last two and - wait a minute - we know how to create souls too !
I just want to point out that proper credit is due to RMS -
who was the one who stood up to consiously prolong the golden age of sharing.