Well, it's true that science only knows facts. As does everyone else. However, science isn't just about what we know, but mostly about what we predict from what we know. That is, we search for patterns, and if a pattern seems to work reasonably well (i.e. we have tried it quite often and it worked on all our tests), we have a theory. Those theories are not facts, the only fact involved is that they describe the past observations well. It is a prediction that they will also describe future observations well, however that is something you cannot know, just expect. This expectation is the more reasonable, the more similar the situation is to those you already have observed.
For example, if you have seen many things fall down, and never have seen anything that raises, then a reasonable theory is that all things fall down. Now say, your experience is mostly on stones and bananas (where you have tested hundreds of them). Now if you find any new, unknown object, it's quite reasonable to expect that it will fall down as well. It's unreasonable to assume you know in advance that it will fall down. If it is a stone or a banana, then it's so likely that you'll again see falling it down that you can treat it almost like a fact. If it is another object, then you shouldn't be too sure. After all, up to now you only observed the pattern with stones and bananas. And indeed, if the object turns out to be a helium balloon, then you'll find that your expectation is not met. That is, you'll find that your theory that everything falls down is in its generality wrong. It is, however, not completely wrong, because it still accurately describes your experience with stones and bananas. So the old theory is not invalid, but only has a limited validity. Now you can look at things which fall down, and things which raise up, and try to find a new pattern which lets you predict if a thing falls down or raises. If you find that pattern (which turns out to be "things which have a greater mass density than air ("are heavier than air") fall, things which are "lighter than air" rise), then you have a new, better theory, which contains your old theory (all things fall down) as special case.
Ok, in this case, it's that most of the stars observed so far (which were the bright stars, because those are the stars you can easily see) were not alone. So the theory was "most stars are not alone" (although in this case, I wouldn't really call it a theory, more an assumption or hypothesis). Now, stars which are sufficiently different from those which were observed before are found to be mostly alone, and there are also more of them. So we find that the old theory's validity is limited to those stars we previously observed (i.e. the bright ones), and a new, better theory replaces it (brighter stars usually have companions, dimmer stars usually don't).
So the fact (most stars previously observed have companions) didn't change, only the conjecture (therefore it probably is true for all other stars).
Note that the working of science isn't really too different from the everyday way of thinking. The main difference is that in everyday life, we often take our conjectures as facts without ever questioning them, while in science, the questioning of theories is institutionalized.
Chances are with this level of technology, health concerns will start to be a moot point. Predisposed to being over weight or having cancer... Well why not just use gene therapy to fix that.
Sure, and if you have a fast disassembler, all software bugs can be immediatly fixed. Right?
Well, the hitchhiker knows how to deal with that situation:
Computer, if you don't open that exit hatch this moment, I shall go straight to your major data banks with a very large axe and give you a reprogramming you'll never forget. Is that clear?
If you are the copyright holder, you can re-release your code any time under any license you want. Now if every kernel contributor re-releases his code under GPLv3, then the whole kernel is re-released under GPLv3. Of course, the code you received under v2 would remain under v2; and if you have an old kernel containing code which isn't in the current one, and isn't relicensed, then you couldn't redistribute it under v3 either. Also I think GPLv2 and GPLv3 are not compatible, because e.g. the GPLv3 DRM stuff is an extra restriction, and GPLv2 explicitly disallows any additional restrictions, therefore you couldn't add that v2 code to the hypothetical v3-licensed kernel and distribute it.
IANAL however, and the above is just my understanding.
Well, the data shows that people already know quite well how to produce global warning and are already doing it, so there's no point in teaching them how to do it in a seminar.
Well, indeed there is an illness which is obviously caused by cell phones: The Phone Shouting Disease (PSD).
My theory: Cell phones affect the part of the brain controlling your lungs when speaking, causing an increased level of neural activity there. This results in speaking extraordinarily loud into the phone.
Well, since AFAIK the left ear signal is decoded on the right brain half and vice versa, I could imagine a cancer which reduces your ability to hear and/or understand to make you more comfortable to put the phone at the cancer side, so that your non-cancer side does the decoding.
Of course, there's also another possibility for the bias:
D. Most people use the phone on the right hand side (because most people are right-handed). For some reason completely unrelated to cell phones, there is more cancer on the right hand side than on the left hand side. Together, this means more cancer on the phone side.
Well, at the basic level your computer is also analog. It works with voltages, currents and charges, which are all perfectly analog magnitudes (Ok, charge can only come in integer multiples of the elementary charge, but then, the charges in your computer are still large enough that this doesn't really matter). Indeed, 0 and 1 are not represented by an exact voltage, but by a whole voltage range (i.e. everything which is below a certain voltage counts as 0, and everything which is above a certain other voltage counts as 1).
Instead of whitelisting only known safe file types (which is easier for them), surely they could blacklist known dangerous file types (which is harder).
I don't think blacklisting file types would have been the right solution. And I'm willing to bet that they didn't choose whitelisting because it's less work (whitelists have to be kept up-to-date as well), but because it's more secure.
However, I think the correct solution would be not to just filter the attachments, but to send a confirmation mail to the sender (e.g. "Your mail contains an attachment 'firmware.bin' which is of an unknown filetype. Did you really intend to send that file?") Now, if it's a virus, then you would not have attached the file yourself, so you surely would answer "no" and the attachment can be deleted. However, it you really intended to attach that (as in the case of the firmware), then you'll answer yes. Since a virus will surely not reply such a confirmation mail (after all, how should it know that it is one), it's safe. It even contains the CYA factor, because if you explicitly confirm a mail attachment which is/contains a virus, then it's clearly your fault, not the IT department's. Most probably this could be automated, thus also reducing the workload of the IT department.
However, glibc is licensed with a special exception to the GPL, so simply linking your program to glibc doesn't cause your program to be covered. I guess this will also hold when re-licensed with GPLv3.
Slashdot Mirror
Well, it's true that science only knows facts. As does everyone else. However, science isn't just about what we know, but mostly about what we predict from what we know. That is, we search for patterns, and if a pattern seems to work reasonably well (i.e. we have tried it quite often and it worked on all our tests), we have a theory. Those theories are not facts, the only fact involved is that they describe the past observations well. It is a prediction that they will also describe future observations well, however that is something you cannot know, just expect. This expectation is the more reasonable, the more similar the situation is to those you already have observed.
For example, if you have seen many things fall down, and never have seen anything that raises, then a reasonable theory is that all things fall down. Now say, your experience is mostly on stones and bananas (where you have tested hundreds of them). Now if you find any new, unknown object, it's quite reasonable to expect that it will fall down as well. It's unreasonable to assume you know in advance that it will fall down. If it is a stone or a banana, then it's so likely that you'll again see falling it down that you can treat it almost like a fact. If it is another object, then you shouldn't be too sure. After all, up to now you only observed the pattern with stones and bananas. And indeed, if the object turns out to be a helium balloon, then you'll find that your expectation is not met. That is, you'll find that your theory that everything falls down is in its generality wrong. It is, however, not completely wrong, because it still accurately describes your experience with stones and bananas. So the old theory is not invalid, but only has a limited validity. Now you can look at things which fall down, and things which raise up, and try to find a new pattern which lets you predict if a thing falls down or raises. If you find that pattern (which turns out to be "things which have a greater mass density than air ("are heavier than air") fall, things which are "lighter than air" rise), then you have a new, better theory, which contains your old theory (all things fall down) as special case.
Ok, in this case, it's that most of the stars observed so far (which were the bright stars, because those are the stars you can easily see) were not alone. So the theory was "most stars are not alone" (although in this case, I wouldn't really call it a theory, more an assumption or hypothesis). Now, stars which are sufficiently different from those which were observed before are found to be mostly alone, and there are also more of them. So we find that the old theory's validity is limited to those stars we previously observed (i.e. the bright ones), and a new, better theory replaces it (brighter stars usually have companions, dimmer stars usually don't).
So the fact (most stars previously observed have companions) didn't change, only the conjecture (therefore it probably is true for all other stars).
Note that the working of science isn't really too different from the everyday way of thinking. The main difference is that in everyday life, we often take our conjectures as facts without ever questioning them, while in science, the questioning of theories is institutionalized.
It does get dubbed, but usually in the same language. The result is called cover version.
Of course it comes with an EULA. What did you think those genes which the researchers claim to have no obvious function are for?
Sure, and if you have a fast disassembler, all software bugs can be immediatly fixed. Right?
Well, the hitchhiker knows how to deal with that situation:
Computer, if you don't open that exit hatch this moment, I shall go straight to your major data banks with a very large axe and give you a reprogramming you'll never forget. Is that clear?
If you are the copyright holder, you can re-release your code any time under any license you want. Now if every kernel contributor re-releases his code under GPLv3, then the whole kernel is re-released under GPLv3. Of course, the code you received under v2 would remain under v2; and if you have an old kernel containing code which isn't in the current one, and isn't relicensed, then you couldn't redistribute it under v3 either. Also I think GPLv2 and GPLv3 are not compatible, because e.g. the GPLv3 DRM stuff is an extra restriction, and GPLv2 explicitly disallows any additional restrictions, therefore you couldn't add that v2 code to the hypothetical v3-licensed kernel and distribute it.
IANAL however, and the above is just my understanding.
Wasn't that because of the CFC?
Well, the data shows that people already know quite well how to produce global warning and are already doing it, so there's no point in teaching them how to do it in a seminar.
Because that's the noodle-string theory.
Just add (setq suck nil) to your
Obviously now the trick of changing the color on subsection stories by just changing the initial part of the URL doesn't work any more ...
Maybe it is "leaked" intentionally by MS in order to keep the news on it going?
I, for one, welcome the classic opportunity.
I guess it's time to start PETR (People for the Ethical Treatment of Robots).
Well, I'm especially waiting for the U+0950-RAM for more esoteric computers :-)
Don't tell that to the military! :-)
Well, only if they get a return ticket. Remember how Australia was colonized? You could do the same with space ...
Well, indeed there is an illness which is obviously caused by cell phones: The Phone Shouting Disease (PSD).
:-)
My theory: Cell phones affect the part of the brain controlling your lungs when speaking, causing an increased level of neural activity there. This results in speaking extraordinarily loud into the phone.
A cure for PSD has not yet been found.
SCNR
Cell phones physically suck? So can I use then as vacuum cleaner?
Well, since AFAIK the left ear signal is decoded on the right brain half and vice versa, I could imagine a cancer which reduces your ability to hear and/or understand to make you more comfortable to put the phone at the cancer side, so that your non-cancer side does the decoding.
Of course, there's also another possibility for the bias:
D. Most people use the phone on the right hand side (because most people are right-handed). For some reason completely unrelated to cell phones, there is more cancer on the right hand side than on the left hand side. Together, this means more cancer on the phone side.
Well, maybe that's intentional, to reduce unemployment ...
:-)
</conspiracy-theory>
Since a good number of bombs are triggered by clocks, maybe we should forbid private clocks.
Well, at the basic level your computer is also analog. It works with voltages, currents and charges, which are all perfectly analog magnitudes (Ok, charge can only come in integer multiples of the elementary charge, but then, the charges in your computer are still large enough that this doesn't really matter). Indeed, 0 and 1 are not represented by an exact voltage, but by a whole voltage range (i.e. everything which is below a certain voltage counts as 0, and everything which is above a certain other voltage counts as 1).
I don't think blacklisting file types would have been the right solution. And I'm willing to bet that they didn't choose whitelisting because it's less work (whitelists have to be kept up-to-date as well), but because it's more secure.
However, I think the correct solution would be not to just filter the attachments, but to send a confirmation mail to the sender (e.g. "Your mail contains an attachment 'firmware.bin' which is of an unknown filetype. Did you really intend to send that file?") Now, if it's a virus, then you would not have attached the file yourself, so you surely would answer "no" and the attachment can be deleted. However, it you really intended to attach that (as in the case of the firmware), then you'll answer yes. Since a virus will surely not reply such a confirmation mail (after all, how should it know that it is one), it's safe. It even contains the CYA factor, because if you explicitly confirm a mail attachment which is/contains a virus, then it's clearly your fault, not the IT department's. Most probably this could be automated, thus also reducing the workload of the IT department.
However, glibc is licensed with a special exception to the GPL, so simply linking your program to glibc doesn't cause your program to be covered. I guess this will also hold when re-licensed with GPLv3.