Thoughts for the day (i)

[unixronin]

Science is an exacting, meticulous process of continuously discovering that most of what we think we know about the universe is wrong, discarding it, and replacing it with something incrementally closer to the truth as measured by how much of the universe it manages to successfully and self-consistently explain.

Comments

(part 2)

[unixronin.livejournal.com]

The most recent new insight on this front proposes that a "theory of everything" called M-theory (http://en.wikipedia.org/wiki/M-theory) offers a possible physical mechanism by which all of our observations to date could be explained more completely than MOND/TeVeS, on a way consistent with relativity and quantum mechanics, without requiring dark matter. Put simply, it postulates that gravitons, the hypothetical carrier particle of the gravitational force, are loosely bound to the brane — roughly, the fabric of space-time — and the to leak off the brane. Once off the brane, they can interact with matter on the brane only gravitationally; they tend to cluster close to the brane at first, but diffuse away over time. This gives a gravitational force that follows Newton's law at local scales, becomes fractionally stronger at galactic and intergalactic scales due to the clouds of extra gravitons that have leaked off the brane (accounting for the rotational observations), and fractionally weaker than Newtonian gravity at truly huge scales as gravitons diffuse away from the brane through the bulk (accounting for the observed expansion of the universe).

So, state of Newton's law? Still hanging in there, but it may be showing cracks.

Well, how about conservation of parity?

Oops. That one went totally out of the window in 1956, when experiments at NIST proved that parity was not conserved in beta decay of cobalt-60 nucleii. (http://physics.nist.gov/GenInt/Parity/expt.html) Junking conservation of parity actually laid the way for making a heck of a lot of physics a lot more consistent, and allowed a lot of major new work.


So, yeah. There will always be rigid thinkers who regard anything called a law as inviolable. But I think for every scientist who assumes without question that something must be incorrect or experimental error because it appears to violate what's believed to be a known physical law, there's another who thinks "How can I explain this phenomenon in a physically consistent way without violating this law?", and another who asks, "What if this law isn't entirely correct, or is actually wrong?"

I'd go so far as to suggest that these kind of questions are very much harder to ask if there is not an existing framework of believed-correct laws. If you have no physical laws that everything is believed to obey, then something that behaves differently than you've hitherto seen is just something that behaves differently. But if you have a law that says it shouldn't, then sooner or later, some scientist is going to look at it, see that it appears not to obey one or another law, and think "Huh, that's funny. I wonder what's going on there...?"

And that's how the majority of the most dramatic and interesting new discoveries happen. Someone, somewhere, says "Huh, that's funny..."

The most true statement I have heard

[ithildae.livejournal.com]

Science changes one funereal at a time.

Laws are simply a hypothesis that has not been challenged successfully by new discoveries in a very long time. It doesn't make the law correct, but it does give it some staying power.

I would argue that an ideal standard, that is never met, is far better for science than an overly complex explanation and model. Something that is simpler is much easier to extrapolate from, providing predictions and directing experimental scientists in what to look for. (Lots of opportunities for a, "That's interesting..." moment.) Something that provides a great mathematical explanation, that is too complex to predict from, is useless to science. My favorite quote along those lines,

It's not even wrong.