Quote:
Originally Posted by Ken G
Quote:
|
Originally Posted by Nereid
I don't know what "astronomy meetings" you're referring to Ken G, but at the very least I think you are missing out on what happens when everyone gets back to the lab/institute/cubicle/whatever.
I mean, if that were all there was, where would ideas for new observations come from? After all, astronomers don't use their zippy new equipment to simply get yet another spectrum of Vega with an x-fold improvement in resolution (to take a ridiculous example)!
But perhaps I misunderstand what you were saying here ...
|
I think you see what I'm saying, but you are choosing to see the glass as half full while I see it as half empty. "Black box" simulations are like fishing nets, and the bigger the net the better the chance of "catching the fish", i.e, including the key physics that describes some astronomical observation. However, they also have the (large) disadvantage that you only know the "fish" is in there somewhere-- they don't tell you which fish was the one you were trying to find, without a lot of additional and painstaking analysis that in my experience rarely actually happens.
Indeed, I see it rarely enough that I take notice when I do see it (it does happen, of course). So that could be improved-- it turns out the skill to create a huge simulation is often quite a bit different than the skill to "boil one down" into what "really happened". It is the latter skill that I see as more and more lacking, whereas decades ago it was much more prevalent (largely because the black boxes had to be much simpler). I'm not saying no one combines those skills-- some do, and admirably at that. It's just not the typical theory or observational talk that does.
As for stimulating future observations, in my experience observations are more often technology driven than theory driven. The CMB was not found because someone said "look for it" (though it almost was, it's true), it was found because the technology to see it was developed. Ditto for quasars, ditto for pulsars, ditto for gamma ray bursts. The theory does often say "you'll need X spatial and Y spectral resolution to see Z effect", but again those can all be callibrated by black box simulations once the discovery of the effect has occured. We do often see "cartoon" level descriptions of "what happened", but that's not what I'm talking about either, as the cartoons don't really tell you much (they won't usually give you factor-2 sorts of estimates, for example). What is (somewhat) lacking is descriptions like "although the full physics that went into the simulation was Y, it turns out that if you just focus on the simple effect X, you get a result whose gross level of accuracy is not that far out of step with the other gross idealizations that are being applied to the question." That is loosely how I would define "understanding".
I should mention, however, that some problems do indeed have a fairly reliable level of precision in the calculations, like stellar interior models. So when the neutrinos come out wrong to a factor of 2 or 3, or if the model won't supernova when it is supposed to, those problems may be taken seriously and lead to new discoveries that can only come from detailed "kitchen sink" simulations (the fishing net, again). So such simulations have their place. I would say they are pretty much all you see, however. (Welcome back Nereid, long time no see!) |
Thanks for the welcome back.
I've read this, several times, and I'm sorry to say I still don't understand what you're saying. This is rather embarrassing, because I nearly always find what you write fairly easy to grasp*, not least because you nearly always write very clearly (to me at least).
I'm not sure if it would help to take some specific examples (it might make matters worse!), but here goes. Oh, and I assume it's not a feature of just astronomy meetings and conferences ... they're just where you see this feature/attribute/way of thinking/whatever most clearly; if this assumption is wrong, I'm sure you'll set me straight post haste.
The Millennium Simulation: huge simulation of a CDM-dominated universe, using GR, which aimed to learn something about the growth of large-scale structure (among other things). One of the many research programmes it (or rather its predecessors and previous analytic work, it just 'shrank the error bars') kicked off was a search for (an OOM more) CDM-dominated dwarf galaxies and other research into dwarfs (merger histories, starburst histories, ...).
Exoplanets: do the doppler programmes qualify as elaborate simulations? After all, to find the n-th planet, you first have to nail down the parameters of the first n-1 ones! Also, the use of microlensing to find planets may be described as hot (and full of models), even though it gives only one shot at each planet. And what of transit searches? and the models used to infer something about the atmospheric composition of the transiting planets (once they're actually identified)?
"Dark Energy": two teams almost simultaneously discovered a consistent trend in high-z Ia SNe data; a flurry of activity followed, much of it involving strenuous efforts to ensure 'accelerated expansion' (and 'cosmic jerk') was a consistent conclusion. Much of this effort necessarily involved elaborate models, but for me the key take-away is the robustness of the conclusion, and that robustness depends critically on (at least some) of the models being quite elaborate.
WMAP and Planck: missions designed explicitly and specifically to study the CMB; compared to COBE the amount of modelling and number-crunching is stupendous; more important however is the combination of robustness and smarts that has gone into the 'how' of digging a cosmological signal out of the raw data (compare, for example, how COBE addressed the zodiacal light component).
Some Galaxy Zoo (GZ) findings: it seems an imbalance of clockwise vs anti-clockwise spirals, reported in some papers, is due to some bias in humans' interpretations of images;
NGC3314 has approval for HST time to investigate Hanny's Voorwerp; a paper on 'blue ellipticals' will be coming out soon; 'zooites' (or 'zooties'!) found 'green peas', this is now been investigated; ... to be sure, even a dozen papers from GZ would be a drop in the bucket of astronomy/astrophysics/cosmology papers.
What am I missing? Can you give some specific examples of the 'half empty' glasses? or rather, how some specific glasses are half empty (rather than half full)?
*
modulo some clarifying questions; of course, I don't always agree with what you write, but it'd be pretty darn boring if I did, right?