Experiments need a theory to even describe what is going on, or if their result is surprising, why it is surprising.
Why publish actual data otherwise? It would be sufficient to publish the theories, and just claim that they work.
However, you're entirely right that the choice of which experiments are significant (and therefore are the ones that should be carried out and published) depends on the theories of the day. So the succession of experiments is guided by the history of the theories. Is that what you mean?
It is physics that represents a small subset of human knowledge. In fact, the more physics we know the less we need to know as the theories become more and more generally applicable. But for the last 30 years theoretical physics has become largely divorced from experiment.
The extent to which research in theoretical physics is directed by the likelihood of coming up with a publishable paper to put in one's CV for the next job placement 3 years down the line is positively postmodern.
Finally, "my" Hegelian dialectic is not about "Knowledge" but about structural narratives/frames/myths of a society.
good papers (though not all) will be cited, while the trash gets forgotten
When you do a literature search properly you quickly encounter the reality of the plagiarizing of reference lists. Papers are not quoted because they are good, but because they are quoted by other papers. Then again, you're right, just because they are listed in the references doesn't mean they have been read by the authors so, yes, the trash is forgotten.
for the last 30 years theoretical physics has become largely divorced from experiment. Now I think you're exaggerating too :)
it's far from clear that we're close to a GUT. And if we were, so what? We still wouldn't be able to calculate a lot of systems for fundamental mathematical reasons
The results speak for themselves, whatever they are.
At the risk of repeating myself, this diary is not about the experimental sciences, it's about the dominant narratives of society.
But, seriously, most modern experiments make no sense without a theory. A vivid image of what should exist acts as a surrogate for reality. Pursuit of the image then prevents pursuit of the reality -- John K. Galbraith
(key word missing...) A vivid image of what should exist acts as a surrogate for reality. Pursuit of the image then prevents pursuit of the reality -- John K. Galbraith
Information theoretically, the meaning is in the prior and the likelihood. The amount of surprise is, too, since entropy is an explicit function of the model, and for a given datapoint it can take any value as soon as you vary the model.
Is knowledge more than a collection of facts? Yes, but I would say it's a construct built on facts. If we lose the theories and the models, we can rebuild them, or substantially equivalent ones, from the facts. If we lose the facts, we can't just simulate new ones and call them real. -- $E(X_t|F_s) = X_s,\quad t > s$
Within physics you have astrophysics or cosmology which are also historical.
And, to a certain extent, stationarity and ergodicity are model-dependent features.
It's not that "postmodern" "textual analysis" doesn't apply to physics, it is that when studying physics one has to take into account the repeatability.
So "repeatability" is a special quality that applies to a subset of human knowledge.
No, it is limited by the extent of repeatability. A vivid image of what should exist acts as a surrogate for reality. Pursuit of the image then prevents pursuit of the reality -- John K. Galbraith
