Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

More questions than answers

  1. Jan 22, 2005 #1


    User Avatar
    Gold Member

    their are so many questions and so few answers, is the universe open
    closed? is it infinite or finite? does dark energy, dark matter exist?
    What are the constants that govern our universe? I have read many
    papers that tweek known parameters, or invoke new theories, most
    have no testability without more new data, or may be impossible to
    So my question is what tests are needed to qualify a theory?
    How do you select which theory is correct?
    what tests are needed to aid a selection?
    can we ever have a perfect model of our universe?
  2. jcsd
  3. Jan 22, 2005 #2


    User Avatar
    Science Advisor

    A theory is useful if it can make predictions which are experimentally testable. Specific tests depend on what the theory is about.
    When there are competing theories, they will usually make predictions which are different. Perform the experiment measuring them. Alternatively, theories that may look different turn out to be equivalent - example, quantum mechanics in the 1920's Heisenberg and Schroedinger.
    Ever is a long time, but based on past experience, it is highly unlikely.
  4. Jan 26, 2005 #3


    User Avatar
    Gold Member

    Somebody writes a paper proposing it, orally proposes it, or thinks about it in their head. Once done if the statement is proposed to explain something it is a theory. Subcategories of theories are theories about physics and publish theories about physics. The really good theories of physics of any significant duration are published theories about physics.

    You do experiments and look at past evidence. You dump the theories that are disproven. You keep the rest in the running and use the ones that are convenient the most.

    Logic and experimental experience.

    What is a perfect model. If a perfect model is analogous to the rules of chess. Yes. If the perfect model is analagous to the rules of chess plus a recounting of every move that got us to how the board looks today, probably not.

    Similarly, we may someday determine every method of genetic change that has ever taken place in evolutionary history. But, because the fossil record is incomplete there will be some species in our evolutionary history which may never be known.
  5. Jan 28, 2005 #4


    User Avatar
    Science Advisor
    Homework Helper

    I think the Universe is infinitely divisible but math will always be able to describe it although only imperfectly. The metaphor I use is zooming into the Mandelbrot Set attempting to find the connections between nested sets only to find they have grown smaller still !

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook