lately I find myself sucked into the disinformation void.
I was hoping I might get some help fighting ignorance by better understanding a dualism that many ppl on the board tend to make reference to.
I speak of the classifications of science/pseudo science. How would one definate each?
As far as I can see, the only discrepency is veracity/number of scientists convinced, no?
Well, one good criteria is that if it can’t be rigorously tested, or if the test can’t be replicated by others, what you’ve got is probably pseudoscience.
Other warning signs are when the theory is so vague that any situation can be applied to it (a new-age remedy that cure everything from dandruff to cancer and AIDS, for example), as well as claims that explanations/data/methodology must be kept secret from ‘the conspiracy’.
This site has some good information on different types of pseudoscience.
http://www.skepdic.com/
As well as some definitions of just what pseudoscience is.
http://www.skepdic.com/pseudosc.html
The distinction is the amount and quality of the evidence supporting a theory. So one might indeed say that if more reputable and critical observers agree with a theory, it is more likely to be genuine than not.
An easy way to definate it is “science” is anything that conforms to the scientific method. (This is not a circular argument; the scientific method is well-defined.) Anything which claims to be science but isn’t really, is “pseudoscience”.
To be “scientific,” research must meet certain requirements. The two that I can think of off-hand are the Repeatability Requirement and the Falsifiability Requirement. Repeatablility means that other people can follow your procedures and get the same results. This was the problem with Cold Fusion: noone could ever duplicate the supposed fusion event. Falsifiability means, basically, that your theories can be tested to prove them true or false. Paranormal research is not a science because all of their theories about ghosts and stuff can’t be proven one way or the other. They’re simply beliefs. This requirement also disqualifies the “science” of archeology, since Prof. Johnson’s theories about the role of nearsighted transvestites in ancient Sumerian society can’t really be tested in any meaningful way.
And of course, there are things like Computer Science, which is really just a skill, like carpentry or plumbing.
It’s not some kind of metaphysical dualism, it’s simply a modifier meaning “false” to denote that something which is being touted as “science” does not stand up to close scrutiny. The criteria for what is, and is not, “science” can be very distinct or slippery slopeish depending on the "scientific fact or paradigm at issue.
“Science” as such is not simply a collection of opinions but a process which generally has certain citeria for validity such as -
If something is being touted as being “science” or “scientific” and it violates any or all of these these criteria one might well conclude it is false or “pseudo”-science.
There are two quantites that are present here. First, there is the question of whether a given hypothsis is scientific. The second is whether it is true.
A scientific hypothesis tends to have certain charteristics:
•Makes claims about the real world which can in principle be known
•Falsifiable - There is the possibility of finding evidence which would refute the hypothsis.
•Occam - The hypothesis is as “simple” (in some sense) as it can be while still accounting for all the facts.
•not ad hoc - the hypothsis came before the evidence was known
A question of whether a given hypothesis is scientific is a different question, however, from asking if it is true.
A true hypothesis will tend to have other charteristics:
•Method - investigated via the scientific method.
•Replicated - The same results have been found by other reseachers
•not falsified - There is not evidence which undermines the hypothesis
There are also proxy indicators which may lead a lay observer to have reasonable confidence a given hypothsis is true and scientific:
•Widely accepted by scientists
•Does not conflict with widely accepted science
•Results which confirm the hypothesis are published in peer reviewed publications
•There is no better confirmed hypothesis
There is, in my view, no magic bullet test which will always differentiate science from non-science, just as there is no magic bullet test to determine if something is true. Each of these criteria are neither necessary nor suffcient to gaurentee a given hypothesis is true or scientific. There are grey areas in both the label “true” and the label “scientific.” It is criteria such as these, coupled with plain old reason, which makes something scientific.
By reason, I mean to look at criteria which are relevant to the hypothesis. It is unreasonable to expect that an astronomical hypothesis about a supernova (a rare event) should be replicated by fifteen supernovae (sp?) before it belived. Simmilarly, the fact that a scientist revises a hypothesis in light of new data once is not as probematic as a revision every time new data arrives (but sometimes the revisions seem reasonable, sometimes they do not).
Answering your question preciely is perhaps one of the most difficult questions in the philosophy of science. There are a number of different answers people have posed, and each has it’s proponents and detractors.
Did this help? Probably not.
What can youo expect from a user with a handle of diceman?
A. You are confusing Computer Science with Computer Programming.
and
B. You are obviously ignorant of “The Art of Computer Programming.”
I have been fortunate to be immortalized in the annals of Computing for a mere 4 lines of code. Really, really good code. It appears in Operating Systems textbooks. No one else figured out how to solve the problem so well.
Furthermore, there’s a bunch of proofs my colleagues and I have come up with that specify how much of what resources are required for those kinds of problems. In some models, there are proofs of impossibility.
And that’s just one itty bitty branch. CS is one of the broadest Sciences. There’s a lot of engineering, some AI people are into Neurology, etc. It isn’t all programming. And even in programming there can be great achievements.
One of the most famous open problems in Science is the “P=NP?” question. Proving “not equal” (the majority opinion) probably would not involve any programming at all. Proving “equal” may or may not involve programming. And if it does, it will be a beauty.
Diceman, go here and look at the slides for an important “classic” result in CS. Go read the original paper. Then tell me it’s carpentry and plumbing.
However, it certainly is the case that Computer Science is not a science in exactly the same way that physics and chemistry are. While I also take exception to diceman’s characterization, I’m willing to grant him a little leeway on this point.
Does that also mean that theoretical mathematics (as an analog to programming) is not a “science” in the same way that the classic “hard” sciences like physics and chemistry are?
Yes. Theoretical mathematics doesn’t deal with the real world. We take a set of statements as true, and apply certain rules of inference to generate other true statements. That’s pretty much all it is. There are no questions of data, or falsifiability: the data is the theorems, and every statement is either false or true (even if we can’t know whether a given statement is false or true).
Programming is not a good example of a non-science. A lot of what you do from day to day consists of applying the scientific method to find program errors, and that is a scientific activity. Computer science is something else.
everything is theoretical.
A lot of scholars do not consider mathematics as a science for the fact that it cannot be falsified.
This is a slight hijack, but what is the P=NP problem?
As I understand it, P and NP properties that certain problems have, it is not a quantity. While P implies NP, NP may or may not imply P. What does it mean for P to be equal to NP?
Regarding CS and the OP, it may be useful to look at two ideas
•Deduction: Logically arriving at conclusions using known premsises
•Induction: Finding patterns in nature based on the idea that the past will be like the future.
Induction is usually considered science. Deduction is sometimes considered science, sometimes not. Certainly no one would argue that if the premises are known and the logic is sound then the conclusion could be anything but true. However, deduction can be used to investigate things which are not real, except in our heads. Mathematical objects can be contructed which do not correspond ot anything in the natural world. Any investigation of these objects would likewise be devoid of empirical content. Thus we can not rightly call it science, since it says nothing about the natural world.
This is, for what its worth, one argument that math is not a science.
Thus to establish CS as a science, perhaps looking at the proofs and derived results is the wrong place. I would look to the trial and error in say the quantum computing project. They are trying to discover what works and what doesn’t. They are, in other words, trying to make statements about the world, and test them against the world by experimentation. In that sense, I would argue it is a science.
Oops, forget my question
P=NP is P iff NP
I get it now. (I must be slow tonight or something)
cheers,
-Short
While this is one of the parts of the scientific method as learned in elementary school, is it totally applicable? Correct concepts have been created after the experiment on many occcasions, and aren’t any less science or less true for it.
Hahn’s hypothesis was that bombarding uranium with neutrons would create a heavier elements. After the experiment failed, Meitner took the evidence and created the new hypothesis of nuclear fission. (That’s just an example on my mind, I’m sure you could think of another thousand that follow the same idea.)
ftg: what four lines?
I believe that Short meant that the hypothesis came before all of the evidence was known. That is, some experiment had to be run after the hypothesis was posited, and the results of that experiment had to support the hypothesis.
Ftg: Sorry I got you pissed off, but when I took a “computer science” course in college, there were no proofs, or theories, or anything that could be considered even remotely scientific. We learned how to use Microsoft Word and Microsoft Excel, and how to program in some computer language that I never used again. Almost all colleges have courses called “Computer Science,” and almost all of these courses consist entirely of programming and learning to use a couple of popular programs. The instructors are calling their profession a science, when they are in fact just teaching skills.
Funny, I though I was going to get flamed for ripping on archeology.
Sure, but I’ll bet that that chemistry set you got as a kid didn’t let you do DNA sequencing, either. Just because the course you took had an immensely deficient self-description doesn’t mean that all Computer Science does - unless Von Neumann was “just a programmer”.
(That’s not a flame, btw, just an explanation of an all-too-common problem with badly designed educational material. When my HS first decided to introduce a “Computer Science” class, it turned out that all they could think of was to have someone teach people how to program in BASIC. Fucking idiots.)
I dont know about you but, aat our university, we have “Information Systems” which is all Excel & Word etc and we have "Computer Science/Software Engineering & Computer Engineering which is all about data structures and Networks and OS’s etc.