Warning: Wall O’ Text. Responses to content-related comments come first, then at the bottom is more on my approach to using/not using submissions.
otsasonr wrote:The tossup on viscosity includes a prompt for "momentum diffusivity", but not for the broader category of "diffusion coefficient", even though the question says "form of this quantity", which leaves it ambiguous what it's actually looking for.
I indicated to accept "momentum diffusivity" because of the sentence about the Prandtl number. However, from my research, it seemed that this term is used somewhat informally by analogy to the more well-known "thermal diffusivity," so kinematic viscosity isn't formally a kind of diffusion coefficient. Your comment did make me realize that I shouldn't have included the clue about the units of kinematic viscosity in the first place, because those units are shared by other quantities, like thermal diffusivity! (Which makes sense for the Prandtl number to be dimensionless…)
otsasonr wrote:The tossup on the distribution function had a completely useless clue saying something like "the time derivative of this function can be calculated as its Poisson bracket with the Hamiltonian", which applies to literally any function of the canonical coordinates in Hamiltonian mechanics.
The text of the clue is “This function’s time derivative equals the negative of its Poisson bracket with the Hamiltonian,” which is one way of stating Liouville’s theorem. As far as I know, Liouville’s theorem applies specifically to the distribution function, but I may be wrong. (I admit that classical, statistical, and to a certain extent fluid mechanics are where I have the least “real” knowledge, which is borne out by the criticisms you and Eric have made in this thread.)
otsasonr wrote:The tossup on curl used a lead-in describing it as being computed at every step of the Yee algorithm, which led to a swift neg with "finite differences". That should be prompted at least (since that's what is actually computed), or better yet the phrasing of that clue should be changed to something to the effect of "a numerical approximation of this operation is computed at every step of the Yee algorithm", which clarifies what the question is looking for.
The text of the lead-in is “This operation takes place at each ‘leapfrog’ time step of the Yee algorithm.” I have not written or run an FDTD code myself, but I did read Yee’s original paper and several sets of lecture notes on the subject. Based on that research, I concluded that the Yee algorithm executes the time derivatives in Maxwell’s equations at two different points in time but a single point in space, while it executes the curl operations in Maxwell’s equations (or spatial derivatives in the 1-D decomposition, hence the anti-prompt) at two different points in space but at a single point in time that “leapfrogs” by a half-timestep in either direction the points in time where the updates to the E-field and H-field (i.e. the results of the algorithm) are actually occurring. Therefore, at a single “leapfrog” timestep value, the only operations being done are curls, not finite difference or time derivatives.
However, your point is valid because it seems that I neglected to consider all the possible interpretations of the phrasing I chose, specifically the use of the word “timestep” to refer to a point in time rather than a step in the algorithm itself. This is a problem I have often faced with physics questions, as both a player and writer. It can be extremely difficult to condense an equation/theorem/etc. that is normally explained with complicated mathematical notation (as well as definitions of quantities, assumptions, exceptions, and so on) into a clear and unambiguous description that is both comprehensible to the ear at game speed and of a reasonable length (i.e. you don’t want one clue to take up 4 lines). This isn’t an excuse, since I must not have been as clear and unambiguous with the lead-in as I thought I was, but my experience in quizbowl has been that no one has found a magic formula for making understandable quizbowlese out of physics material as it is taught and learned in an academic setting. The ease of translation, if you will, is much easier for most humanities content and arguably for all of the other areas of science as well, since they don’t rely as heavily on mathematics as the default language of description. (Obviously mathematics itself is an exception to this assertion, although I still suspect it is easier to turn math into quizbowl clues because to understand a physics concept you need to understand both the math jargon and the physics jargon, so it is two levels removed from normal English.)
otsasonr wrote:I thought the AdS/CFT correspondence was a bit early in the de Sitter tossup, but that might just be a bias.
I put the AdS/CFT sentence before the sentences on de Sitter precession because the latter topic has more quizbowl exposure, even if AdS/CFT is the most-cited high-energy physics paper OF ALL TIME. Determining what topics are “more famous” than others is a task fraught with danger, since everyone knows different things to different degrees, but that was my judgment here, and Billy and Seth didn’t disagree with me, so that’s how I ordered the clues.
otsasonr wrote:Did anyone convert sawtooth collapse in the tokamak bonus? I've taken a whole course on fusion reactors and didn't, but I might just be salty.
I’m reasonably sure sawtooths are an important topic in tokamak research (and one that I learned while studying for quizbowl long before I wrote this bonus). In the textbooks I used as reference, there were extensive discussions of why sawtooths occur and attempts to model them with various degrees of accuracy. Incidentally, this illustrates one of my main efforts in writing bonus hard parts for this tournament, which was to avoid things named after people whenever possible and include a noticeable number of descriptive/math-based answers. While a high-level topic to be sure, “sawtooth” is an easy answerline to recall because it’s just what the signal looks like. I hope these went over well for the most part. As a player, I longed for more answerlines in the style of “time to the two-thirds power” in the Friedmann equations bonus because they hewed more closely to how I learned physics in the classroom.
otsasonr wrote:Describing what the six-factor formula calculates as a property of neutrons themselves is a pretty serious abuse of terminology.
The six-factor formula determines k, which is called the effective neutron multiplication factor. Even if calling this quantity a property possessed by a group of neutrons were an abuse of terminology, which I don’t think it is, there’s no possible way somebody who knows what this clue is referring to and has heard the moderator say “these particles” 3 times is going to give anything other than the correct answer. You’ve chosen a very small nit to pick here.
otsasonr wrote:While I will grant that Goldstone's theorem is the more common name, it is also definitely referred to as Nambu's theorem and the Nambu-Goldstone theorem, so those answers should be acceptable
I have never seen it referred to as anything other than Goldstone’s theorem, and a Google search of those other names does not turn up any results on the first page, so as far as I’m concerned your argument fails the reasonableness test, i.e. the question writer should not need to devote an excessive amount of effort to hunt down every possible alternate name that someone might use for something. Physicists using nonstandard or idiosyncratic nomenclature is definitely a thing, but I disagree with people who want to stuff answerlines with alternatives that are questionably valid and have an extremely small chance of even being said in the first place, just because one source somewhere says it differently.
otsasonr wrote:The first clue of the pendulum tossup is not unique, and is completely unhelpful. Two-timing analysis and effective potentials are used to analyse many different systems, not just Kapitza's pendulum.
I found this description of how to solve Kapitza’s pendulum and it seemed specific enough, with quite a few distinct mathematical characteristics, that I assumed it was uniquely identifying. If it’s not, mea culpa.
The Quest for the Historical Mukherjesus wrote:1708 is also a critical value for the Rayleigh number in convection (in the Taylor tossup)
I had no idea. Oops!
The Quest for the Historical Mukherjesus wrote:The Hypernetted Chain and Percus-Yevick equations are used to find the radial distribution function, which is also called the pair correlation function, or sometimes just correlation function. I'd push for this not to just be prompted, but accepted.
As far as I recall, I didn’t see “correlation function” used by itself in the sources I used when writing this tossup, so I put it in as a prompt figuring that anyone buzzing in that early in the question would be able to give a more specific answer. If that’s one of your two complaints about my work, I’ll take it (although the 1708 gaffe is embarrassing, I didn't expect to make zero mistakes, especially in the areas where my knowledge is entirely self-taught).
otsasonr wrote:The only other thing that I will say is in response to Austin's decision to write almost all of his categories from scratch, instead of editing submissions. I think it's disrespectful to participating teams to have them contribute questions to those categories if it is known that they are not going to be used. It's one thing for questions to be cut if they are repeated or of poor quality, completely another to knowingly let people write questions which will not be used even if they are good and unique. I understand that things like this are not necessarily known beforehand, but since this pattern apparently began back in August, I think this could have been foreseen.
First of all, I want to emphasize that I took this approach because I didn’t think I would have enough time to properly edit my categories otherwise. When I was offered the editing position (rather unexpectedly, I might add) around the end of July, I was unemployed but going through the interview process at the company where I now work. With regards to my work on ACF Nationals, I had to assume that I’d get a job, with my current employer or somewhere else, in the near future, so I decided I needed to write as many tossups as possible while I still had a significant amount of free time. (Bonuses are a lot easier to think up and a lot faster to write, for me and probably for most other writers, so I wasn’t as worried about those.) I’m a very slow writer/editor, so it was not an appealing prospect to me to churn out half of my questions, even for what amounted to a tad under 10% of the set, in the 2 months or so between when the first packets trickle in and the tournament needs to be ready. In fact, I’ll go further and state that if I didn’t have that free time in August and, to a lesser extent, September, I wouldn’t have agreed to join the editing team in the first place. It’s also important to note that for quite a while, I thought we’d need to produce in the range of 23-25 packets rather than the 20 that ended up being required after (I think) Jerry and Cody came up with a more efficient schedule, so the tossups I wrote or at least conceived of before January turned out to cover a higher percentage of needs than I had thought they would.
A relatively minor but still significant consideration was the need to weight subcategories properly. I tried as much as possible to use submitted bonuses directly or convert submitted tossups to bonuses, but this got progressively harder for the prelim packets as the number of gaps I was looking to plug narrowed. Ensuring adequate representation of all the major subfields of physics is the kind of editorial work that never gets noticed if it’s done right, but is immediately noticed if it’s done wrong. I did the same for earth science, making sure that there wasn’t a huge bias toward either geology or the hydrosphere + atmosphere, and astronomy, making sure to hit topics from the scale of the solar system, to stars and galaxies, on up to cosmology. And I’ve already talked about how I covered mythology.
If you look at the numbers, there really weren’t that many teams that I supposedly “disrespected” with this approach. (Notice I’ve completely avoided discussing the quality of said submissions, which is always a problem with science.) I can see where you could get that feeling, but the real problem is that the editors of ACF Nationals will inevitably throw out a huge number of submissions every year, since the number of teams that write packets is much larger than the number of non-editors’ packets that are needed for the tournament. I don’t know if there is an easy way to reform this scenario, but there’s certainly room for discussion to springboard off the lessons learned from ACF Fall, which has experienced this problem to an even greater degree. As a player, I used required packet-writing as a way to study and wasn’t concerned whether anything I wrote would ever see the light of day, but that’s not how editors should assume everyone thinks. And it’s not what I assumed either – in the end, I figured it was better to run the risk of people being mad that I didn’t use their submissions than to flirt with disaster and end up with a rushed, lower-quality product.