Naked Science Forum
On the Lighter Side => New Theories => Topic started by: malform on 05/01/2006 00:28:54
-
-What makes any theory good?
1 Unifies prevailing yet presently incompatable theories, instead of adding one more incompatable theory.
2 Applies the smallest change to get the largest results.
3 Disprobable. A theory is useless if it cannot be disproven.[If I say the Universe dissapears every other second, and returns with no visible evidence, that would not be disprovable]
4 No reliance on major errors in observation or accepted theory.
5 No reliance on new forms of energy or matter.
6 No reliance on phenomenon unobserved and unneccesary by proven theory.
7 No reliance on paradoxical phenomenon. [for example: multiple universes, space surrounding the big bang "that doesn't exist", travelling in a straight line to end up where you started, or matter/energy creation from nothing]
8 No reliance on the creation of adjustable parameters,
Please read all of my theory:
http://higgs-gluon.com
-
1) Agreed to an extent. If the new theory brings the 2 closer then I'd class that as good.
2) Not necessarily. Sometimes massive leaps are good. Take heliocentricism as an example. Or relativity. The latter was a phenomenal leap (yeah, I know it was based on other peoples' theories)
3) Sort of agree
4) yup
5) Disagree. Many theories have come forward where the physical evidence has only been seen later.
6) As #5
7) As #5
8) Do you mean in the theory itself or adjusting the parameters later to fit in with observations?
I'll read your theory when I'm not so tired
-
Here is my take on the matter.
A scientific theory is a tool to help predict observable phenomena.
Thus it needs two features, firstly the ability to accurately and unambiguously predict, and secondly, as with any tool, to be usable and user friendly.
If it unambiguously predicts an observable event, then the success or failure in observing the predicted event will constitute provability (hence meeting your third criteria).
Usability and user friendliness can be achieved in a number of ways. It may be achieved by being familiar, and thus providing little change from earlier theory; or else it can be achieved by innate simplicity. Either scenario will provide the shallowest learning curve for the prospective user base.
If the theory provides enormous increase in the amount that can be predicted over that which could previously be predicted, then the user base will tolerate a significantly greater learning curve than if it only marginally increases the amount that can be predicted.
If the theory does unify two disparate preceding theories, then even if it in no way increases the number of predictions, it will nonetheless increase the simplicity of the theory base by having one theory where previously there were two – but this only makes sense if the new theory does not end up being twice as complex as the theories it supersedes.
Your criteria 5, 6, 7, and 8 are really just variations on the a theme. Each of the criteria, if violated, can increase the complexity of the theory, and if they do, then they are only justifiable if they create new predictions (predictions that could not be made by existing theories), and those predictions are supported by observations.
-
another_someone - very eloquently stated.
However...
quote:
If it unambiguously predicts an observable event, then the success or failure in observing the predicted event will constitute provability (hence meeting your third criteria).
I think that's a bit narrow. Certain theories predict events or conditions that we currently do not have the technology to prove. Just because the predicted event is presently unobservable, doesn't mean the theory is not good.
Plus, one has to consider that it's easier to prove something isn't rather than that it is. A theory can be tested in a thousand different ways and seem to hold true. But who is to say that the 1001st test won't disprove it? One can never test for all eventualities, so for any theory there is always the possibility of failure. However, once a theory has failed, there is no way back. Or, at least, no way back without modification.
-
I've read your theory & I must say I found it very interesting. As a non-physicist, I can't comment authoritatively, but it certainly has an elegant feel to it. Good luck with it.
-
quote:
Originally posted by DoctorBeaver
another_someone - very eloquently stated.
However...
quote:
If it unambiguously predicts an observable event, then the success or failure in observing the predicted event will constitute provability (hence meeting your third criteria).
I think that's a bit narrow. Certain theories predict events or conditions that we currently do not have the technology to prove. Just because the predicted event is presently unobservable, doesn't mean the theory is not good.
Plus, one has to consider that it's easier to prove something isn't rather than that it is. A theory can be tested in a thousand different ways and seem to hold true. But who is to say that the 1001st test won't disprove it? One can never test for all eventualities, so for any theory there is always the possibility of failure. However, once a theory has failed, there is no way back. Or, at least, no way back without modification.
If all of the predictions the theory are unobservable in the current time, then the theory has no contemporary utility. If all of the predictions that are currently observable are only those that may be predicted with current alternative theory, and it predicts with equal accuracy all of those events that existing theory would predict, then the utility of the new theory exists purely in whether the new theory is easier to use than its preceding theory.
The fact that it predicts events that are not presently observable would be as if it predicts unobservable events, except that the utility of the theory might in future increase as those events at some future date might become observable. The difference between a theory that predicts wholly unobservable events, and one that predicts events that may in future be observable but is not currently observable is that the latter has the potential to increase utility over time if and when these potentially observable predictions are proven to be true.
Ofcourse, a theory that has the potential to increase utility (or find its utility drastically diminished if its predictions at some time in the future are shown to contradict the observable facts) is a more exciting theory just as a sporting fixture where the future outcome is unknown but is expected to become known is more exciting than one whose future outcome is already as well known as it ever will be.
-
quote:
If all of the predictions the theory are unobservable in the current time, then the theory has no contemporary utility. If all of the predictions that are currently observable are only those that may be predicted with current alternative theory, and it predicts with equal accuracy all of those events that existing theory would predict, then the utility of the new theory exists purely in whether the new theory is easier to use than its preceding theory.
Again, I find myself disgreeing. Take the example of general relitivity. When Einstein 1st published it, many of its predictions were unobservable (e.g. time dilation) & it was certainly no easier to use than Newtonian mechanics.
-
quote:
Originally posted by DoctorBeaver
Again, I find myself disgreeing. Take the example of general relitivity. When Einstein 1st published it, many of its predictions were unobservable (e.g. time dilation) & it was certainly no easier to use than Newtonian mechanics.
Time dilation was first predicted by Special Relativity in 1907, while General Relativity came in 1915. Special Relativity ofcourse came about because of the failure of theories using the notion of a luminiferous aether.
General Relativity would have been a dead duck except that it predicted anomalies in the orbit or mercury that could not be accurately modelled by Newtonian mechanics, and in 1919 the predicted deflection of light by the Sun was confirmed.
As you say, General Relativity is complex, and I have no doubt that if any other theory had surfaced at the time that could have as accurately modelled the orbit of Mercury, and would have accurately predicted the deflection of light in a gravitational field (or if the 1919 observations had failed to agree with General Relativity) then the theory would at very least have been shelved pending further experimental supporting evidence.
-
oops! [:I]
But I think you got the point I was trying to make. Just because a theory can't be proven at the time of it's inception, that doesn't make it bad.
I take your point about utility.
-
Thank you DoctorBeaver for your support.
My list of what makes a theory good is not accurate.
It probably should have been titled:
"What makes a theory less flamable on the internet"
-
quote:
Originally posted by malform
It probably should have been titled:
"What makes a theory less flamable on the internet"
The basis of the answer is the same, but the hurdles are simply that much higher.
In essence:
It must explain observed phenomena that are not presently explained, or where the explanation lacks accuracy.
It must not make impossible phenomena that are presently observed and explained (if need be, this can be done with exclusion clauses that make the theory only applicable in certain contexts – but having too many of these will make the theory less attractive).
It must be easily understandable by your audience. It doesn't matter how good your theory is at predicting observables if your audience misunderstands the theory, or fails to use it properly.
It doesn't matter how many unobservable predictions it makes. These may matter in the long term survivability of the theory, but in the first instance the theory will be based upon what people know today, not on what they might know tomorrow.
And finally, however rational or otherwise, the reputation of the author will be significant. If you have a dozen proven theories to your credit, then if you introduce a new wacky theory, you will be given far more leeway than if the same theory had been introduced by an unknown.
-
another_someone - I agree entirely
-
I forgot to add "providing references". I have provided refences to those who ask or are unaware of some discovery or observation.
Most say the lack of formalism is my death-nail.
-
I have had a quick look through your "theory" and I find it totally unscientific.
What you have presented is really a hypothesis an idea that MAY contain a useful basis for scientific investigation.
However my quick look through revealed several fundamental first level errors and misunderstandings of the generally accepted laws of physics that it leads me to believe that you may not be a fully trained scientist.
Whilst I accept that classical imagary can be useful in describing physical processes and use them mysself I feel that imagary alone is not enough ans serious mathematical or experimental evidence is essential.
The most glaring error was your statement that there was a problem with the fact that the alpha particle was less massive than the individual masses of its components. this has been fully understood and explained by the einstein equation e=mc^2 for more than one hundred years.
Learn, create, test and tell
evolution rules in all things
God says so!
-
Soul surfer, when was the last time you made a post that didn't say "That's completely unscientific!"
-
Go to the index and read my posts I don't do it all that often but do not like wild hypoteses on very shaky foundations and am pepared to say so. They tend to confuse people who don't know much by wasting their minds on attractive but fundamentally sterile thinking. There are plenty of good ideas about and plenty of important gaps that needs looking at. It is important to get a good grounding in the way currently accepted theories work before going on the wild side. I do not wish to stifle creative thinking, it comes from personal experence of being attracted down blind alleys that I could have seen were useless with a bit of thought before I entered them.
If you read thoroughly you will also find quite a few wilder ideas of my own to demonstrate that I really am interested in new thinking and not trying to stifle other peoples originality.
Learn, create, test and tell
evolution rules in all things
God says so!