Writing science questions for science players

Old college threads.
User avatar
setht
Auron
Posts: 1193
Joined: Mon Oct 18, 2004 2:41 pm
Location: Columbus, Ohio

Writing science questions for science players

Post by setht »

As I mentioned over here, I feel there's been a trend in recent science questions towards lots of clues of the form "fill in the blank in the following multiply-eponymous thing" without much scientific context. In addition, I think many of the "multiply-eponymous things" that have been showing up as clues have been of little use to science players (or at least myself) because they aren't things that show up in science classes. I think it's great that writers are being diligent in searching for fresh clues, but I think people should redirect their efforts a bit. I think overuse of clues of this sort results in questions that give little advantage to people with deeper science knowledge (and possibly disadvantage them, if motivated players start memorizing lots of random eponymous things that have showed up in previous questions).

First off, I claim there is a science canon that is much more clearly delineated than for any other area of the distribution, and I claim that that science canon is defined by what shows up in science classes and standard textbooks, not by chestnuts that have shown up as clues in a bunch of quizbowl questions. I'm not 100% sure of this for bio and chem, but the fact is that there is a large and nearly-uniform body of knowledge covered in undergraduate curricula in physics, astro, earth science, and math. My sense of things is that the shared body of course material is much larger in the (physical) sciences than it is in the humanities or social sciences--all physics undergrads learn about Maxwell's equations; not all English undergrads read Hamlet. Moreover, the core sequence of classes is (I believe) more uniform in the sciences than it is in other areas: going back to physics, it seems like most departments have a 1-1.5 year introductory sequence (basic mechanics, EM, maybe some basic thermodynamics, then maybe waves/introductory quantum), then a set of upper division core classes in classical mechanics, EM and maybe some optics (usually more than one upper division class), thermodynamics/statistical mechanics, and quantum mechanics (usually more than one upper division class). This set of classes seems largely uniform and there is a lot of overlap from program to program in the material covered in these classes. I am confident in claiming that nearly all physics undergrads see Maxwell's equations in differential and integral form in multiple classes, the Heisenberg uncertainty principle in terms of the commutator of incompatible observables, etc. Furthermore, it's not just physics majors taking the introductory sequence: at UC Berkeley, chem majors have to take the first 2/3 of the sequence, most engineering students have to take the sequence, astro students have to take it, earth and planetary science students have to take it, etc. Meanwhile, pre-meds have to take a slightly condensed version that still shares a lot of the same material.

Second, I claim that science clues of the form "fill in the blank in the following multiply-eponymous thing" with little science context are at least as bad as literature clues of the form "name the author of the following title" with little context about plot, characters, critical reaction, etc. In some cases the situation is worse in science questions: a question on an author with some titles and a little context may give knowledgeable players a chance to narrow down time period, area of the world, genre, etc. The same really can't be said for almost any question on a scientist: at best a science player can narrow down time period, but "physicist from the early-to-mid 20th century who worked on all sorts of things" really doesn't get you anywhere. Similarly, bad descriptions of plots, jamming several descriptions of works together before giving titles, and confusing wording in literature questions are (I think) less problematic for literature players than analogous gaffes in science questions are for science players: as I said before, even bad/confusing descriptions of works may contain clues that are useful in narrowing down an answer, and if I write a literature question that sounds like the plots from two short stories are actually descriptions of a single work, there's little chance that my chimeric plot description is going to sound enough like some third work to get a knowledgeable player to neg. If I write a science question that makes it sound like I'm talking about the high-temperature limit of some situation instead of the low-temperature limit, there's a decent chance there is some named thing associated with the high-T limit and some knowledgeable play might neg. I guess this is all by way of saying that I think people should try to provide goodly amounts of science context in science questions, but they shouldn't go overboard with lots of context from material people don't cover in classes, and writers should try to be clear in indicating which pieces of science context are connected and which are not.

I don't think every science question has to have an answer and all of its clues taken direct from the science canon I described above (and I certainly don't think everything has to come from the more restrictive pool of material common to all core sequences--there's plenty of good stuff that deserves to come up and typically shows up in elective courses, for instance). However, given that one of the goals of good quizbowl is to reward people with knowledge, I think the way people learn science means that we need to put more emphasis on that science canon. No one sits down and reads up on all the stuff Zeeman worked on besides the Zeeman effect; if someone is going to learn more about Zeeman than the basic stuff covered in quantum perturbation theory, they're going to learn more about the Zeeman effect (perhaps I'm wrong and Zeeman has other stuff that people still really care about, but I think the basic point stands: no one sits down to read up on all things Zeeman). Science students learn what they need to learn; if they want to go beyond what they learn in classes, they almost always pick material based on connections to other material of interest. I get the impression this is not the case with, say, literature: my impression is that most literature players in quizbowl do most of their reading outside of classes, and that many people find an author they really like and then read through a bunch of their stuff. So, in a literature question on an author it often does advantage the more knowledgeable players to include clues on obscure works, because the most knowledgeable players read them; in a science question on Zeeman, on the other hand, clues on papers and research not connected with the Zeeman effect are likely to be useless even to the most knowledgeable physics players, and the solution is to give less space to those clues and more space to clues that are useful to someone who has studied the Zeeman effect in more depth.

I'll try to write up some thoughts some time later on resources for non-science people to identify "canonical" material (and it would be cool to have other people jump in with suggestions). Also, it would be cool if some bio and chem people could jump in and talk about their categories--I believe Mike and Jerry agree with me that a lot of recent physical science questions have been too shy about using solid middle clues from canonical course material, but I'm not sure how bio and chem players feel about recent questions in their areas.

-Seth
Seth Teitler
Formerly UC Berkeley and U. Chicago
President of NAQT
Emeritus member of ACF
User avatar
DumbJaques
Forums Staff: Administrator
Posts: 3086
Joined: Wed Apr 21, 2004 6:21 pm
Location: Columbus, OH

Re: Writing science questions for science players

Post by DumbJaques »

I remember that Evan Adams once espoused the idea of having literature questions on authors talk mainly about 2 or 3 works at the most, so as to serve as sequential mini-tossups on those works instead of allowing people who know the name of a random work to outbuzz people who read stuff. I thought that idea was reactionary and mostly bad for reasons that are more central to the literature canon, but I'm not convinced much of the same principles wouldn't benefit science tossups (at least of the "this name" variety). Anyone can learn random crap named for people, especially if it's paired with another person's name - that doesn't even require me to understand the concept, which I usually at least attempt to do anyway. I once beat Sorice on a bullshit Debye buzz because I memorized the like, 4 most obscure but still important things Debye names. Hilariously, if the question had talked about whatever Debye is best-known (or second best-known) for, I would have been shit out of luck, because not only am I sure Mike knows more about those things than I do, I don't even think I really know what those things are.

To return to Sorice-related points, Mike made a post in another thread that I very much agree with - that finding good middle clues is often the hardest part of writing tossups (once you reach a certain level of proficiency). I think there's been a big trend in questions lately, and perhaps particularly among newer writers including myself, to construct questions that, for some reason, end up with a whole lot of lead-in level clues. I personally fall into this trap because I end up doing lots of connected reading whenever I write a question on something and put in anything that sounds interesting until you're reading a 15 line question on the Warlord period, but I think most people do it either out of lack of canon mastery or the perception that tossups are supposed to be like that. Honestly, questions do not need that many very difficult clues. The amount of people who will even *get* a given nationals or plus level answer is already kind of small, and the amount of people who are going to get buzzes in the first few lines of many questions is most likely very, very few. If your leadins have lasted four and five lines into the question, you've not only eclipsed your utilitarian purposes, you've also overdone whatever canon expansion justification you might have had. Now if you want to really do your job on the rest of the tossup, you need to make it like 12 lines (which is not really ideal) or skimp on middle clues (which is much, much worse). If people have a phobia of middle clues, it's time to lose it. Keep in mind that, as easy or fraudable as the clues you're typing up might sound to you, you just read and wrote them; when players are deprived of this advantage, you'll notice that they very rarely go as early as you think they will. I think this also extends to over-opacity due to concerns for linguistic or other fraud that leads to people (at times, perhaps foolishly, but still) becoming needlessly confused (I buzzed in on the first line of the 2nd amendment tossup at nationals after it had said "this law" and could not figure out what they wanted, so I said "gun bans"). That's a separate topic of discussion, I suppose.

I don't actually know if this overuse of largely trivial leadins is as pronounced in science as I think it might be, but Seth's post suggests that I may be right. Based on said post, I will surely attempt to construct my science people tossups in ways that spend much of the tossup discussing the things they really are best known for, but I'm interested to know if there's a way for me to more readily identify lead-in clues as a non-science person. I'm a sucker for naturalism and don't use the database while writing much anyway, but if I'm reading correctly, it seems like showing up a lot as clues in quizbowl hardly makes something important in a scientific sense. I liked reading through that book with the dead cat on the back, but that's where my science textbook experiences are going to end. So I'm interested in what Seth (or Jerry/Mike/Eric/etc.) might suggest on how to go looking for those clues.

It also strikes me as somewhat amusing that, in our original communal attempt to shift from fake science clues on trivial things like biography that enabled non-science people to beat more knowledge science players, we have now arrived at a point where we may be going so overboard with "hardcore" academic clues that we've to some degree leveled the playing field once again - and now the science people are out to deprive us of our deliciously fake buzzes once again. You people are all a bunch of haters. Science clues now, science clues tomorrow, science clues forever, right Teitler? You make me sick.
Chris Ray
OSU
University of Chicago, 2016
University of Maryland, 2014
ACF, PACE
User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

Re: Writing science questions for science players

Post by grapesmoker »

I agree with virtually everything Seth has written; perhaps I'll have more substantive comments later. However,
DumbJaques wrote:I remember that Evan Adams once espoused the idea of having literature questions on authors talk mainly about 2 or 3 works at the most, so as to serve as sequential mini-tossups on those works instead of allowing people who know the name of a random work to outbuzz people who read stuff. I thought that idea was reactionary and mostly bad for reasons that are more central to the literature canon, but I'm not convinced much of the same principles wouldn't benefit science tossups (at least of the "this name" variety).
I don't want to hijack this thread for discussion of literature questions, but I think you're very mistaken in calling this writing style "reactionary." Evan has a point here; writers are prolific and it's not necessary to mention every work (or every other work, or every third work even) in order to write a good tossup. There's no reason why 3 or 4 works shouldn't suffice to write a good tossup on an author. I personally think rewarding deep knowledge of a few important works is better than rewarding cursory familiarity with 10.
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance
User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

Re: Writing science questions for science players

Post by grapesmoker »

DumbJaques wrote:It also strikes me as somewhat amusing that, in our original communal attempt to shift from fake science clues on trivial things like biography that enabled non-science people to beat more knowledge science players, we have now arrived at a point where we may be going so overboard with "hardcore" academic clues that we've to some degree leveled the playing field once again - and now the science people are out to deprive us of our deliciously fake buzzes once again. You people are all a bunch of haters. Science clues now, science clues tomorrow, science clues forever, right Teitler? You make me sick.
It's not so much that we've gone overboard with "hardcore academic clues," it's that a lot of clues now being used are of very remote significance to anyone who doesn't work in the subfield to which they belong. Take tossups on the Stark effect, for example: I'll be damned if I've ever used that Autler-Townes effect (I believe it's used in spectroscopy) but pretty much anyone who's been reading packets for the last few years, scientist or not, knows that when they hear that it's time to buzz with "Stark." So as you yourself pointed out, this produces anomalous results like you beating Mike to a Debye tossup even though it's obvious that Mike's knowledge of all things Debye is way better. I think this is more of a problem with fill-in-the-blank questions than with other types of questions.
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance
User avatar
Skepticism and Animal Feed
Auron
Posts: 3209
Joined: Sat Oct 30, 2004 11:47 pm
Location: Arlington, VA

Re: Writing science questions for science players

Post by Skepticism and Animal Feed »

The point of a canon, to me, is that people can read old packets and improve. A canon based on anything else adds information costs to the process of improving; not only must you look at old packets, you must also look at textbooks or take classes.
Bruce
Harvard '10 / UChicago '07 / Roycemore School '04
ACF Member emeritus
My guide to using Wikipedia as a question source
User avatar
DumbJaques
Forums Staff: Administrator
Posts: 3086
Joined: Wed Apr 21, 2004 6:21 pm
Location: Columbus, OH

Re: Writing science questions for science players

Post by DumbJaques »

I don't want to hijack this thread for discussion of literature questions, but I think you're very mistaken in calling this writing style "reactionary." Evan has a point here; writers are prolific and it's not necessary to mention every work (or every other work, or every third work even) in order to write a good tossup. There's no reason why 3 or 4 works shouldn't suffice to write a good tossup on an author. I personally think rewarding deep knowledge of a few important works is better than rewarding cursory familiarity with 10.
Eh, I don't think I'm representing this correctly. As I recall, that post suggested that the ideal author tossup talked about two works or something. That seems extreme to me - certainly there are some authors where you'd be best-served talking mostly about 3-4 works and maybe briefly discussing another, but that's a far cry from a two-work author tossup (which I maintain is, in fact, way too far in the other direction when you submit that as a model), and there are plenty of authors who you wouldn't want to limit in that way. I certainly would be interested to see if people want to try writing author tossups with fewer works in them - I feel as though many of us fall into some very formulaic stuff when we do author tossups and it would be nice to have some variety.
I think this is more of a problem with fill-in-the-blank questions than with other types of questions.
Well, there are certainly named things of questionable actual relevance (although I know people whose work covers the Autler-Townes effect rather heavily) in all other kinds of science tossups. I think the difference is that, when you frame a clue like "Dipshit's Law of Disequilibrium must be modified with the Fathead Coefficient to account for this phenomena," you at least need to have a cursory understanding of how things work. I think this can be eliminated if, as Seth proposes, you simply make these tossups on people focus on their better-known things rather than hurling out minor things they co-name with random other people. I think Eric's post in the Missouri Open thread demonstrated that, even when you diligently research a host of minor but important eponymous clues, it often ends up creating the "buzz when you name-associate" problem anyway. I certainly don't want to see a movement away from tossups on scientists as I find the information in these questions very interesting and think it keeps things accessible towards the end even with higher difficulty subjects.
Chris Ray
OSU
University of Chicago, 2016
University of Maryland, 2014
ACF, PACE
User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

Re: Writing science questions for science players

Post by grapesmoker »

Whig's Boson wrote:The point of a canon, to me, is that people can read old packets and improve. A canon based on anything else adds information costs to the process of improving; not only must you look at old packets, you must also look at textbooks or take classes.
Oh shit, you might have to learn something!
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance
User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

Re: Writing science questions for science players

Post by grapesmoker »

DumbJaques wrote:Well, there are certainly named things of questionable actual relevance (although I know people whose work covers the Autler-Townes effect rather heavily) in all other kinds of science tossups.
My point is not that AT is irrelevant or useless. The point is that it's a clue that's been used enough where people who have no idea of what it is or its significance or its relation to the price of tea in China are buzzing on word association alone. So now if you put AT in the first clue of a Stark effect tossup, you have generated buzzer races between a bunch of people who read old packets and buzz on reflex. My second point is that AT is not something you cover in any undergraduate class nor any graduate class I've ever taken; it's literally something that only a specialist knows a lot about.
I think the difference is that, when you frame a clue like "Dipshit's Law of Disequilibrium must be modified with the Fathead Coefficient to account for this phenomena," you at least need to have a cursory understanding of how things work. I think this can be eliminated if, as Seth proposes, you simply make these tossups on people focus on their better-known things rather than hurling out minor things they co-name with random other people.
Why do you want to eliminate this? That sounds like a bad idea and directly contrary to everything Seth is arguing.
I think Eric's post in the Missouri Open thread demonstrated that, even when you diligently research a host of minor but important eponymous clues, it often ends up creating the "buzz when you name-associate" problem anyway. I certainly don't want to see a movement away from tossups on scientists as I find the information in these questions very interesting and think it keeps things accessible towards the end even with higher difficulty subjects.
Well, the problem with that question, in my mind, was that it referenced a lot of random things named after a dude, with no clear progression from difficulty to ease. Moreover, many of those things were not particularly well known to most people.
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance
User avatar
Captain Sinico
Auron
Posts: 2869
Joined: Sun Sep 21, 2003 1:46 pm
Location: Champaign, Illinois

Re: Writing science questions for science players

Post by Captain Sinico »

Whig's Boson wrote:The point of a canon, to me, is that people can read old packets and improve. A canon based on anything else adds information costs to the process of improving; not only must you look at old packets, you must also look at textbooks or take classes.
Well, yeah! A topic having come up isn't and can't be sufficient (or necessary) for it to come up in the future. On the contrary, academia needs to dictate what's fair game for quizbowl, or we're not playing an academic game in any proper sense.
Even laying that to one side, I don't see any compelling reason we should restrict the type of knowledge we reward to things that have come up in packets before. Your argument seems to be that we have to consistently reward the same knowledge for the purposes of making the game easier to learn. I disagree: if a type of knowledge isn't meritorious, then rewarding it/easing its acquisition isn't a positive end.
And even ignoring all that, you've begged the question! Even supposing we accept your idea that we ought to have some fixed canon based only on quizbowl and reward only that says nothing important about what that canon ought to contain. In other words, accepting your premises, one would read Seth's argument as: "The science canon ought to contain knowledge of some kinds and not others; it is not currently what it ought." It seems, then, that you must either surrender the point or contend that we cannot change what's in the canon, which I don't think you believe.

MaS
Mike Sorice
Coach, Centennial High School of Champaign, IL (2014-2020) & Team Illinois (2016-2018)
Alumnus, Illinois ABT (2000-2002; 2003-2009) & Fenwick Scholastic Bowl (1999-2000)
Member, ACF (Emeritus), IHSSBCA, & PACE
User avatar
at your pleasure
Auron
Posts: 1716
Joined: Sun Aug 03, 2008 7:56 pm

Re: Writing science questions for science players

Post by at your pleasure »

I personally fall into this trap because I end up doing lots of connected reading whenever I write a question on something and put in anything that sounds interesting until you're reading a 15 line question on the Warlord period, but I think most people do it either out of lack of canon mastery or the perception that tossups are supposed to be like that.
Another part of the problem is that if a writer is especially interested in something, that's going to skew the writer's difficulty meter when it comes to that topic. To provide a concrete example, someone who studies art history pretty deeply is more likely to underestimate the diffculty of the easiest clue possible on Massacio's second-or-third best known work than someone who does not study art history pretty deeply. This is also more likely to be a problem for newer writers, since they have less experience gauging question difficulty.
Douglas Graebner, Walt Whitman HS 10, Uchicago 14
"... imagination acts upon man as really as does gravitation, and may kill him as certainly as a dose of prussic acid."-Sir James Frazer,The Golden Bough

http://avorticistking.wordpress.com/
User avatar
DumbJaques
Forums Staff: Administrator
Posts: 3086
Joined: Wed Apr 21, 2004 6:21 pm
Location: Columbus, OH

Re: Writing science questions for science players

Post by DumbJaques »

Why do you want to eliminate this? That sounds like a bad idea and directly contrary to everything Seth is arguing.
Ha, this is a poorly constructed sentence. I meant eliminating the problem with fill-in-the-blank questions, not eliminating the obviously useful clue I had outlined before.
Well, the problem with that question, in my mind, was that it referenced a lot of random things named after a dude, with no clear progression from difficulty to ease. Moreover, many of those things were not particularly well known to most people.
Yeah, I agree, but reading through Eric's account of writing the question, it certainly seemed like he made an effort to seek at things that were scientifically important even if he had never personally encountered them. My point was that even when you adhere to the real scientific relevance clue criteria, it seems, you can end up with stuff that has the same practical effect if they end up just not being known in quizbowl. Since we've already established that previous packet occurrence is not a great metric for evaluating actual scientific accessibility, then you can't necessarily look at something that seems scientifically relevant and check that based on its presence in the packet archive. Certainly, as you said, these things were amplified greatly (perhaps from insignificance) by the excessive difficulty of the answer.
Chris Ray
OSU
University of Chicago, 2016
University of Maryland, 2014
ACF, PACE
Tower Monarch
Rikku
Posts: 360
Joined: Sun Oct 07, 2007 6:23 pm
Location: Richmond, VA

Re: Writing science questions for science players

Post by Tower Monarch »

Whig's Boson wrote:The point of a canon, to me, is that people can read old packets and improve. A canon based on anything else adds information costs to the process of improving; not only must you look at old packets, you must also look at textbooks or take classes.
While I think the point has been made clearly that at face this sounds dumb, I would like to point out that it contains some necessary points. I own around 15 science-related textbooks from all levels of undergraduate courses (not to mention another 20+ math textbooks from linear algebra up), but just flipping through them reveals few things as interesting as the stuff I can find by typing "superconductor" into the packet archive. If you (meaning Seth/Jerry/Mike/Eric, etc) want to make a huge list of upper-undergraduate and graduate textbooks from which you want to make a decently-sized answerspace that can still contain interesting clues, I will grab some of those and study until I'm beating you on these questions. But, until it's more clear where information can come from, I am actually learning science just as someone like Evan Adams might learn literature: I hear some interesting clue in the middle of a tossup, and I read a paper on it or look at an online tutorial, no matter if it is in the average graduate course on the subject.
It is obviously necessary to avoid some of the issues that have definitely come up recently: even as someone who cannot (and should not, yet, be able to) buzz before the giveaway of upper-level science questions, I notice early buzzes coming from people who probably don't understand what they're doing (I'll use Chris Ray as an example, but there are many players like this) while players who have clearly spent time studying the general field let it go by (I've seen Jerry have to wait until the end of physics questions many times). That is definitely an issue. However, if you want someone such as myself (or Andy Watkins, if you want someone who has already made significant progress) to become a high-level player as easily as someone else could become highly competitive at history or whatever, then there needs to be a way to do this. Right now, that way is to read old packets and decipher the context for each clue. With the new model, the way seems less clear: unless I am missing something, you're asking me to double my textbook collection and maybe get tossups in the middle...
Cameron Orth - Freelance Writer/Moderator, PACE member
College: JTCC 2011, Dartmouth College '09-'10, '11-'14
Mathematics, Computer Science and Film/Media Studies
High School: Home Schooled/Cosby High '08-'09, MLWGSGIS A-E '06-'08
User avatar
DumbJaques
Forums Staff: Administrator
Posts: 3086
Joined: Wed Apr 21, 2004 6:21 pm
Location: Columbus, OH

Re: Writing science questions for science players

Post by DumbJaques »

It is obviously necessary to avoid some of the issues that have definitely come up recently: even as someone who cannot (and should not, yet, be able to) buzz before the giveaway of upper-level science questions, I notice early buzzes coming from people who probably don't understand what they're doing (I'll use Chris Ray as an example, but there are many players like this) while players who have clearly spent time studying the general field let it go by (I've seen Jerry have to wait until the end of physics questions many times).
Well, this definitely didn't happen at the Missouri Open mirror!

I mean, I don't think this is totally accurate. I think pretty much everyone has early buzzes where they don't really understand what they're doing (and surely I've done this in areas that aren't science) - that just ends up being the nature of the game. I think with most players, they can play percentages and try to come up with some good buzzes, and if things work out end up playing a bit better than they are. I don't know that it has anything to do, intrinsically, with how people construct questions (or science questions, in particular). I think I'd try to find a way to do this regardless - it's not like I set out to get buzzes I'm less than sure of the way I set out to learn about things that interest me, but I challenge that my ability to remember that some random modification of a certain reaction has come up before is somehow less legitimate than anyone else remembering the title of somebody's only play, or that somebody won a certain battle, etc. You should make sure that people can't fraud a clue faster than someone who really knows it can answer it (and should obviously address things like obvious linguistic fraud or whatever), but if you're proposing somehow actively stripping tossups of proper names in a quest to rob non-science players of buzzes, well, that's silly! Besides, I'm not really sure what it is that I (or those other people) do on science questions, but whatever it is, it strikes me that science people should be able to do it too (but better, since they have the knowledge). I guess I'm unclear as to what you're citing here as problematic in science questions - or perhaps questions in general - that we for some reason ought to address. It's not like quizbowl is beset by a systemic problem of me beating Eric Mukherjee to chem questions.
Chris Ray
OSU
University of Chicago, 2016
University of Maryland, 2014
ACF, PACE
Tower Monarch
Rikku
Posts: 360
Joined: Sun Oct 07, 2007 6:23 pm
Location: Richmond, VA

Re: Writing science questions for science players

Post by Tower Monarch »

DumbJaques wrote: I guess I'm unclear as to what you're citing here as problematic in science questions - or perhaps questions in general - that we for some reason ought to address. It's not like quizbowl is beset by a systemic problem of me beating Eric Mukherjee to chem questions.
The part your referring to is just a poor restatement of the rest of the thread. My point (and one of the points hit upon in all the previous posts) is that someone genuinely interested in learning science for real and for quizbowl (currently rather different goals) right now has to compete in the memorization game of what do I buzz on when I hear this named thing (see: Jerry on AT) because the context given that should make science players have the advantage over those who couldn't care less about real science knowledge (I used you as an example, but I could substitute in Evan A, if you want) is either nonexistent, ambiguous or confused (see: Jerry on high-temperature). The solution put forth in this thread is to move away from the arms race of named things that appear in random papers that the writer happened to see on google books or pubmed and towards upper-level yet "canonical" textbooks. The purpose for my last post was that I would like to ensure that the interesting things that currently come up will continue to come up and that there is a way for writers and up-and-coming players (I cited myself) to find which of these things will be canonical.
As for the beating Eric thing, my point was that you have a high chance of beating Will Butler (substitute any undergraduate, interested/strong science player not quite at the level of Eric/Mike/Jerry/Seth) on many science topics for which he has the better science basis but you have the better quizbowl basis. With the high level science players, the problem isn't that they will lose to you but that they either have to wait 5 lines to find something they've actually studied (this is bad because they have studied more than anyone else and deserve to get it as early as Andrew Yaphe can get an average humanities question) or will not be able to answer a question on something that is not academically significant (therefore forcing them to give up the advantage of science knowledge and decide games outside of what is likely there best subject). With these two, different points (which you merged into your last sentence), I just wanted to confirm that there is definitely a need for this discussion/ change in general for science writing.
Right now, I would say that I, as someone who studies academic science, face two difficulties: science is probably the hardest of any quizbowl subject to break into (due to the reasons outlined in this thread) and I have minimal (sometimes no) advantage over those who choose less-than-academic approaches (named thing memorization) to learning science. I thought it would be worthwhile to add to the discussion among top science players the problems faced by rising science players.
Cameron Orth - Freelance Writer/Moderator, PACE member
College: JTCC 2011, Dartmouth College '09-'10, '11-'14
Mathematics, Computer Science and Film/Media Studies
High School: Home Schooled/Cosby High '08-'09, MLWGSGIS A-E '06-'08
User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

Re: Writing science questions for science players

Post by grapesmoker »

Hey, if you want to look at some upper-level textbooks on physics, here are my recommendations:

E&M:
Griffiths, Introduction to Electrodynamics: A standard undergraduate text. Used in many upper-division E&M courses.
Jackson, Classical Electrodynamics: The Bible; it's dense and hard, but contains a wealth of useful information.

Solid state:
Kittel, Introduction to Solid State Physics: used in many upper-division courses. Covers a decent amount of information but is not too rigorous.
Ashcroft and Mermin, Solid State Physics: The equivalent of Jackson for solid state physics. Covers a large variety of topics and is quite dense.

Quantum mechanics:
Griffiths, Introduction to Quantum Mechanics: Has Schrodinger's cat on back and front. About the same level for QM as the E&M book, covers most standard topics but is not particularly adventuresome.
Sakurai, Modern Quantum Mechanics: Small and dense; contains a lot of neat stuff but is a very difficult book.
Bransden and Joachain, Quantum Mechanics: Not a standard book but one I used for my QM class. More complex and in-depth than Griffiths and covers a lot more ground, with more emphasis on atomic applications.
Cohen-Tannoudjui, Quantum Mechanics: A huge tome in several volumes. Everything and then some, but is also very difficulty. From my perspective it contains some of the best straight up explanations of how to understand QM.

Thermodynamics:
Kittel and Kromer, Thermal Physics: Used in many undergrad classes. Not particularly rigorous or in depth but has decent breadth.
Reif, Statistical and Thermal Physics: My personal favorite. Contains in-depth discussions of all important aspects of thermodynamics. A little out of date in terms of modern applications but very clear.

I hope that helps people who are looking for sources for their question writing material. This isn't meant to be an exhaustive list, but it's the list of books that many people who take physics classes are likely to encounter. In addition, any volume of Landau and Lifschitz is guaranteed to contain mountains of useful information (it's a 10 volume set on all of the above topics and then some) but L&L is notoriously difficult so be warned.
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance
User avatar
Sima Guang Hater
Auron
Posts: 1881
Joined: Mon Feb 05, 2007 1:43 pm
Location: Nashville, TN

Re: Writing science questions for science players

Post by Sima Guang Hater »

Tower Monarch wrote:The part your referring to is just a poor restatement of the rest of the thread. My point (and one of the points hit upon in all the previous posts) is that someone genuinely interested in learning science for real and for quizbowl (currently rather different goals) right now has to compete in the memorization game of what do I buzz on when I hear this named thing (see: Jerry on AT) because the context given that should make science players have the advantage over those who couldn't care less about real science knowledge (I used you as an example, but I could substitute in Evan A, if you want) is either nonexistent, ambiguous or confused (see: Jerry on high-temperature).
Some of this is inevitable to quizbowl, and I don't think moving towards "canonical" textbooks will really help with it (though I think its a good idea for other reasons). Even if I were to smooth out the pyramid in my Kramers tossup as Seth suggested, for example, there's nothing stopping a non-specialist from memorizing things about the WKB approximation without really knowing what they mean, just as I can still (rarely) beat Jonathan to a good literature tossup without having read the book. I can never have solved an Airy equation in my life but still buzz on that clue on WKB tossups, just as people can buzz on transparent PCR tossups early without ever having pipetted.

Specifically, though, ambiguity and confusion is easier to correct, and I think a lot of what causes science players to get [unfairly?] beaten to or confused on tossups is strictly the phrasing and NOT the type/source of the clues present. I caught a little bit of it this weekend when I recognized (and have even attempted to isolate for ~4 years!) a protein complex that came up in a question about telomeres, only to have Chris beat me to it because the question mentioned that these structures have "ends", and another tossup on methylation that mentioned trithorax (I know exactly what it does, but I couldn't figure out what the question wanted - this one was much less costly). Another case in point, if I had cut out the first two clues in the Kramers tossup and spent a little more time describing the Kramers-Kronig relations and the WKB approximation (ie using textbook clues), I'm sure that tossup would have gone over a lot better (but as I said above, would have still been fraudable).

In biology, I feel like the memorization game and the "real" science game go hand-in-hand a little better than it does in other fields - my classes have often asked me to memorize things from papers that in turn come up in questions. This is why as long as you're writing about a protein family or process that has some weight of research about it, you can't really go wrong using it as a leadin. When a knowledgeable person is writing, these things go hand-in-hand in every subject. Take the tossup on the prime number theorem, which I got based on the fact my class spent a week proving it and not due to any stock clues whatsoever (even though I'm sure that really big number that bounds the Logarithmic Integral thing will become stock sooner or later, along with the von Mangoldt function).
Tower Monarch wrote:The solution put forth in this thread is to move away from the arms race of named things that appear in random papers that the writer happened to see on google books or pubmed and towards upper-level yet "canonical" textbooks. The purpose for my last post was that I would like to ensure that the interesting things that currently come up will continue to come up and that there is a way for writers and up-and-coming players (I cited myself) to find which of these things will be canonical.
I don't think the random paper arms race is as bad as people think, and I certainly don't think its a problem in the general sense. The usual caveats of only using pubmed when you know what you're doing apply, but I don't see anything wrong with using a current paper in a leadin, especially when its a new variation of an effect (cf those boron nitride nanotubes undergoing the Giant Stark effect). As long as your middle clues aren't cluttered with research, the question will usually turn out fine. I try to put a clue about derivation in the middle to upper middle of a question (for example, putting the selection rule for the Raman effect in such a tossup, perhaps right after plasmon resonance, which a lot of biologists have heard of). I don't think moving to textbooks (as opposed to online sources explaining your clue) for middle clues will really make the questions fraud-proof, but it will make them more "real", which I can support.
Tower Monarch wrote:As for the beating Eric thing, my point was that you have a high chance of beating Will Butler (substitute any undergraduate, interested/strong science player not quite at the level of Eric/Mike/Jerry/Seth) on many science topics for which he has the better science basis but you have the better quizbowl basis.
I'm not sure if you're referring to Chris in particular, or non-specialists in a broad sense, but this is a problem that plagues all of quizbowl. I manage to edge Jerry out of physics tossups that he has a greater overall grasp on with some regularity, simply because I've heard the clue or I've read about/written a question on the answer - and its not always because the questions are poorly written.
Tower Monarch wrote:With the high level science players, the problem isn't that they will lose to you but that they either have to wait 5 lines to find something they've actually studied (this is bad because they have studied more than anyone else and deserve to get it as early as Andrew Yaphe can get an average humanities question) or will not be able to answer a question on something that is not academically significant (therefore forcing them to give up the advantage of science knowledge and decide games outside of what is likely there best subject).
Maybe its different for bio, but I really don't feel like this is a big problem on tossups. I'm generally out of the loop on the leadin, but know what's going on around middle-clue time, and I think for specialists that's the way it should be. Where the idea of academic significance applies, though, is on bonuses, where forcing people to remember academically unimportant facts can really, REALLY destroy a specialist advantage. Maybe for physics, these textbooks will help find better examples of bonus third parts (assuming that's a problem now)

Tower Monarch wrote:Right now, I would say that I, as someone who studies academic science, face two difficulties: science is probably the hardest of any quizbowl subject to break into (due to the reasons outlined in this thread) and I have minimal (sometimes no) advantage over those who choose less-than-academic approaches (named thing memorization) to learning science. I thought it would be worthwhile to add to the discussion among top science players the problems faced by rising science players.
You're going to face this problem for some time, no matter how we alter the canon or change sources. You should either take a hybrid approach (ie memorize, then gain real knowledge), or be prepared for a bit of a lag while your buzzing catches up to your knowledge. I've defaulted to the former. I'm not sure whether this is really any different from other subjects - I think science just gets a bad rap because you have to wrap your head around it in a specific way (which a good intro textbook will teach you how to do).



LIST OF TEXTBOOKS FOR BIO:

-The best solution is to grab some class notes from advanced undergrad courses, as these are usually 100% article-driven, unlike many other upper-level sciences at the UG level. This gives you a good idea of what research and techniques are important. If someone wants to show me some webspace, I can make mine available.

Good textbooks to have:
-Basic: Campbell's Biology, 7th edition. Still the king, and is both very readable and more than enough to understand the rest of these.
-Immuno: Kuby's Immunology
-Cell Bio: Albert's Molecular Biology of the Cell. Some people also like Lodish's Molecular Cell Biology.
-Genetics: Griffiths' Introdution to Genetic Analysis (Basic - you can substitute this one out for others) and Lewin's Genes IX (Advanced - the title changes based on the edition, so older ones will be called Genes V, VI, etc). Keep in mind with genetics I almost exclusively relied on class notes.
-Biochemistry: Lehninger Principles of Biochemistry.

I'm sure in a few months I'll have some more constructive offerings. Most of these are very expensive and you should definitely not infringe any copyrights by trying to download them on BitTorrent, by the way.


LIST OF TEXTBOOKS FOR CHEM:
-Organic Chemistry by Thomas N. Sorrell. Good for understanding what the hell is going on.
-Other than that I just end up using MIT OpenCourseWare and http://www.organic-chemistry.org a lot. Not many of my upper-level chemistry classes really relied on a textbook, instead preferring, you guessed it, class notes.


As a final addendum, LIST OF TEXTBOOKS FOR MATH:
-Serge Lang's Algebra, for Abstract Algebra (and I think Galois theory?)
-Free Probability textbook yaaay: http://www.dartmouth.edu/~chance/teachi ... /book.html
-Silverman's A Friendly Introduction to Number Theory, and the more advanced and dense Apostol's Introduction to Analytic Number Theory (I have the pdf)
-I cant remember the topology or analysis textbooks my friends like, and I never took them, if someone wants to take this
Last edited by BuzzerZen on Sun Jun 14, 2009 10:46 am, edited 1 time in total.
Reason: hsquzibowl does not condone illegal filesharing. nope.
Eric Mukherjee, MD PhD
Brown University, 2009
Perelman School of Medicine at the University of Pennsylvania, 2018
Medicine Intern, Yale-Waterbury, 2018-9
Dermatology Resident, Vanderbilt University Medical Center, 2019-

Writer, NAQT, NHBB, IQBT, ACF, PACE
User avatar
theMoMA
Forums Staff: Administrator
Posts: 5802
Joined: Mon Oct 23, 2006 2:00 am

Re: Writing science questions for science players

Post by theMoMA »

I don't know if I can add much to this thread in terms of textbook discussion. I have a very e-dusty copy of Campbell's on some hard drive or another, but that's about it. Since I've been writing science questions more and more regularly lately, I thought I'd create a little addendum for writing (and to an extent, learning) science for those who don't have traditional science classroom backgrounds. Many times, teams don't have the luxury of turning loose a classroom-educated scientist on the fourth of the distribution that science takes up. With the growing "realification" of quizbowl science, it's important for all of us to know where to find good clues and how to write good tossups in that part of the distribution.

The first step is paying attention. You have to put the same effort into listening to and attempting to understand science clues as you do any other part of the distribution. Pyramidality is always going to be a challenge when writing outside your fields of expertise, but it's exacerbated when writers without classroom background write on science. It's fairly obvious that you shouldn't drop "Sal Paradise" into the first line of that tossup on On the Road, but you have to pay attention in order to realize that it's equally egregious to talk about phosphorus early in that Wittig reaction tossup.

Second, I encourage anyone who is trying to become a quizbowl generalist to ask good questions about science things. Quizbowl is full of experts who would be thrilled to tell you what a Fermi level is. Even a layman's understanding of terms will help you understand what's going on in a tossup, or how to parse a dense paragraph of reference material for clue-worthy material.

Finally, you have to be willing to put in work on your own. Packet searches are incredibly helpful for knowing what came up in the past, but they don't always facilitate the kind of understanding that is necessary to create that framework of background information that helps you order clues and convert tossups. You have to be willing to read papers, sections of textbooks, course notes, and the like in order to figure out what's actually going on with the topic at hand. There is no magic trick for finding and identifying good sources; use good judgment about whether the author can be trusted, and if necessary, conduct further research to make sure he information is reliable. If you know someone like Jerry, Eric, Mike, or Gautam won't be playing a tournament, read your questions to them to see if you're on the right track. This is the best way to make sure that the wording of your questions is exactly correct and unambiguous.

In a short time, all of these things work together to produce a network of knowledge that is extremely beneficial not only to writing and playing quizbowl, but to outside applications. One of the most rewarding things I've gotten out of quizbowl is a broad understanding of topics in science that I never would have encountered otherwise. Amusingly, I shocked my ChemE-major roommate by starting a conversation about his column chromatography homework (which actually came in really handy to understanding what the hell theoretical plates were).

It will also really help when it comes to playing science questions. Aside from knowing more bonus parts, you will be able to understand the direction that science tossups are heading, and narrow down possible answers accordingly. Being able to parse the language of tossups to yield even basic understanding is an incredibly important skill that has gotten me more points against really good science players than knowing a bevy of buzzwords.

Sorry to derail the topic a bit. To the point of the thread: If I have written tossups that favor quizbowl-famous over actual knowledge, I apologize. It's hard for me (and the vast majority of people in quizbowl) to know what comes up in science academia, so I just do what I can and hope that the editors have better knowledge of what people actually know.

I guess the only thing I have to say about the topic is that it makes complete sense to me that clues should be rewarding what people actually know, especially when that knowledge is academically important. So it seems perfectly fair that what is emphasized in textbooks should be privileged over stock clues and minor eponymous things. As long as the purpose is to reward important knowledge, and not just learning something through a certain textbook, I see no concerns that using these types of clues is creating "science only for scientists" or whatever.
Andrew Hart
Minnesota alum
User avatar
pray for elves
Auron
Posts: 1060
Joined: Thu Aug 24, 2006 5:58 pm
Location: 20001

Re: Writing science questions for science players

Post by pray for elves »

The Quest for the Historical Mukherjesus wrote:LIST OF TEXTBOOKS FOR MATH:
-Serge Lang's Algebra, for Abstract Algebra (and I think Galois theory?)
-Free Probability textbook yaaay: http://www.dartmouth.edu/~chance/teachi ... /book.html
-Silverman's A Friendly Introduction to Number Theory, and the more advanced and dense Apostol's Introduction to Analytic Number Theory (I have the pdf)
-I cant remember the topology or analysis textbooks my friends like, and I never took them, if someone wants to take this
I'd take Gallian's Contemporary Abstract Algebra as a great starting place, as it's more readable. Artin's book on Galois Theory is the best. For general topology, Munkres's Topology is widely regarded as the best. Apostol also has a good real analysis text (Mathematical Analysis).
Evan
Georgetown Law Alum, Brandeis Alum, Oak Ridge High Alum
Ex-PACE, Ex-ACF
User avatar
The Toad to Wigan Pier
Tidus
Posts: 528
Joined: Mon Oct 10, 2005 6:58 pm
Location: Seattle

Re: Writing science questions for science players

Post by The Toad to Wigan Pier »

The Quest for the Historical Mukherjesus wrote: -Immuno: Kuby's Immunology
Janeway's and Parham's are also very good textbooks
William Butler
UVA '11
Georgia Tech 13
User avatar
setht
Auron
Posts: 1193
Joined: Mon Oct 18, 2004 2:41 pm
Location: Columbus, Ohio

Re: Writing science questions for science players

Post by setht »

DumbJaques wrote:Mike made a post in another thread that I very much agree with - that finding good middle clues is often the hardest part of writing tossups (once you reach a certain level of proficiency)... That's a separate topic of discussion, I suppose.


Yeah, I'm also convinced that there's a pretty good-sized pool of people that have figured out how to write lead-ins and giveaways that are consistently good (or at least acceptable), but haven't figured out how to write consistent middles for tossups. I also think that's a separate topic, since I don't think this is endemic to writing science questions.
DumbJaques wrote:I don't actually know if this overuse of largely trivial leadins is as pronounced in science as I think it might be, but Seth's post suggests that I may be right. Based on said post, I will surely attempt to construct my science people tossups in ways that spend much of the tossup discussing the things they really are best known for, but I'm interested to know if there's a way for me to more readily identify lead-in clues as a non-science person. I'm a sucker for naturalism and don't use the database while writing much anyway, but if I'm reading correctly, it seems like showing up a lot as clues in quizbowl hardly makes something important in a scientific sense. I liked reading through that book with the dead cat on the back, but that's where my science textbook experiences are going to end. So I'm interested in what Seth (or Jerry/Mike/Eric/etc.) might suggest on how to go looking for those clues.


I'm actually not as concerned about lead-ins in science tossups as I am about middles--after all, there are also plenty of non-science tossups with hopelessly hard/over-specialized lead-ins. In any case, I think the "classes-defined science canon" (as opposed to the "packets-defined science canon" Bruce seems to be advocating) I described above should be a very useful source for middle clues (and hopefully some lead-ins too). Suppose you want to write a physics tossup: if you don't really have time, I guess I'd recommend browsing Wikipedia, maybe the David Darling science encyclopedias, etc., but chances are you're not going to write a good question (science or not) if you don't have time to put into it. If you do have time, I would recommend looking up a physics department (I think it doesn't really matter which one you pick, since I think there's lots of overlap in course listings and material covered) and finding a class that has a website with a syllabus you can look at, or course notes or the like. That should pop up a decent number of possible answers. Depending on what area you choose an answer from, you may have multiple courses with possibly interesting material--for instance, if I go to the physics department webpage for UMD and pick something in electromagnetism, it looks like courses 411, 375, 272, 270 and 260 might all be of interest (and possibly others). It also looks like there are often webpages for one or more iterations of each of those UMD courses as taught by different people. If you try a department page and aren't finding much publicly-available material, go to another university. It looks like UMD has a fair amount of stuff freely available online; I know UC Berkeley does as well, and I'm sure there are many other universities with good material.

Textbooks are a second good source. There are some fairly standard texts out there, but I don't think it matters all that much which one(s) you pick out, just as I don't think it matters all that much which particular course website you take material from. If you go to course websites they'll generally show which books are being used. If you have access to those books they should be good; if not, just try to find another book that seems to be similar in scope and depth, I guess. If you pick up a book and don't like it, try another one.
DumbJaques wrote:It also strikes me as somewhat amusing that, in our original communal attempt to shift from fake science clues on trivial things like biography that enabled non-science people to beat more knowledge science players, we have now arrived at a point where we may be going so overboard with "hardcore" academic clues that we've to some degree leveled the playing field once again - and now the science people are out to deprive us of our deliciously fake buzzes once again. You people are all a bunch of haters. Science clues now, science clues tomorrow, science clues forever, right Teitler? You make me sick.
I'm giving myself a pat on the back.

-Seth
Seth Teitler
Formerly UC Berkeley and U. Chicago
President of NAQT
Emeritus member of ACF
User avatar
setht
Auron
Posts: 1193
Joined: Mon Oct 18, 2004 2:41 pm
Location: Columbus, Ohio

Re: Writing science questions for science players

Post by setht »

grapesmoker wrote:It's not so much that we've gone overboard with "hardcore academic clues," it's that a lot of clues now being used are of very remote significance to anyone who doesn't work in the subfield to which they belong. Take tossups on the Stark effect, for example: I'll be damned if I've ever used that Autler-Townes effect (I believe it's used in spectroscopy) but pretty much anyone who's been reading packets for the last few years, scientist or not, knows that when they hear that it's time to buzz with "Stark." So as you yourself pointed out, this produces anomalous results like you beating Mike to a Debye tossup even though it's obvious that Mike's knowledge of all things Debye is way better. I think this is more of a problem with fill-in-the-blank questions than with other types of questions.
I've also never used the Autler-Townes effect, and I've never covered it in any class (in fact, I think I've had very little class time spent on the Stark effect). I would guess that there are now a lot more good college players that can associate Autler-Townes with Stark (and maybe even say "AC/dynamic Stark," even if they don't know what that means) than can give a quick description of the variational principle or link it to any other buzzwords. I think it's fine that quizbowl isn't a perfect reflection of what people learn in classes, both in science and in other categories; the main point I'm trying to make on this front is that "what people learn in science classes" really is so uniform that it represents a very good resource for writing questions that will reliably reward the most knowledgeable people.

-Seth
Seth Teitler
Formerly UC Berkeley and U. Chicago
President of NAQT
Emeritus member of ACF
User avatar
setht
Auron
Posts: 1193
Joined: Mon Oct 18, 2004 2:41 pm
Location: Columbus, Ohio

Re: Writing science questions for science players

Post by setht »

DumbJaques wrote:I think Eric's post in the Missouri Open thread demonstrated that, even when you diligently research a host of minor but important eponymous clues, it often ends up creating the "buzz when you name-associate" problem anyway. I certainly don't want to see a movement away from tossups on scientists as I find the information in these questions very interesting and think it keeps things accessible towards the end even with higher difficulty subjects.
I guess one thing I would urge writers to do is redirect their question-writing research efforts a bit: rather than diligently digging up a host of minor but important eponymous clues, dig up one or two, then spend more time digging up middle clues. In the case of science questions, some good sources for middle clues are important topics that show up in science classes and textbooks. If you pick the right kind of answer, it'll be something that science students come back to again and again as they progress through classes, so you should be able to find course material (notes, textbooks, whatever) from advanced and intermediate classes to furnish middle clues.

If writers do a better job of furnishing middle clues that come from typical course work, I see no problem with going for "fill in the blank" type giveaways (or late middle clues) when those are the best available giveaways (or late middle clues). For instance, I think Eric's Missouri Open tossup on Kramers probably falls into this category, since I don't imagine there are enough players that know any "Kramers science" well enough to come up with Kramers just off of science context without any "fill in the blank" clues.

-Seth
Seth Teitler
Formerly UC Berkeley and U. Chicago
President of NAQT
Emeritus member of ACF
wd4gdz
Tidus
Posts: 740
Joined: Mon Jan 19, 2004 5:40 pm
Location: Tallahassee

Re: Writing science questions for science players

Post by wd4gdz »

I can't wait for Ryan Westbrook's tossup on the Autler-Townes effect.
Billy Beyer, formerly of FSU
User avatar
setht
Auron
Posts: 1193
Joined: Mon Oct 18, 2004 2:41 pm
Location: Columbus, Ohio

Re: Writing science questions for science players

Post by setht »

DumbJaques wrote:My point was that even when you adhere to the real scientific relevance clue criteria, it seems, you can end up with stuff that has the same practical effect if they end up just not being known in quizbowl. Since we've already established that previous packet occurrence is not a great metric for evaluating actual scientific accessibility, then you can't necessarily look at something that seems scientifically relevant and check that based on its presence in the packet archive. Certainly, as you said, these things were amplified greatly (perhaps from insignificance) by the excessive difficulty of the answer.
I think we're on the same page, here. I'll say this: I believe that most of the clues that have shown up in science questions have some scientific relevance, even if they don't show up in typical class sequences that majors would take (undergraduate or graduate). I think that "this clue is relevant to the field in question" isn't enough to justify using a clue in a quizbowl question--there has to be some consideration of whether the clue is one that will be known to quizbowlers in general (and to players with good backgrounds in the field in particular). If we wrote history questions where the first 2/3 of the clues required research-in-that-specific-area level knowledge, history players would be up in arms--or there'd be no more history players since everyone would realize they should focus their efforts on other parts of the distribution. I think the problem is worse in the case of science in the sense that it generally takes more background work to be able to understand what's going on with cutting-edge research than in fields like literature or history. I could be wrong about that, but I think that's the case.

I agree that previous packet occurrence is not a great metric for deciding whether a particular science answer is something science players will know much about, especially at lower levels. For low-level events, I really think going to stuff like course notes/syllabi/typical textbooks is the only good, reliable way to figure out what's reasonable to ask about. I suppose at high-level events there may be some need to go to packet searches to figure out whether certain clues have shown up so much in recent events that they are significantly easier than one would imagine from where they show up in a typical sequence of science classes (e.g., Autler-Townes).

-Seth
Seth Teitler
Formerly UC Berkeley and U. Chicago
President of NAQT
Emeritus member of ACF
User avatar
setht
Auron
Posts: 1193
Joined: Mon Oct 18, 2004 2:41 pm
Location: Columbus, Ohio

Re: Writing science questions for science players

Post by setht »

Tower Monarch wrote:
Whig's Boson wrote:The point of a canon, to me, is that people can read old packets and improve. A canon based on anything else adds information costs to the process of improving; not only must you look at old packets, you must also look at textbooks or take classes.
While I think the point has been made clearly that at face this sounds dumb, I would like to point out that it contains some necessary points. I own around 15 science-related textbooks from all levels of undergraduate courses (not to mention another 20+ math textbooks from linear algebra up), but just flipping through them reveals few things as interesting as the stuff I can find by typing "superconductor" into the packet archive. If you (meaning Seth/Jerry/Mike/Eric, etc) want to make a huge list of upper-undergraduate and graduate textbooks from which you want to make a decently-sized answerspace that can still contain interesting clues, I will grab some of those and study until I'm beating you on these questions. But, until it's more clear where information can come from, I am actually learning science just as someone like Evan Adams might learn literature: I hear some interesting clue in the middle of a tossup, and I read a paper on it or look at an online tutorial, no matter if it is in the average graduate course on the subject.
It is obviously necessary to avoid some of the issues that have definitely come up recently: even as someone who cannot (and should not, yet, be able to) buzz before the giveaway of upper-level science questions, I notice early buzzes coming from people who probably don't understand what they're doing (I'll use Chris Ray as an example, but there are many players like this) while players who have clearly spent time studying the general field let it go by (I've seen Jerry have to wait until the end of physics questions many times). That is definitely an issue. However, if you want someone such as myself (or Andy Watkins, if you want someone who has already made significant progress) to become a high-level player as easily as someone else could become highly competitive at history or whatever, then there needs to be a way to do this. Right now, that way is to read old packets and decipher the context for each clue. With the new model, the way seems less clear: unless I am missing something, you're asking me to double my textbook collection and maybe get tossups in the middle...
Cameron, I think what you're saying is that it's hard to find really cool-sounding nuggets of information to put into science questions. I agree--I've written lots of science questions over the years, and I still frequently find myself taking longer to write science questions than questions in other subjects. Having said that, if someone writing a science question faces a choice between producing a question chock-full of clues that they think are interesting but no one will know (or some people will know only because those same clues have appeared in previous quizbowl questions) and producing a question chock-full of clues that seem less exciting but do a really good job of advantaging people with knowledge, I think the latter type of question is the right choice to make--except perhaps for a vanity event where answerability matters less than cool-sounding clues. I think there are plenty of good, juicy science clues that would serve both ends (cool-sounding clues and rewarding a non-empty set of people with knowledge), but if a writer can't find those clues I'll take boring clues that reward non-packet-based knowledge over the other stuff.

I don't think there's any need for you to ask me, Jerry, Mike, Eric, or anyone else for reading lists: if you're serious about this, go look at course websites. If you go through a couple universities' physics department websites, I'm very confident you'll have more than enough physics textbook recommendations.

I sympathize with wanting to get results fast, but I'm not sure that's ever going to be as easy with science as it is with most (possibly all) other areas. I think the study model you've outlined (pick out interesting clues from questions and study them) is okay, but I'm not sure that's actually easier or more time-efficient than just reading through sets of course notes or textbooks or whatever. It also seems like it might mislead you (and Andy Watkins and whoever else) into spending time looking up material that is really specialized and is unlikely ever to come up again--or, worse yet, it might mislead you (and whoever else) into writing lots of questions down the road featuring material that people either don't know at all, or only know because "it shows up in quizbowl" (and at that point, we're back to Fakey McFakerton beating all the science students to the science questions).

I'm also not sure your study model is actually going to do a better job of preparing you for the kind of questions I'm railing against, compared with how my study model (learn some science in a progressive fashion, so you actually understand what's going on at each step) does in preparing people for the kind of questions I'm promoting. Unless I'm missing something, right now you're reading through large archives of packets, and you're getting some questions in the middle when a writer regurgitates the same clue some previous writer used. If writers start writing science questions with new middle clues, you're SOL; if they don't, everyone memorizes the same limited set of middle clues and you're buzzer racing in the middles of most science questions. If instead you learn a bunch of science in a way that lets you really understand what's going on, and writers start writing science questions of the sort I'm promoting, it should be possible for you to get science questions consistently ahead of people that haven't put in the work you have--because I'm talking about a large number of lead-in and middle clues with science context that can be presented in multiple ways, making it very hard for people to memorize a limited set of middle clues and run with that.

-Seth
Seth Teitler
Formerly UC Berkeley and U. Chicago
President of NAQT
Emeritus member of ACF
User avatar
Important Bird Area
Forums Staff: Administrator
Posts: 5715
Joined: Thu Aug 28, 2003 3:33 pm
Location: San Francisco Bay Area
Contact:

Re: Writing science questions for science players

Post by Important Bird Area »

setht wrote:If we wrote history questions where the first 2/3 of the clues required research-in-that-specific-area level knowledge, history players would be up in arms--or there'd be no more history players since everyone would realize they should focus their efforts on other parts of the distribution.
Not to derail the science thread: but this is simply a particular case of the general principle "don't write ridiculous unbuzzable leadins, because then you will create late-clue difficulty cliff buzzer races." There simply aren't research-specific history clues that are buzzable by anyone, because there is no large population of players doing research in any given area of history. History leadins instead either use really influential historiography, or use bits of evidence that would show up in an advanced undergraduate class or masters-level graduate class regardless of the historiography cited.
Jeff Hoppes
President, Northern California Quiz Bowl Alliance
former HSQB Chief Admin (2012-13)
VP for Communication and history subject editor, NAQT
Editor emeritus, ACF

"I wish to make some kind of joke about Jeff's love of birds, but I always fear he'll turn them on me Hitchcock-style." -Fred
User avatar
setht
Auron
Posts: 1193
Joined: Mon Oct 18, 2004 2:41 pm
Location: Columbus, Ohio

Re: Writing science questions for science players

Post by setht »

DumbJaques wrote:I challenge that my ability to remember that some random modification of a certain reaction has come up before is somehow less legitimate than anyone else remembering the title of somebody's only play, or that somebody won a certain battle, etc. You should make sure that people can't fraud a clue faster than someone who really knows it can answer it (and should obviously address things like obvious linguistic fraud or whatever), but if you're proposing somehow actively stripping tossups of proper names in a quest to rob non-science players of buzzes, well, that's silly! Besides, I'm not really sure what it is that I (or those other people) do on science questions, but whatever it is, it strikes me that science people should be able to do it too (but better, since they have the knowledge). I guess I'm unclear as to what you're citing here as problematic in science questions - or perhaps questions in general - that we for some reason ought to address. It's not like quizbowl is beset by a systemic problem of me beating Eric Mukherjee to chem questions.
I agree that there isn't anything qualitatively different between remembering some names that associate with other names in a science question as opposed to a literature question. I also agree that we don't need to strip all science questions of proper names until the last line from now on, but I do think we are often devoting too much space in science questions to clues whose main (or sole) commendation is that they involve some people's names.

-Seth
Seth Teitler
Formerly UC Berkeley and U. Chicago
President of NAQT
Emeritus member of ACF
Tower Monarch
Rikku
Posts: 360
Joined: Sun Oct 07, 2007 6:23 pm
Location: Richmond, VA

Re: Writing science questions for science players

Post by Tower Monarch »

On a slightly different note, while we are talking about science writing: Are there any suggestions for subdistributions across tournaments within broad fields like physics, chemistry or biology? I ask because I will be writing most of my portion of science for CaTO/TaCO in the next two weeks and beyond a set distribution among those three, Math, CS and Earth & Space Sci, I can ask pretty much anything I want.
My question is, about how much of the, say 15/15, Physics (substitute any other field of science) across a tournament should be Quantum Mechanics vs Thermodynamics (or some other subfield)? I think if we emphasize using textbooks (that likely have somewhat specific subjects) as sources rather than packet archives (which usually don't separate beyond "Physics" or "Chemistry"), there ought to be a few more guidelines as to how many questions in one tournament should come out of a book with a specific title like "Classical Theory of Fields."
I remember Seth saying "who else is going to write geophysics" when he read his Emergency question on that topic, and if it is (I don't know whether it is or not) an important enough field, it obviously deserves more than the 1/0 or 0/1 that he is willing to submit to any given tournament. There must be other examples of fields that are under-/overrepresented in quizbowl versus their academic weight, and to reward the most "real" knowledge possible, that should be corrected as much as possible. Any thoughts?
Cameron Orth - Freelance Writer/Moderator, PACE member
College: JTCC 2011, Dartmouth College '09-'10, '11-'14
Mathematics, Computer Science and Film/Media Studies
High School: Home Schooled/Cosby High '08-'09, MLWGSGIS A-E '06-'08
User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

Re: Writing science questions for science players

Post by grapesmoker »

The standard physics areas covered by any plausible curriculum are statistical mechanics, quantum mechanics, electrodynamics, and classical mechanics. After that, people typically take electives; popular choices include mathematical methods, solid state physics, astrophysics (either cosmology or stellar interiors), plasma physics, and various math and engineering classes (the latter less popular). I think it's not unreasonable to suggest that these topics be roughly equally represented. Diversity is important just like in other categories, but don't feel confined to the categories I've listed.
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance
User avatar
DumbJaques
Forums Staff: Administrator
Posts: 3086
Joined: Wed Apr 21, 2004 6:21 pm
Location: Columbus, OH

Re: Writing science questions for science players

Post by DumbJaques »

I remember Seth saying "who else is going to write geophysics" when he read his Emergency question on that topic, and if it is (I don't know whether it is or not) an important enough field, it obviously deserves more than the 1/0 or 0/1 that he is willing to submit to any given tournament
Well, every tournament really does need more geophysics. I also think that at the higher levels, people sometimes skimp on biology topics like infectious diseases and major organs. I would imagine people who study anatomy academically focus more on knowing a lot of things about important stuff (the body of MDs who know fairly substantial amounts about the liver or thyroid presumably dwarfs the amount who know much at all about Lesch-Nyhan syndrome, but I guess I don't really know).

Actually, I'd say the number one issue in terms of overlooked areas arises from things not fitting in neatly to one category. If you wrote a purely biochem question on cholera, you're never going to cover cholera's fairly substantial history (which itself might be divided into more science-based and more "true" history-based). Meanwhile people are rarely even attempting to talk about history of science within the history distribution (and I'm not really suggesting that they should - it's similarly hard to write comprehensive tossups on those things without getting appropriately sciency). But the result is some stuff that's academically studied by MDs and historians alike getting largely overlooked.

EDIT: To prevent de-railing this thread, I've moved what I originally wrote here to this post
Chris Ray
OSU
University of Chicago, 2016
University of Maryland, 2014
ACF, PACE
User avatar
Mechanical Beasts
Banned Cheater
Posts: 5673
Joined: Thu Jun 08, 2006 10:50 pm

Re: Writing science questions for science players

Post by Mechanical Beasts »

DumbJaques wrote:Lesch-Nyhan syndrome
is one of the major major major X-linked diseases you study in (generally intro) biology classes. It's certainly harder than a tossup on muscular dystrophy or scurvy, because while everyone has felt like a workout, not everyone feels driven to bite his lips off. This is one of those things that feels naturally important to a biologist and felt naturally important to me once Hannah linked me to references since I'm not really a biologist, and feels obscure to non-scientists.

What I wanted to ask more, Jerry, is how to deal with engineering. That is, besides tossups on things you would learn in a circuits or ee class or whatever, how can you put things specific to engineering in the cluespace or answerspace? I'm sympathetic to a lot of marginalized fields, particularly because some neat chemistry comes from those fields (inorganic polymers? inorganic small-molecule drugs? gimme more!) and because I will gasp and leave the room unless quizbowl's one millionth tossup on the Claisen condensation contains a new, fresh clue somewhere (and those areas are great to explore for fresh, new clues).
Andrew Watkins
User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

Re: Writing science questions for science players

Post by grapesmoker »

DumbJaques wrote:Actually, I'd say the number one issue in terms of overlooked areas arises from things not fitting in neatly to one category. If you wrote a purely biochem question on cholera, you're never going to cover cholera's fairly substantial history (which itself might be divided into more science-based and more "true" history-based). Meanwhile people are rarely even attempting to talk about history of science within the history distribution (and I'm not really suggesting that they should - it's similarly hard to write comprehensive tossups on those things without getting appropriately sciency). But the result is some stuff that's academically studied by MDs and historians alike getting largely overlooked.
Insofar as this relates to science, I've argued before that there is no reason not to write, say, a more medical-like tossup on cholera (or what have you). It's just that the population of people writing bio and chem questions comes largely from a molecular rather than a clinical background (and clinical clues are often difficult to write well, since many symptoms may not be uniquely identifying). Also, I'm not sure what you mean by cholera's history. It would seem that if you wanted to write a history question on a disease, you should be doing that; otherwise, I'm not sure I see the connection.
everyday847 wrote:What I wanted to ask more, Jerry, is how to deal with engineering. That is, besides tossups on things you would learn in a circuits or ee class or whatever, how can you put things specific to engineering in the cluespace or answerspace?
I'm somewhat confused by your question. Are you asking how one might incorporate engineering-type clues into physics tossups, for example? If that's your question, the answer is that there's obviously a good deal of cross-pollination between physics and engineering, with much of the distinction being largely academic rather than substantive. For example, the quantum Hall effect was discovered in the inversion layer of a MOSFET, so there's some good material for asking about QHE, MOSFETs, or inversion layers. Other than that, I would say that if you want to see more engineering content in the science distribution, you'd probably do well to just ask about them. A lot of the physics distribution tends to be quite theoretical as, in my experience, experimental physics is harder to write about, so that will probably prevent some cross-usage of clues.

As long as we're on this topic, allow me to rant a little bit about questions on experiments. There are some pivotal experiments in physics that every physicist learns about: Franck-Hertz, Davisson-Germer, and Stern-Gerlach can be considered the main foundational experiments of early QM. Other notable experiments are Michelson-Morley, Pound-Rebka, C.S. Wu's parity violation experiment, and Cronin and Fitch's CP violation experiment are others that come up and that people should know about. On the other hand, derivative experiments like Trouton-Noble and Trouton-Rankine don't seem particularly important to me and they certainly aren't discussed in most classes; they're interesting in the context of their time but today they're mostly historical curiosities. My general opinion is that these are not particularly noteworthy things and probably shouldn't be written about. Attendant to this main point is a secondary one, which is that lots of people start questions on better-known experiments with a discussion of the setup, e.g. "This experiment used a wire-mesh for Faraday shielding." Well, this is a very bad way to start such a question for the simple reason that lots of experiments used Faraday shielding (if you're working with electrodynamic interactions, you'd better shield your experiment), and in general, many clues about experimental setups are not unique at all. Those that are unique (e.g. silver atoms for SG, nickel target for DG, etc.) are too easily identifiable. My recommendation is that people use the more famous experiments as easy or medium bonus parts and connect them to other interesting physics rather than tossup material, especially at the level of ACF Regionals or above.
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance
User avatar
Mechanical Beasts
Banned Cheater
Posts: 5673
Joined: Thu Jun 08, 2006 10:50 pm

Re: Writing science questions for science players

Post by Mechanical Beasts »

Yeah; that basically answered my question. I wanted to know, I guess, if there were (somehow) some sort of engineering-specific topics that don't have high overlap with canonical physics that are important in their own right.

As to your point about experimental setup clues, this is one of those times where it really pays to do some research on what your clues mean. I bet that clue could find itself in a tossup if a writer thinks that using a wire-mesh for Faraday shielding is an innovation unique to experiment x, which I could imagine someone believing even if they infer that Faraday shielding is hardly a unique goal. Two minutes of research shows that saying that is like saying that the experiment used paper towels for their water-absorbing powers.

(I may make a thread about saying what you mean, and knowing what you mean, in chemistry tossups (particularly, but not exclusively, organic) a bit later, since this problem is pretty big there.)
Andrew Watkins
User avatar
Lagotto Romagnolo
Tidus
Posts: 533
Joined: Wed Jan 23, 2008 12:43 pm
Location: Alexandria, VA

Re: Writing science questions for science players

Post by Lagotto Romagnolo »

everyday847 wrote:I wanted to know, I guess, if there were (somehow) some sort of engineering-specific topics that don't have high overlap with canonical physics that are important in their own right.
I've been considering this for a while now. Jerry's definitely right in that there is a large amount of overlap between physics and engineering (e.g. Maxwell's equations and thermodynamics). But sometimes I get the sense that engineering doesn't lend itself well to quizbowl. Not to say you can't write about it at all (for example, that bonus on the Rankine cycle at nationals was well done).

Still, the engineering-specific tossups that don't overlap with physics can be a problem. In my circuits course last semester, we spent well over a month studying techniques like node-voltage analysis, mesh analysis, and source transformations, all of which are important tools. But I imagine it would be hard to write a solid tossup on those topics (though if anyone can prove me wrong, I'd love to see such a tossup).

I suppose part of the problem is that engineering is based on creative design, more so than other sciences; you can't exactly ask someone to design a catapult as part of a bonus.

Still, I wouldn't mind seeing more engineering clues within pure science questions, like Jerry's MOSFET example above. Any thoughts?
Aaron Rosenberg
Langley HS '07 / Brown '11 / Illinois '14
PACE
User avatar
Captain Sinico
Auron
Posts: 2869
Joined: Sun Sep 21, 2003 1:46 pm
Location: Champaign, Illinois

Re: Writing science questions for science players

Post by Captain Sinico »

As an engineer, I occasionally write questions and see questions by others on per se engineering topics. In the past couple tournaments I played, off the top of my head I recall having written a bonus on structural engineering and heard questions on the Antoine correlation, Moody diagram, and Thevénin's theorem.
That said, I find it's more useful in most cases frequently to write on the science concepts underlying those topics and only infrequently to write on those topics themselves. For example, more or less the entire field of electrodynamics underlies those circuit theory topics you've cited, so questions on Maxwell's equations, Kirchoff's laws, or any number of other topics can be and sometimes are (by me) approached from a circuit-theory perspective in a question.

MaS
Mike Sorice
Coach, Centennial High School of Champaign, IL (2014-2020) & Team Illinois (2016-2018)
Alumnus, Illinois ABT (2000-2002; 2003-2009) & Fenwick Scholastic Bowl (1999-2000)
Member, ACF (Emeritus), IHSSBCA, & PACE
User avatar
No Rules Westbrook
Auron
Posts: 1232
Joined: Mon Nov 22, 2004 1:04 pm

Re: Writing science questions for science players

Post by No Rules Westbrook »

Okay, a few general observations.

I hope this thread doesn't create an effect whereby people purposely avoid writing on the work of people like Kramers just because they fear that this is being met with disapproval. Like Seth says, if you spend some time trying to create real middle clues that reward bona fide acquired knowledge, you'll be fine and noone will quibble. It's perfectly fine to create a giveaway that lets people get the tu simply because they know that there is a thing called "Kramers-Kronig" - that's what a giveaway should be.

Anyway, here's the bigger point I want to make. People need to understand that there are inherent limitations and parameters on the game of quizbowl. There are not an infinite number of ways that you can write tossups - certain clues invariably work a lot better than other clues, for reasons which can be fairly unrelated to the academic importance of those clues. Within the context of the game of QB, you have to worry about things like "linguistic transparency" and "clue uniqueness" and "buzzability" - those considerations constrain how you write, and yet they are completely irrelevant to the actual academic study of whatever you're writing on. There will always be a "canon" and players who are focused on memorizing shit will always thrive, because this is a game about buzzing off of clues. I don't care how academically relevant or science-y you make the clues (unless they're positively indecipherable in a science-y sort of way), you can still memorize them without having to know a whole lot of context.

Bruce's first post is a tad simplistic, but it hints at a very good point. There will always be a canon in good quizbowl, it can't be any other way.
Ryan Westbrook, no affiliation whatsoever.

I am pure energy...and as ancient as the cosmos. Feeble creatures, GO!

Left here since birth...forgotten in the river of time...I've had an eternity to...ponder the meaning of things...and now I have an answer!
User avatar
cvdwightw
Auron
Posts: 3446
Joined: Tue May 13, 2003 12:46 am
Location: Southern CA
Contact:

Re: Writing science questions for science players

Post by cvdwightw »

As probably the worst "real science" player in this thread (compared to "fake science" players like Chris that get most of their science knowledge from old packets), I think it gets harder and harder to find good, applicable science-y clues. People are writing on crazy stuff no one cares about because "that thing that came up in my grad-level class" has already been a lead-in three times.

One thing that I do try to do is use my fairly non-traditional background to write leadins from fields that don't get lots of play. For instance, several of my lead-ins have contained stuff from MEMS/microfluidics, because it's stuff I've taken a class on, stuff that's legitimately important, and stuff that doesn't come up in quizbowl all that much. Maybe it's just because I write questions like this, but I think this is the kind of direction that science lead-ins should be moving in - not because it's not "crazy stuff no one is expected to know" but because it's "legitimately important stuff people might learn about if it comes up a couple of times."
Dwight Wynne
socalquizbowl.org
UC Irvine 2008-2013; UCLA 2004-2007; Capistrano Valley High School 2000-2003

"It's a competition, but it's not a sport. On a scale, if football is a 10, then rowing would be a two. One would be Quiz Bowl." --Matt Birk on rowing, SI On Campus, 10/21/03

"If you were my teammate, I would have tossed your ass out the door so fast you'd be emitting Cerenkov radiation, but I'm not classy like Dwight." --Jerry
User avatar
Quantum Mushroom Billiard Hat
Rikku
Posts: 435
Joined: Tue Jun 12, 2007 1:16 pm
Location: Midland, MI

Re: Writing science questions for science players

Post by Quantum Mushroom Billiard Hat »

everyday847 wrote: I wanted to know, I guess, if there were (somehow) some sort of engineering-specific topics that don't have high overlap with canonical physics that are important in their own right.
I am not especially familiar with the physics curriculum; these are things I've run into in my engineering classes (chemical and materials science mostly) that already come up in QB, but could probably be expanded. I am not thinking too hard about whether a good tossup could be constructed out of these for now, only thinking about possible topics. I also haven't checked the archives thoroughly, so I don't know how much has come up before.

Fluid mechanics: There are plenty of things there that seem interesting and reasonable for some tournaments to me beyond Navier-Stokes and laminar flow. Bingham plastics and power-law fluids, for instance, caught my attention as interesting topics, but I don't know how important they are. Flow through packed beds, boundary layers, and flow around falling objects are important.

Stress/strain stuff: Young's Modulus, stress, strain, yield stress, ultimate tensile stress, ductility, and Poisson's Ratio show up pretty often. Couples, stuff about truss structures, and Mohr's circle seem important. Cantilevered beams might be as well.

Material structures: I think I've heard Bravais' Lattices come up before, but they are important. My class spent a long time going through various phases of steel (martensite, austenite, ferrite...) so my professor obviously thought they were important. Elastomers, thermoplasts, and thermosets also seem really important.

Thermo: Raoult's and Henry's Laws show up a lot. Fugacity comes up. Both it and activity coefficients must be important based on the time we spent learning them, although I haven't quite figured how yet. Equations of state are important, but very hard to ask about in a uniquely-identifying way. The Gibbs Phase Rule seems important, but might be hard to write a tossup on. Power and refrigeration cycles are important, as mentioned earlier.

That's all I remember at the moment
Michael Hausinger
Coach, Bay City Western High School
formerly of University of Michigan and East Lansing High School
User avatar
Quantum Mushroom Billiard Hat
Rikku
Posts: 435
Joined: Tue Jun 12, 2007 1:16 pm
Location: Midland, MI

Re: Writing science questions for science players

Post by Quantum Mushroom Billiard Hat »

squareroot165 wrote:
everyday847 wrote: I wanted to know, I guess, if there were (somehow) some sort of engineering-specific topics that don't have high overlap with canonical physics that are important in their own right.
I am not especially familiar with the physics curriculum; these are things I've run into in my engineering classes (chemical and materials science mostly) that already come up in QB, but could probably be expanded. I am not thinking too hard about whether a good tossup could be constructed out of these for now, only thinking about possible topics. I also haven't checked the archives thoroughly, so I don't know how much has come up before.

Fluid mechanics: There are plenty of things there that seem interesting and reasonable for some tournaments to me beyond Navier-Stokes and laminar flow. Bingham plastics and power-law fluids, for instance, caught my attention as interesting topics, but I don't know how important they are. Flow through packed beds, boundary layers, and flow around falling objects are important.

Stress/strain stuff: Young's Modulus, stress, strain, yield stress, ultimate tensile stress, ductility, and Poisson's Ratio show up pretty often. Couples, stuff about truss structures, and Mohr's circle seem important. Cantilevered beams might be as well.

Material structures: I think I've heard Bravais' Lattices come up before, but they are important. My class spent a long time going through various [url-http://en.wikipedia.org/wiki/Template:Steels]phases of steel[/url] (martensite, austenite, ferrite...) so my professor obviously thought they were important. Elastomers, thermoplasts, and thermosetsalso seem really important.

Thermo: Raoult's and Henry's Laws show up a lot. Fugacity comes up. Both it and activity coefficients must be important based on the time we spent learning them, although I haven't quite figured how yet. Equations of state are important, but very hard to ask about in a uniquely-identifying way. The Gibbs Phase Rule seems important, but might be hard to write a tossup on. Power and refrigeration cycles are important, as mentioned earlier.

That's all I remember at the moment

Edit for more links. Also, liquid-glass transition temperature seemed pretty cool.
Edit 2: Crystal defects seem important too. Doping, band gaps, and related stuff in semiconductors comes up, and is really important too.
Michael Hausinger
Coach, Bay City Western High School
formerly of University of Michigan and East Lansing High School
User avatar
cvdwightw
Auron
Posts: 3446
Joined: Tue May 13, 2003 12:46 am
Location: Southern CA
Contact:

Re: Writing science questions for science players

Post by cvdwightw »

squareroot165 wrote:Fluid mechanics: There are plenty of things there that seem interesting and reasonable for some tournaments to me beyond Navier-Stokes and laminar flow. Bingham plastics
We're not at the point where we can start tossuping specific types of non-Newtonian fluids at anything below ACF Nationals. This would make a decent third part of a Regionals-level bonus (viscosity/non-Newtonian/Bingham plastics).
squareroot165 wrote:Stress/strain stuff: Young's Modulus, stress, strain, yield stress, ultimate tensile stress, ductility, and Poisson's Ratio show up pretty often.
Yeah, this stuff is pretty important. But they're not the easiest things in the world to write good, non-transparent tossups on (certainly Poisson's Ratio would be better off as a solid, important middle clue on most of the more generic things than as an answer in and of itself).
squareroot165 wrote:Material structures: I think I've heard Bravais' Lattices come up before, but they are important.
I believe that Bragg's Law/Bravais Lattices/Miller Indices has been described by Ray Luo as the "only three answers" on a crystallography bonus (though, incidentally, he's wrong - there's plenty of other stuff you can ask about, some of which has been mentioned). It's really hard to write non-transparently on Bravais Lattices or Miller Indices.
Dwight Wynne
socalquizbowl.org
UC Irvine 2008-2013; UCLA 2004-2007; Capistrano Valley High School 2000-2003

"It's a competition, but it's not a sport. On a scale, if football is a 10, then rowing would be a two. One would be Quiz Bowl." --Matt Birk on rowing, SI On Campus, 10/21/03

"If you were my teammate, I would have tossed your ass out the door so fast you'd be emitting Cerenkov radiation, but I'm not classy like Dwight." --Jerry
User avatar
Quantum Mushroom Billiard Hat
Rikku
Posts: 435
Joined: Tue Jun 12, 2007 1:16 pm
Location: Midland, MI

Re: Writing science questions for science players

Post by Quantum Mushroom Billiard Hat »

cvdwightw wrote:
squareroot165 wrote:Fluid mechanics: There are plenty of things there that seem interesting and reasonable for some tournaments to me beyond Navier-Stokes and laminar flow. Bingham plastics
We're not at the point where we can start tossuping specific types of non-Newtonian fluids at anything below ACF Nationals. This would make a decent third part of a Regionals-level bonus (viscosity/non-Newtonian/Bingham plastics).
squareroot165 wrote:Stress/strain stuff: Young's Modulus, stress, strain, yield stress, ultimate tensile stress, ductility, and Poisson's Ratio show up pretty often.
Yeah, this stuff is pretty important. But they're not the easiest things in the world to write good, non-transparent tossups on (certainly Poisson's Ratio would be better off as a solid, important middle clue on most of the more generic things than as an answer in and of itself).
squareroot165 wrote:Material structures: I think I've heard Bravais' Lattices come up before, but they are important.
I believe that Bragg's Law/Bravais Lattices/Miller Indices has been described by Ray Luo as the "only three answers" on a crystallography bonus (though, incidentally, he's wrong - there's plenty of other stuff you can ask about, some of which has been mentioned). It's really hard to write non-transparently on Bravais Lattices or Miller Indices.
Agreed. I just wanted to throw a few more topics out there.
Michael Hausinger
Coach, Bay City Western High School
formerly of University of Michigan and East Lansing High School
User avatar
No Rules Westbrook
Auron
Posts: 1232
Joined: Mon Nov 22, 2004 1:04 pm

Re: Writing science questions for science players

Post by No Rules Westbrook »

Dwight brings up a good point that applies to science and a few other categories at higher levels of QB.

The canon is getting big enough that its increasingly really tough to meet all of the demands of good question writing. You can choose to write yet another Meissner Effect or Stark Effect tossup - but at some point, the usable clues kind of run out. The fact that we live in a "public archive era" of quizbowl with a fair number of good players who read and keep up with tournaments presents a continual challenge to writing pyramidal and fresh tossups - it's perfectly easy to write ass-hard/unbuzzable clues and it's perfectly easy to use clues that have come up before, but finding new and yet buzzable/important clues - that's not so easy these days. Some subjects, like literature, are fairly immune to this effect - but others, like science, just have a more limited answer and clue space.
Ryan Westbrook, no affiliation whatsoever.

I am pure energy...and as ancient as the cosmos. Feeble creatures, GO!

Left here since birth...forgotten in the river of time...I've had an eternity to...ponder the meaning of things...and now I have an answer!
User avatar
setht
Auron
Posts: 1193
Joined: Mon Oct 18, 2004 2:41 pm
Location: Columbus, Ohio

Re: Writing science questions for science players

Post by setht »

No Rules Westbrook wrote:Dwight brings up a good point that applies to science and a few other categories at higher levels of QB.

The canon is getting big enough that its increasingly really tough to meet all of the demands of good question writing. You can choose to write yet another Meissner Effect or Stark Effect tossup - but at some point, the usable clues kind of run out. The fact that we live in a "public archive era" of quizbowl with a fair number of good players who read and keep up with tournaments presents a continual challenge to writing pyramidal and fresh tossups - it's perfectly easy to write ass-hard/unbuzzable clues and it's perfectly easy to use clues that have come up before, but finding new and yet buzzable/important clues - that's not so easy these days. Some subjects, like literature, are fairly immune to this effect - but others, like science, just have a more limited answer and clue space.
I think Ryan's touched on some of the issues that are resulting in science questions that advanced science students can't buzz on: writers sometimes prioritize finding completely new answers/clues over finding buzzable/important answers/clues. I'll add that I think writers sometimes look preferentially for new clues that involve proper names, and I think this is not a good strategy for producing decent science questions. Instead, I think writers that have little or no background in a particular field of science should approach writing a question in that field with the mentality that the goal isn't to try to produce a really exciting tossup on an answer near the edges of the quizbowl science canon; rather, the goal is to produce a competent tossup that will either be fine as submitted or can easily be edited to decency by a competent editor. A solidly canonical answer (canonical in a curriculum-based sense) is much more likely to have lots of buzzable/important clues available, which means it's more likely that a non-knowledgeable writer will do a decent job to start with, and also more likely that a competent editor will be able to turn it into a decent question quickly, if it isn't fine to start with. Speaking for myself, I have no objection to hearing high-level tossups on "pressure" or "Fermi" or "the Milky Way." In fact, I'd rather hear competent tossups on any of those answers than an ill-advised tossup on "the Pockels effect."

Taking Ryan's comparison of science and literature and running with it for a moment: the really nice thing about writing literature questions is that lots of quizbowlers read extensively, and many works of literature have lots of possible middle clues (important events, memorable secondary characters, famous line, whatever). This makes it relatively easy to write half a dozen tossups on, say, Mrs. Dalloway without feeling like all the possible lead-ins and middle clues have been used up (repeated giveaways don't bother me). At the same time, there are some works that I think we can all agree would not make for a single good tossup (e.g. Williams' "This Is Just To Say" or Poe's unfinished play Politian)--they might be fine as clues in tossups on related subjects or occasionally as bonus parts, but they are not meant to be part of the canon-expansion process. Moving over to science, I think there are fewer quizbowlers who study science and/or read extensively about science on their own than there are quizbowlers who read lots of literary works. One thing I think this means is that canon-expanding answers in science are much less likely to go over well: just because the Pockels effect has been mentioned six times in the last two years or whatever doesn't mean that a bunch of motivated science players have gone out and learned lots of exciting facts about the Pockels effect. There's a decent chunk of the quizbowl science canon that I think is best left at the "use as a clue in tossups on better-known, related answers, and occasionally as a bonus part" stage. I think the science equivalent of Mrs. Dalloway (a good writer could squeeze out ~6 tossups without using up all the lead-ins/middle clues, and produce questions that science players could buzz on before the end) really has to be a fairly low-level answer. For writers that don't feel confident about picking out which topics near the edges of the quizbowl science canon would actually make for good tossup answers and which should be left at the tossup clue/bonus answer stage, I heartily recommend sticking with safer answers: basic concepts and famous scientists (stick to scientific accomplishments, of course) are generally safe, and plenty of good clue material should be available from almost any introductory textbook or course notes. I don't think this advice is unique to science, actually--if I had to write a philosophy tossup I'd almost certainly do a better job on something like "Hegel" than I would on something like "philosophical zombies" (which may or may not be a really exciting topic for philosophy players, but the chances of my producing a decent tossup on it [if that's even possible] are really slim).

I should say that while I've been touting curriculum-based clue and answer selections, I don't think every science question has to come from that sort of source: if there are interesting news stories about science or popular science books that people are reading about, those could presumably also be good sources for science questions. However, if an advanced physics major with little or no quizbowl background can't get solid buzzes on the majority of the physics questions in a set, I think there's something wrong with that. I don't think we're quite there yet, but I think some of the trends I mentioned earlier in this thread are pointing in that direction. If more question writers approach science writing with a default model* of "write on something safe, and look for fresh clues that people taking classes in this field would encounter (or look for new ways to package/present important clues that have been used before)" I think that will be an improvement.

-Seth

* By "default model" I mean what a question writer does when they don't have some really exciting idea to start with. My impression is that a decent number of writers currently use a "default model" along the lines of "look to the edges of the quizbowl science canon for question ideas."
Seth Teitler
Formerly UC Berkeley and U. Chicago
President of NAQT
Emeritus member of ACF
User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

Re: Writing science questions for science players

Post by grapesmoker »

Just for the record, I think a tossup on philosophical zombies would be pretty great. It's been done at least once, I think, but it's actually something that one would know about if one is interested in contemporary analytic philosophy.

Here's where I am at variance with Seth: I don't want to hear high level tossups on the Milky Way or pressure. Not that those aren't important things, but I think at the higher levels of the game we can allow ourselves a little more variety than that. Those kinds of questions are great for what I envision something like ACF Regionals looking like, but I think at Nationals other interesting topics should be covered. Which brings me to my next point:

People need to recognize that there's a difference between writing for different kinds of tournaments. I've gone on record before (notably during discussion of EFT 2008 and the "law" tossup) as saying that I don't find it problematic that good players can get early buzzes. If noted physicist Seth Teitler beats everyone in the room on a physics tossup by buzzing on the first or second clue, that's not a problem; that's what's supposed to happen. To that effect, some (most) tournaments do not require (and indeed, cannot support) unlimited excavation of new clues for well-established topics (e.g. Stark effect, etc.). That's why for tournaments like CO and ACF Nationals, I think it's fine to branch out into something that's more adventurous, provided it's written in a way that rewards knowledge and isn't just a word soup or a listing of names. There's no shortage of good topics in the sciences that are both advanced enough to be interesting at higher levels and open to being written well.
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance
User avatar
setht
Auron
Posts: 1193
Joined: Mon Oct 18, 2004 2:41 pm
Location: Columbus, Ohio

Re: Writing science questions for science players

Post by setht »

Tower Monarch wrote:On a slightly different note, while we are talking about science writing: Are there any suggestions for subdistributions across tournaments within broad fields like physics, chemistry or biology? I ask because I will be writing most of my portion of science for CaTO/TaCO in the next two weeks and beyond a set distribution among those three, Math, CS and Earth & Space Sci, I can ask pretty much anything I want.
My question is, about how much of the, say 15/15, Physics (substitute any other field of science) across a tournament should be Quantum Mechanics vs Thermodynamics (or some other subfield)? I think if we emphasize using textbooks (that likely have somewhat specific subjects) as sources rather than packet archives (which usually don't separate beyond "Physics" or "Chemistry"), there ought to be a few more guidelines as to how many questions in one tournament should come out of a book with a specific title like "Classical Theory of Fields."
I remember Seth saying "who else is going to write geophysics" when he read his Emergency question on that topic, and if it is (I don't know whether it is or not) an important enough field, it obviously deserves more than the 1/0 or 0/1 that he is willing to submit to any given tournament. There must be other examples of fields that are under-/overrepresented in quizbowl versus their academic weight, and to reward the most "real" knowledge possible, that should be corrected as much as possible. Any thoughts?
I'm in agreement with Jerry on subdistributing physics--I might give a little more weight to quantum and EM over stat. and classical mechanics (and possibly stuff some stat. mech. and maybe some more quantum into physical chemistry), but as long as none of them are absent or overwhelming you should be in good shape.

I've thought for a while that earth science as a whole is underrepresented in quizbowl. I think geophysics should get slightly less coverage than traditional geology, with oceanic and atmospheric sciences each getting less than geophysics (but not zilch). I'm going to guess that the total number of earth science questions will be small enough that the details won't really matter.

In astronomy, I think a curriculum-based distribution would have chunks of stellar physics, radiative transfer, galactic astronomy, cosmology, and optics/instrumentation, plus a decent amount of space for "electives": extragalactic astronomy, astrophysical fluid dynamics/plasmas, interstellar medium (and maybe intergalactic medium), relativistic astrophysics, high energy astrophysics, and compact objects (although this maybe should just be included in stellar physics). There should also be some real planetary science (no questions on the 8th largest moon of Uranus, please); whether that really belongs to astronomy or to earth science is somewhat a question of taste and the clues used. For the purposes of quizbowl science, I think some of those categories should be augmented or diminished: radiative transfer probably shouldn't get nearly as much coverage as it does in a real astronomy curriculum, for instance.

I've come around to the opinion that the math sub-distribution should shift a bit to include more "applied" stuff like statistics, differential equations, and other topics that students might encounter in non-math classes, but that's a topic for some other time: as long as you don't have tons of questions on algebra or analysis (or none) I think the details won't really matter.

-Seth
Seth Teitler
Formerly UC Berkeley and U. Chicago
President of NAQT
Emeritus member of ACF
User avatar
setht
Auron
Posts: 1193
Joined: Mon Oct 18, 2004 2:41 pm
Location: Columbus, Ohio

Re: Writing science questions for science players

Post by setht »

grapesmoker wrote:Just for the record, I think a tossup on philosophical zombies would be pretty great. It's been done at least once, I think, but it's actually something that one would know about if one is interested in contemporary analytic philosophy.

Here's where I am at variance with Seth: I don't want to hear high level tossups on the Milky Way or pressure. Not that those aren't important things, but I think at the higher levels of the game we can allow ourselves a little more variety than that. Those kinds of questions are great for what I envision something like ACF Regionals looking like, but I think at Nationals other interesting topics should be covered. Which brings me to my next point:

People need to recognize that there's a difference between writing for different kinds of tournaments. I've gone on record before (notably during discussion of EFT 2008 and the "law" tossup) as saying that I don't find it problematic that good players can get early buzzes. If noted physicist Seth Teitler beats everyone in the room on a physics tossup by buzzing on the first or second clue, that's not a problem; that's what's supposed to happen. To that effect, some (most) tournaments do not require (and indeed, cannot support) unlimited excavation of new clues for well-established topics (e.g. Stark effect, etc.). That's why for tournaments like CO and ACF Nationals, I think it's fine to branch out into something that's more adventurous, provided it's written in a way that rewards knowledge and isn't just a word soup or a listing of names. There's no shortage of good topics in the sciences that are both advanced enough to be interesting at higher levels and open to being written well.
I'm not saying there can be no questions on more adventurous answers. I am suggesting that writers that don't feel confident about their ability to write a good science question on an adventurous answer should go ahead and play it safe. There are a decent number of teams with good bio/chem/physics writers; I don't imagine that encouraging less-knowledgeable writers to play it safe will spell the death of the good, adventurous science tossup. If a tournament set has 10 "safe" physics tossups and 5 "adventurous" physics tossups, all written competently, I think that's preferable to 10 "adventurous" and 5 "safe" with 5 poorly-written tossups. If a less-knowledgeable writer is willing to put in the time to research and write a good tossup on an adventurous topic, that's great; if not, I'd rather see them write a good tossup on a safe topic.

-Seth
Seth Teitler
Formerly UC Berkeley and U. Chicago
President of NAQT
Emeritus member of ACF
User avatar
Lagotto Romagnolo
Tidus
Posts: 533
Joined: Wed Jan 23, 2008 12:43 pm
Location: Alexandria, VA

Re: Writing science questions for science players

Post by Lagotto Romagnolo »

setht wrote:I've thought for a while that earth science as a whole is underrepresented in quizbowl. I think geophysics should get slightly less coverage than traditional geology, with oceanic and atmospheric sciences each getting less than geophysics (but not zilch)
Earth science is a genuinely important field and probably deserves more than it gets, but some parts of it have a tendency to degenerate into list knowledge, namely the time scale.

Our friend the the QB Wiki claims:

The presence of these materials in a "science" question will usually identify it as Colvin science:

* biographical clues about scientists

* taxonomy

* the geologic time scale

(http://www.doc-ent.com/qbwiki/index.php ... in_science)

It seems like most questions on the time scale tend to take the form of "name this period which came after A and before B. I admit that I once wrote such a tossup on the Devonian period which named a bunch of subdivisions of said period for half the question, which is a similar writer's trap to 'fill in the blank in this multiply eponymous thing'.

Then again, I know nothing about the standard geology curriculum (and I'd be interested in learning what topics are covered), so if the Devonian period is something that is widely studied, perhaps you could write a good tossup on it with important academic clues.
Aaron Rosenberg
Langley HS '07 / Brown '11 / Illinois '14
PACE
User avatar
Theory Of The Leisure Flask
Yuna
Posts: 843
Joined: Fri Aug 29, 2003 11:04 am
Location: Brooklyn

Re: Writing science questions for science players

Post by Theory Of The Leisure Flask »

The Gold Gringo wrote:
setht wrote:I've thought for a while that earth science as a whole is underrepresented in quizbowl. I think geophysics should get slightly less coverage than traditional geology, with oceanic and atmospheric sciences each getting less than geophysics (but not zilch)
Earth science is a genuinely important field and probably deserves more than it gets, but some parts of it have a tendency to degenerate into list knowledge, namely the time scale.

Our friend the the QB Wiki claims:

The presence of these materials in a "science" question will usually identify it as Colvin science:

* biographical clues about scientists

* taxonomy

* the geologic time scale

(http://www.doc-ent.com/qbwiki/index.php ... in_science)
Going only by what my currentinternship has taught me about groundwater, the geologic time scale is genuinely important. Also, I suspect much of the reason so little earth science gets asked is that it's pretty closely connected to geography, and interdisciplinary questions tend to (unfairly IMO) get a bad rap these days.

Granted, the fact that I (a non-scientist) am calling geologic time periods important does nothing to change its status as a Colvin science, and if anything probably confirms said status. But just because it's a Colvin science doesn't mean it's unimportant or trivial.
Chris White
Bloomfield HS (New Jersey) '01, Swarthmore College '05, University of Pennsylvania '10. Still writes questions occasionally.
User avatar
Lagotto Romagnolo
Tidus
Posts: 533
Joined: Wed Jan 23, 2008 12:43 pm
Location: Alexandria, VA

Re: Writing science questions for science players

Post by Lagotto Romagnolo »

Theory Of The Leisure Flask wrote:But just because it's a Colvin science doesn't mean it's unimportant or trivial
Far from it. A science question can't be called SCIENCE just because of the answer choice.
Aaron Rosenberg
Langley HS '07 / Brown '11 / Illinois '14
PACE
User avatar
Important Bird Area
Forums Staff: Administrator
Posts: 5715
Joined: Thu Aug 28, 2003 3:33 pm
Location: San Francisco Bay Area
Contact:

Re: Writing science questions for science players

Post by Important Bird Area »

The Gold Gringo wrote:Our friend the the QB Wiki claims:

The presence of these materials in a "science" question will usually identify it as Colvin science:

* the geologic time scale

It seems like most questions on the time scale tend to take the form of "name this period which came after A and before B. I admit that I once wrote such a tossup on the Devonian period which named a bunch of subdivisions of said period for half the question, which is a similar writer's trap to 'fill in the blank in this multiply eponymous thing'.
I wrote that wiki entry; all I intended to claim was that "many questions with geologic-time-scale subdivisions as answers fell into the traps Aaron mentions, and thus fail to reward real science knowledge." (Not "all questions about this thing are inherently fake.")
Jeff Hoppes
President, Northern California Quiz Bowl Alliance
former HSQB Chief Admin (2012-13)
VP for Communication and history subject editor, NAQT
Editor emeritus, ACF

"I wish to make some kind of joke about Jeff's love of birds, but I always fear he'll turn them on me Hitchcock-style." -Fred
Locked