Minnesota/Penn Open IV Discussion

[Quantum Mushroom Billiard Hat]

I really enjoyed this tournament a lot. A few of the answers seemed unnecessarily hard / overly specific, but in general the really hard questions seemed to be on interesting topics worth learning about. My only real (small) complaint comes is the way the giveaways were structured on tossups like Janus Kinase and Kepler, since the last clue seemed to be completely unrelated to the rest of the tossup. My impression is that a tossup answer should be important and convertable enough to stand on its own without a tangentially related giveaway.

I thought the contents of the Anfinsen's Experiment tossup were an especially good idea, even though I didn't know the name. I also was happy to see tossups on Jim Casey and the Abbe Faria since they test knowledge of books many, many people are likely to have read.

Finally: Eric, as much as I like the BET isotherm, the edited form of my catalyst/adsorption bonus really doesn't have an easy part.

[setht]

From what Eric listed it looks like CS had 3/4, math had 4/4, astro had 7/3 (metallicity was an astro TU that seems to have been pasted into the earth science list by accident), and earth science had 1/4. I think math + CS ~ astro + earth science is a decent rubric, with maybe a little more space for math (in particular, some space for applied math and statistics--stuff that lots of scientists should be learning as tools of the trade). The science subdistribution for MO was clearly a bit imbalanced, but I think within acceptable variation for a packet-submission event.

I guess Eric has already commented on this in connection with the Feit-Thompson question, but: I thought there were some tossups on topics that are real and important, but probably shouldn't be tossup topics even at a non-vanity/subject tournament--stuff like deep inelastic scattering, vacuum polarization, and Feit-Thompson theorem (and possibly my own tossup on the Kepler mission; possibly also some stuff like dust and metallicity? I don't know, those were fine for me but I wonder how they played for other people). I'm not saying these absolutely should have been watered all the way down to tossups on "scattering" and "vacuum" and "math," but presumably there's some middle ground of related answers that a decent number of teams can muster that would still allow lead-ins rewarding knowledge of these topics.

Also, it felt like some science bonuses really had two hard parts rather than a medium and a hard part--e.g. the edited version of my stats bonus, the KAM theorem bonus, the plesiosaurs bonus, possibly the Hidden Markov/Bayes/Viterbi bonus.

Eric, I think Selene and I can offer more science feedback after the set is posted.

-Seth

[Auks Ran Ova]

packets, until whatever's happening with quizbowlpackets works itself out: here

EDIT: if this link doesn't work for you, try just going to angelfire.com/mech/strifeheart and clicking on "few"; if you don't want to do that, email me for the packets instead.

[Notably Not Pierre]

I was also on the team with Ben and Marshall that went the whole tournament hearing 4/4 of the Math and Computer Science. Having seen the rest of the packets, it doesn't look like the subdistribution was as skewed as it seemed to us; we just happened to miss 3/4 of the Math and Computer Science in the 5 packets we didn't play on.

Obviously as someone whose primary real-life interests lie in Math and CS it would be to my benefit if those areas were better represented in quizbowl, and I certainly overvalue their importance relative to the quizbowl and academic communities at large. That said, I think we should find some way to get more Math and CS (and Stats!) into tournaments, although I'm not sure what categories it would be at the expense of.

Some points about specific Math/CS questions (many of which I didn't actually get to play on):

Emil Post - People have already mentioned that this seems somewhat ill-advised. I'm not sure if it's actually too hard, but he isn't someone I know a whole lot about despite being particularly interested in recursion theory and formal languages. As far as I could tell, it wasn't really possible to power the question solely with actual knowledge of the PCP (rather than knowledge about his "influential" address to the AMS - according to Wikipedia). I think this is just an example of the pitfalls in tossing up people who are only famous for one thing, though.

Feit-Thompson Theorem - I'm proud to say that this question was powered in my room thanks to the fact that between the two teams we had 7 quarters of experience in Chicago's Honors Algebra sequence. I'm also pretty sure that only 1 out of the 4 math majors in the room actually remembered anything about it. As Trevor Davis' experience seems to suggest, it's entirely possible to learn tons of abstract algebra, and learn nothing about this theorem. A cursory inspection of the index of Hungerford's graduate Algebra seems to suggest it's a "hard result." (It's notable absent.) Dummit and Foote, a very comprehensive undergraduate algebra text gives the theorem without proof and remarks in a footnote that the proof is really hard. Problems in that section also prohibit the use of the Feit-Thompson theorem.

Obviously this question is an outlier in terms of difficulty, but I think there is a more general lesson here: we should stay away from requiring deep knowledge of theorems and problems whose proofs are too long/complicated/technical to be presented in textbooks or classrooms. There are plenty of hard things it's eminently possible to have real, deep knowledge about for us to just ask about those. Obviously important and challenging results, like Fermat's Last Theorem, 4 Color Theorem, Feit-Thomspon theorem, the Poincare Conjecture, Riemann Hypothesis etc are all important and should come up, but they should come up in ways that don't pretend to reward the deep knowledge no one has.

Normal subgroups/Lagrange's Theorem/Cosets - I think this is an example of a question that does everything right. (This is based solely on the answer lines; it was never read in a round I played). These are all things that any abstract algebra class would treat in detail, including a proof of the theorem in question. Perhaps this was a bit too easy or whatever, but this is definitely the kind of thing we should be asking about.

Lagrange Multipliers - Was this even really a math question? I didn't actually play on this question but from what I heard, it was basically about a bunch of applications to economics. That's a good idea since I think this is exactly the kind of thing we should be testing...with economics questions. From second-hand information it sounds like anyone who had taken an Intermediate Micro class would crush someone with an arbitrarily high amount of math knowledge on this question. I like economics too, and I'm glad this a question was in the tournament, it's just too bad that the only way we can get a real economics question (rather than one which rewards knowledge of scandals and titles related to an economist) is by putting it in the Math distribution.

Central Limit Theorem, and Null Hypothesis/Something I'd never heard of/ Neyman-Pearson Lemma - We didn't hear the CLT tossup, but it's a good idea for a tossup answer line: it's both accessible and something about which people have deep knowledge. Also Neyman-Pearson is a good idea for a hard part, although I've never heard of the 2nd thing and can't vouch for its authenticity. I would just like to point out that Statistics is not Math. Quizbowl needs more statistics questions since it's obviously very important and something people know about, but those questions probably shouldn't be coming at the expense of Math/CS. I'm not totally sure how Stats currently fits into the distribution, but as far as I can tell, it's just "Math." Can we just call Stats another one of the privileged minor sciences and treat it like we do CS? By that I mean making it explicitly acceptable as the mandatory Math question, but also making sure people can write about Math and Stat in the same packet without considering that a repeat of a minor science.

KAM theorem / Torus / Whatever - Is the KAM theorem math? I have no idea what that actually is and it's never come up in any math class I've taken. Based on a very quick inspection of the Wikipedia article, it seems interesting and worth asking about but it also seems like something only physicists could know. Am I just wrong here? If not, it seems strange to count this as a Math question, although both my ignorance and strong personal bias might account for my position on this one.

All that said, I thought it was overall a well-written tournament and was glad to get the chance to play it.

[The Toad to Wigan Pier]

Emil Post - People have already mentioned that this seems somewhat ill-advised. I'm not sure if it's actually too hard, but he isn't someone I know a whole lot about despite being particularly interested in recursion theory and formal languages. As far as I could tell, it wasn't really possible to power the question solely with actual knowledge of the PCP (rather than knowledge about his "influential" address to the AMS - according to Wikipedia). I think this is just an example of the pitfalls in tossing up people who are only famous for one thing, though.

This tossup was very powerable, but of course you need to know more than his correspondence problem. His lattice and work on Turing degrees are essential to computability theory.

[ThisIsMyUsername]

I really enjoyed playing this tournament, even though it was harder than its advertised difficulty (and clearly not just because of submissions, because the hardest packets were the editors'). Quality of writing will always trump difficulty for me, and the quality for this tournament was high.

The Editors 3 packet contained three questions worth commenting on:

- The Jim Casy tossup: I was glad to see this come up, a great idea for a character tossup, which I am surprised has not been done before.
- The Battle of Mohacs tossup: What the hell is Louis II doing in the fourth line of an eight-line tossup on Mohacs, still in power?
- Ein Heldenleben: This tossup claimed that the recurring "hero" theme is horns and celli playing E triads. It is in fact E-flat triads. This prevented me from buzzing and consequently we were beaten to this question. There is a tendency for music-question writers to be careless and do something like write "major" when they mean "minor" or to just outright confuse keys (there were tossups at ACF Fall this year and ACF Regionals last year that did this). The key that something is in is always a significant fact, rather than an unimportant detail, especially so in this case, where the piece is carrying on the tradition of "heroic" E-flat pioneered by Beethoven's Eroica.

And though I have nothing against Canadian history, I will agree that it felt over-represented. (Were there more questions about Canada than Great Britain, or is that just an impression I got?)

[Masked Canadian History Bandit]

And though I have nothing against Canadian history, I will agree that it felt over-represented. (Were there more questions about Canada than Great Britain, or is that just an impression I got?)

Definitely an impression.

Canadian history in this set: 3/1

Sir Wilfred Laurier (Maryland)
On-to-Ottawa Trek (Editors 3)
New Democratic Party of Canada (NDP) (Editors 7)

Trudeau/Just Watch Me/Turner (Editors 1)

Overall Canadiana: 4/2

Above plus,
Moshe Safdie (Editors 4)

Group of Seven/Pine Tree/Fauvism (Editors 3)

Meanwhile, British History in the first 7/16 rounds played at the Penn site: 5/3

Bloody Assizes (Bentley/Davis)
Repealing the Corn Laws (Vinokurov + Michigan)
Henry VIII's Dissolution of the Monasteries (Snorlaw+Sethlene)
John Wilkes (Editors 1)
Francis Drake (WUSTL)

Alex III/Robert the Bruce/Auld Alliance (Maryland)
Clement Attlee/Windscale Disaster/Polaris Missiles (Illinois/Yale)
Mountbatten/Prince Phillip/Harold Wilson (Snorlax+Sethlene)

It's probably just that there was more Canadiana than normal.

[grapesmoker]

Feit-Thompson Theorem

I took a year of undergrad abstract algebra and Galois theory and never encountered this. Maybe I was taking the wrong classes, but this strikes me as a very difficult question.

Obviously this question is an outlier in terms of difficulty, but I think there is a more general lesson here: we should stay away from requiring deep knowledge of theorems and problems whose proofs are too long/complicated/technical to be presented in textbooks or classrooms. There are plenty of hard things it's eminently possible to have real, deep knowledge about for us to just ask about those. Obviously important and challenging results, like Fermat's Last Theorem, 4 Color Theorem, Feit-Thomspon theorem, the Poincare Conjecture, Riemann Hypothesis etc are all important and should come up, but they should come up in ways that don't pretend to reward the deep knowledge no one has.

I agree with this though it seems like everything in your list other than the FT theorem is actually much more accessible in the sense that you could write a deep question on those things that rewards knowledge that people do have.

Normal subgroups/Lagrange's Theorem/Cosets - I think this is an example of a question that does everything right.

Agreed. Excellent bonus.

Lagrange Multipliers - Was this even really a math question? I didn't actually play on this question but from what I heard, it was basically about a bunch of applications to economics. That's a good idea since I think this is exactly the kind of thing we should be testing...with economics questions. From second-hand information it sounds like anyone who had taken an Intermediate Micro class would someone with an arbitrarily high amount of math knowledge on this question. I like economics too, and I'm glad this a question was in the tournament, it's just too bad that the only way we can get a real economics question (rather than one which rewards knowledge of scandals and titles related to an economist) is by putting it in the Math distribution.

I wrote this question. This was not a "pure" math question but neither was it only about economics. The initial clues were from an econ application while the later clues were from physics and math. I perused several textbooks and it actually turned out to be quite hard to find multiple clues on Lagrange multipliers, so I chose instead to focus on their applications, since they are ubiquitous. It was not intended to fill the econ distribution (and I'm not convinced that scandals and titles are a more valid econ answer) and I think the set of people who have taken intermediate econ is not that large.

KAM theorem / Torus / Whatever - Is the KAM theorem math? I have no idea what that actually is and it's never come up in any math class I've taken. Based on a very quick inspection of the Wikipedia article, it seems interesting and worth asking about but it also seems like something only physicists could know. Am I just wrong here? If not, it seems strange to count this as a Math question, although both my ignorance and strong personal bias might account for my position on this one.

Was this listed in math? To me this is a physics question (c.f. ACF Nationals 2011). The KAM theorem is basically the most important result in the theory of dynamical systems; it's a statement about which phase space orbits survive perturbations (those would be the quasiperiodic orbits, the answer to the last bonus part). However, even though I would have classed this under physics, mathematical physics is certainly a branch of mathematics too, and that's where this theorem comes from. You are right that it's not taught in undergrad math classes, but it was covered (briefly) in my undergrad dynamics class and more extensively in my graduate mechanics class.

[MLafer]

Well, I've now gone through the packets including tiebreakers and unread packets and my feeling was correct: there was only one tossup in the entire set on history that occurred before 500 CE (Aurelian). This seems like a gross imbalance, chronologically, even considering the decent amount of ancient history bonuses. I think subdistributions in categories should attempt to have balance between the number of tossups and bonuses.

The only other real gripes I had have already been mentioned - some really hard biology and a crazy amount of astronomy. I'll especially praise the physics, which was gotten consistently early by Seth, as it should be, and on very important concepts; as well as the American history, an area where the answer choices are usually boring but here were consistently creative.

[grapesmoker]

I guess Eric has already commented on this in connection with the Feit-Thompson question, but: I thought there were some tossups on topics that are real and important, but probably shouldn't be tossup topics even at a non-vanity/subject tournament--stuff like deep inelastic scattering, vacuum polarization, and Feit-Thompson theorem (and possibly my own tossup on the Kepler mission; possibly also some stuff like dust and metallicity? I don't know, those were fine for me but I wonder how they played for other people). I'm not saying these absolutely should have been watered all the way down to tossups on "scattering" and "vacuum" and "math," but presumably there's some middle ground of related answers that a decent number of teams can muster that would still allow lead-ins rewarding knowledge of these topics.

I thought questions like "dust" and "metallicity" were actually great; the latter may not be quite as accessible as the former, but it's still a very important topic (not that I need to tell you that!) and I'd much rather see that than questions on Herbig-Haro objects or whatever. Vacuum polarization is the one physics question that I thought was just not successful at all; I probably don't know all that much about the phenomenon, and I suppose I should have figured it out from the Lamb shift clue, but it strikes me as basically impossible for almost anyone else playing this tournament.

[grapesmoker]

At what point was Laurier powered in your room?

Jerry recognized the first clue, and then spent the next few clues trying to remember where he knew the first clue from before buzzing in and receiving 15 points.

Slight correction: I actually recognized (and subsequently buzzed on) the clue that indicated that his predecessor served 69 days. Further investigation reveals that the short-serving PM was Charles Tupper; since I couldn't come up with any other PMs around that time, I went with the obvious answer of the famous guy.

Fun fact: I know all this purely from listening to the Three Dead Trolls' song about five short-serving PMs.

[grapesmoker]

I wish that Kepler tossup hadn't begun with "These totally Scandinavian people are in charge of it." Maybe this didn't actually help anyone (I only processed this after the fact) but it felt gimmicky.

I'm not really sure what this means. Could you clarify?

I wrote the Kepler tossup and after seeing it play in one room and talking with some people in other rooms at the Minnesota site I think it probably wasn't a good idea for a tossup, but for reasons that don't seem connected with Jerry's note here. Did anyone get this one before the giveaway?

Obviously I'm in a somewhat privileged position with regards to playing this tossup. Here's as much of the question as I heard before getting this tossup:

This project’s Asteroseismic Investigation component is led by Ron Gilliland and Jørgen Christensen-Dalsgaard, among others. It utilizes a very precise photometer with a large field of view that takes data every 30 minutes from almost 150,000 stars

I'm not sure the first clue is useful. I've been to talks on and read articles about Kepler (hence being able to buzz from the experiment description), and those names ring no bells for me. That second name really sounds totally Scandinavian though, which didn't strike me until after the fact. Anyway, it's not a big deal, it just doesn't seem to me that "these people are the project leaders" is a very useful piece of information to anyone.

[Cheynem]

The Mohacs one was my mistake. I attempted to arrange the clues in the order that I was most familiar with them and that left Louis II, a clue that for whatever reason I was not familiar with, in the middle. Sorry about that. I hope you liked the Jim Casy tossup--I've never read the book.

In regards to ancient history, as Trevor Davis will attest, I dislike writing ancient history questions and I did attempt to make more of an effort on it this year by writing bonuses but I should have spread it out to tossups as well. I think it didn't help that each editor writing history was not interested in writing ancient history tossups either. That's also definitely something I want to work on in the unlikely event I work on this tournament again.

There was a lot of Canada, but there was way more British stuff. I enjoy writing on (modern) British history a lot.

[Auroni]

One thing that I just remembered.. should the pyridine tossup have "bipy" that early? Not that people who know what that is don't deserve a power, because they totally do, but it's right there in the name.

[Sima Guang Hater]

I'll cop to the lack of ancient history. I actually do like writing ancient but I should have cordoned off more space for it.

One thing that I just remembered.. should the pyridine tossup have "bipy" that early? Not that people who know what that is don't deserve a power, because they totally do, but it's right there in the name.

So I thought about that, but I though it was real knowledge rewarding enough to warrant being a leadin.

[Auks Ran Ova]

- Ein Heldenleben: This tossup claimed that the recurring "hero" theme is horns and celli playing E triads. It is in fact E-flat triads. This prevented me from buzzing and consequently we were beaten to this question. There is a tendency for music-question writers to be careless and do something like write "major" when they mean "minor" or to just outright confuse keys (there were tossups at ACF Fall this year and ACF Regionals last year that did this). The key that something is in is always a significant fact, rather than an unimportant detail, especially so in this case, where the piece is carrying on the tradition of "heroic" E-flat pioneered by Beethoven's Eroica.

Yeah, I'm sorry about this. This was a typo; all of my sources said "E-flat" and that's what I fully intended to include in the question. Carelessness or laziness were certainly not my intent, as I certainly understand the importance of exactness here; I think the word "flat" just got accidentally deleted while I was writing and I never noticed.

[women, fire and dangerous things]

A slight erratum from the Bentley et al. packet: Aira's El congreso de literatura was translated into English recently. Also, it's awesome.

[DumbJaques]

So, I definitely enjoyed playing this tournament, and in particular I thought that once you removed the hardest 1-4 tossups from a given packet, the answer selection did an excellent job of being reasonable for a harder, open/Nats level event. Though I haven't looked at the packets yet, I also recall some extraordinarily laudable things like tossups on easier answers using hard but not abjectly obscure clues (Geertz springs to mind). While this approach did fail sometimes and create problematic difficulty cliffs (Bertrand Russell) or result in 13-line tossups on Kenzaburo Oe that only get easier than The Pinch-Runner Memorandum at line 12, I hope everyone recognizes that it's one of the critical cogs in the wheel that gets us out of the difficulty pit we've found ourselves in.

On that note, there were definitely a few not-so-great difficulty issues with the set. I'm kind of surprised nobody has mentioned this yet, but take a look at the stats from the Penn site. The team of Jerry, Ted, Aaron, and Bruce put up 71 powers over 13 rounds. Maryland's A team (me, SJ, Arun, Dan Puma) put up 24 - just about 1/3 of their total. The difference in ppb between these two squads? About 1.2. I guess that's not ridiculous, but I do think's it's reflective of this tournament's major problem with hard parts - specifically, some significant percentage of them were just impossible. No matter how you slice it that's a big issue with hitting that appropriate knowledge gradient, but let's be honest - impossible third parts are usually the thing you worry least about when writing your set. The reasons are fairly obvious (after all, they're SUPPOSED to be the hardest parts of the set), but in general we tend to reason that the impact of a hard part that maybe nobody gets is really low (far lower than the impact of a medium part that's really hard, or a tossup going dead, etc.).

However, I'd suggest that the problems of impossible third parts go further than that. First of all, when they're as widespread as they were at MO, it causes significant muddy battlefield problems, which is a problem for any tournament. You also have to consider that when you get into the mindset of just letting it fly with third parts, that means that when you make the occasional mistake of thinking something is easier than it really is, that something is probably now going to be a MIDDLE part - and that can cause serious issues with bonus variability. You're also inevitably going to have plenty of reasonable hard parts that top teams are going to convert, so it becomes less about whether you get something in your better areas and more about if you happen to get a bonus where the third part was feasibly answerable (this problem is amplified in sets that, like MO, have a committee of editors).

Another thing I think we might be overlooking with impossible hard parts is that people will often try to compensate by making sure the middle parts are gettable - in general that's good, but at tournaments like this you DO want middle parts that differentiate teams when they aren't as good at various categories. So you can very easily end up in a situation where a team like Maryland would be getting the same conversion on a lot of music bonuses as, say, UVA (because the middle parts would be relaxed, and the hard parts may or may not always even be answerable). That's a point where you're failing to differentiate knowledge levels, and it's a part of how you end up with wonky results.

Bottom line, as we continue our push to scale back difficulty this year, we need to make sure that we aren't using hard parts as some kind of outlet for our raging quizbowl id, because the impact is definitely noticeable.*

*To be clear, this wasn't some MO-specific problem, but something I've noticed at a lot of events. It also didn't keep MO from being fun to play, though I think we could have done with some shorter questions/less focus on really obscure stuff.

[theMoMA]

I partially agree with Chris; as I was reading through the packets, I thought that there were a significant number of bonuses that didn't have answerable third parts. On the other hand, I thought that Seth's team had a noticeable advantage on the bonuses, so I'm not sure that the muddy battlefield problem that Chris identifies really played out, at least at our site. I will also note that powers are probably not the best backdrop against which to measure team competence; oftentimes, a high number of powers has to do as much with playing strategy as knowledge base.

From my perspective, this was one of the better harder-difficulty events that I've had the pleasure to read. I thought that Rob and Mike did their best work to date on this set, and I hope that my contributions (if I'm not mistaken, about 40 questions total, the vast majority tossups) were well-received. Eric's science was, in my amateur opinion, extremely high-caliber as usual, though I thought that the biology shaded harder that was probably warranted. As a reader, I will observe that Eric probably needs someone to look over his questions to fix grammar mistakes and rambling sentences (there weren't too many, but the only place I saw them regularly was in the science), and that everyone could use to write more eight-liners than thirteen-liners, but these were pretty negligible shortcomings compared to the general quality of this set.

[Auks Ran Ova]

problematic difficulty cliffs (Bertrand Russell)

Hmm, you're right, there probably should've been something a little bit deeper to lead into the "On Denoting" section of the question--noted for the future.

or result in 13-line tossups on Kenzaburo Oe that only get easier than The Pinch-Runner Memorandum at line 12,

This, on the other hand, was deliberate--sort of a morbid joke/final extension into absurdity of the old model of difficulty-racing in Asian lit questions combined with a genuine curiosity and desire to test knowledge of that stuff. I'd argue that the pyramid isn't as steep as you make it sound, but ultimately it certainly could've been normalized a bit just by cutting off the earlier, crazier clues. (Also, you'll surely be "delighted" to know that Ike Jose won a buzzer race on "the Flaming Green Trees trilogy", so there's that.)

[Magister Ludi]

I partially agree with Chris; as I was reading through the packets, I thought that there were a significant number of bonuses that didn't have answerable third parts. On the other hand, I thought that Seth's team had a noticeable advantage on the bonuses, so I'm not sure that the muddy battlefield problem that Chris identifies really played out, at least at our site. I will also note that powers are probably not the best backdrop against which to measure team competence; oftentimes, a high number of powers has to do as much with playing strategy as knowledge base.

I would humbly submit that a team of Jerry, Aaron, Bruce, and myself earned our powers more from our knowledge base than a superior "playing strategy." And I was often frustrated with the third parts of bonuses at this tournament because teams with deeper expertise were often denied a chance to pull away with thirties. The more I think about it, the more I agree with Chris's sentiments because there seems to be a trend at recent hard tournaments where bonuses become somewhat perfunctory as editors focus on making the second part very gettable while ignoring difficulty of the third part. I think this mentality has insidious side effects in which the bonuses my team actually 30ed at this tournament had little to do with how much we actually knew about a topic and was based much more on getting lucky. For example, V. S. Naipaul is one of my favorite writers and I've read all his major works, but I 20ed that bonus because the hard part was on his ninth or tenth most important novel, which is exactly as many points as I got on the Gao bonus (about whom I know practically nothing). Moreover, the bonuses my team did thirty seemed to be based on the whims of the editor rather than our actual depth of knowledge in the topic, so we 30ed the Art Nouveau bonus because we got lucky while we 20ed the Bolano bonus because the hard part is about The Distant Star even though I actually know a lot more about Bolano than Art Nouveau. I think Chris very acutely points out a problem with hard tournaments that are edited towards the "universal twenty."

I personally think it's a more productive editing strategy to pick hard parts based on what is important and likely to be known within a topic. For example, if I wrote that Naipaul bonus I would think to myself: What would make an important third part that someone who is knowledgeable about Naipaul is likely to answer? And I would probably pick the first story from In a Free State or a character from in A Bend in the River or Biswas rather than a minor novel. I think a problem with several bonuses in this set was asking for depth in the wrong places where players were unlikely to have knowledge. Additional examples that come to mind off the top of my head are the Akutagawa bonus with the hard part on some third tier story or the Congreve bonus with the third part on his first play. Rather than constructing a bonus to allow the editor to expand the canon by sneaking in a hard part about Ba Jin or a Caroline Gordon novel, I wish editors adopted a more empathic approach geared towards asking third parts that are meant to be answered.

I don't mean to detract from a solid tournament that I would rate as an outstanding 8.5/10, but I think for editors of this caliber I feel comfortable pointing out small scale issues for them to address or ignore as they see fit.

[Maxwell Sniffingwell]

This, on the other hand, was deliberate--sort of a morbid joke/final extension into absurdity of the old model of difficulty-racing in Asian lit questions combined with a genuine curiosity and desire to test knowledge of that stuff. I'd argue that the pyramid isn't as steep as you make it sound, but ultimately it certainly could've been normalized a bit just by cutting off the earlier, crazier clues. (Also, you'll surely be "delighted" to know that Ike Jose won a buzzer race on "the Flaming Green Trees trilogy", so there's that.)

Didn't this already happen when Matt W. wrote a 12-line Murakami tossup for CO Lit?

[Mike Bentley]

I personally think it's a more productive editing strategy to pick hard parts based on what is important and likely to be known within a topic. For example, if I wrote that Naipaul bonus I would think to myself: What would make an important third part that someone who is knowledgeable about Naipaul is likely to answer? And I would probably pick the first story from In a Free State or a character from in A Bend in the River or Biswas rather than a minor novel. I think a problem with several bonuses in this set was asking for depth in the wrong places where players were unlikely to have knowledge. Additional examples that come to mind off the top of my head are the Akutagawa bonus with the hard part on some third tier story or the Congreve bonus with the third part on his first play. Rather than constructing a bonus to allow the editor to expand the canon by sneaking in a hard part about Ba Jin or a Caroline Gordon novel, I wish editors adopted a more empathic approach geared towards asking third parts that are meant to be answered.

Maybe I'm in the minority here, but I typically find character questions very difficult to answer for works I've ready. Obviously asking about a protagonist or something is a very reasonable thing to do, but is it really common for people to remember second-tier characters without explicitly memorizing them? At that point we're probably running into the same problem (if we want to call it a problem) as people who sit down and memorize second and third tier works by authors.

For instance, having read both The Martyr and Hell Screen, I'd find it easier to answer a question on The Martyr than something like the daughter's name from Hell Screen (assuming she had one, I don't recall at this point).

[The King's Flight to the Scots]

I personally think it's a more productive editing strategy to pick hard parts based on what is important and likely to be known within a topic. For example, if I wrote that Naipaul bonus I would think to myself: What would make an important third part that someone who is knowledgeable about Naipaul is likely to answer? And I would probably pick the first story from In a Free State or a character from in A Bend in the River or Biswas rather than a minor novel. I think a problem with several bonuses in this set was asking for depth in the wrong places where players were unlikely to have knowledge. Additional examples that come to mind off the top of my head are the Akutagawa bonus with the hard part on some third tier story or the Congreve bonus with the third part on his first play. Rather than constructing a bonus to allow the editor to expand the canon by sneaking in a hard part about Ba Jin or a Caroline Gordon novel, I wish editors adopted a more empathic approach geared towards asking third parts that are meant to be answered.

Maybe I'm in the minority here, but I typically find character questions very difficult to answer for works I've ready. Obviously asking about a protagonist or something is a very reasonable thing to do, but is it really common for people to remember second-tier characters without explicitly memorizing them? At that point we're probably running into the same problem (if we want to call it a problem) as people who sit down and memorize second and third tier works by authors.

For instance, having read both The Martyr and Hell Screen, I'd find it easier to answer a question on The Martyr than something like the daughter's name from Hell Screen (assuming she had one, I don't recall at this point).

Yeah, I think this is a legitimate issue. Sometimes characters have very memorable names, sometimes they're major enough that they're worth asking about anyway, but the character bonus part is overdone sometimes. I like it when people ask about a plot detail as a third part--at Regionals, maybe your hard part could ask for what profession Yoshihide has in Hell Screen and have "ANSWER: A _Painter_ [or _artist_; accept equivalents]." Some people responded negatively to the _mendacity_ bonus part I wrote for Terrapin last year, but I think that's another thing that can be done.

[grapesmoker]

I think a bonus like the Salinger bonus (Laughing Man/ Salinger/ Raise High the Roofbeam, Carpenters) is a good example of something like this done right. The Laughing Man is a very memorable character from Nine Stories (also the title of the story), so it's a good thing to ask about because people who have read a lot of Salinger will know that.

Sure, the hard part were too hard. They should have probably been easier. But it was mostly pretty rough across the board and with the exception of a few imbalanced bonuses things were relatively even. It was just That Kind of Tournament.

[women, fire and dangerous things]

I'm glad I'm not the only one who sucks at remembering character names from stuff they've read. Also, while I agree with Ted's overall point about the hard parts of bonuses, I don't think the Bolano example is a good parallel to the Naipaul example. Bolano is one of my favourite writers and I've read about a dozen of his books, and consequently I would have 30ed that bonus, just like Ted would ideally 30 most Naipaul bonuses. Fluky stuff like 30ing a bonus on Art Nouveau when you're not particularly an expert on Art Nouveau is always going to happen; you shouldn't expect to also 30 a Bolano bonus just because you know more about Bolano than Art Nouveau.

[Magister Ludi]

I personally think it's a more productive editing strategy to pick hard parts based on what is important and likely to be known within a topic. For example, if I wrote that Naipaul bonus I would think to myself: What would make an important third part that someone who is knowledgeable about Naipaul is likely to answer? And I would probably pick the first story from In a Free State or a character from in A Bend in the River or Biswas rather than a minor novel. I think a problem with several bonuses in this set was asking for depth in the wrong places where players were unlikely to have knowledge. Additional examples that come to mind off the top of my head are the Akutagawa bonus with the hard part on some third tier story or the Congreve bonus with the third part on his first play. Rather than constructing a bonus to allow the editor to expand the canon by sneaking in a hard part about Ba Jin or a Caroline Gordon novel, I wish editors adopted a more empathic approach geared towards asking third parts that are meant to be answered.

Maybe I'm in the minority here, but I typically find character questions very difficult to answer for works I've ready. Obviously asking about a protagonist or something is a very reasonable thing to do, but is it really common for people to remember second-tier characters without explicitly memorizing them? At that point we're probably running into the same problem (if we want to call it a problem) as people who sit down and memorize second and third tier works by authors.

For instance, having read both The Martyr and Hell Screen, I'd find it easier to answer a question on The Martyr than something like the daughter's name from Hell Screen (assuming she had one, I don't recall at this point).

Yeah, I think this is a legitimate issue. Sometimes characters have very memorable names, sometimes they're major enough that they're worth asking about anyway, but the character bonus part is overdone sometimes. I like it when people ask about a plot detail as a third part--at Regionals, maybe your hard part could ask for what profession Yoshihide has in Hell Screen and have "ANSWER: A _Painter_ [or _artist_; accept equivalents]." Some people responded negatively to the _mendacity_ bonus part I wrote for Terrapin last year, but I think that's another thing that can be done.

Maybe I didn't explain my position clearly. I'm not trying to say that editors should force bonus parts on characters for every author, but rather that editors should approach each topic differently and pick appropriate elements from their oeuvre as bonus parts. I thought the "mendacity" bonus part at TIT last year was exemplary of exactly the kind of third parts we should be asking more often. For some authors characters are important but for other authors like Akutagawa a bonus should focus on other elements from their work. In the past I wrote third parts on the Milton quote "they also serve who only stand and wait" and the Hemingway phrase "a separate peace." I just used characters as an example of one potential type of hard part. The key is to pick things that are memorable and important.

[Auks Ran Ova]

Maybe I didn't explain my position clearly. I'm not trying to say that editors should force bonus parts on characters for every author, but rather that editors should approach each topic differently and pick appropriate elements from their oeuvre as bonus parts. I thought the "mendacity" bonus part at TIT last year was exemplary of exactly the kind of third parts we should be asking more often. For some authors characters are important but for other authors like Akutagawa a bonus should focus on other elements from their work. In the past I wrote third parts on the Milton quote "they also serve who only stand and wait" and the Hemingway phrase "a separate peace." I just used characters as an example of one potential type of hard part. The key is to pick things that are memorable and important.

What did you think about the Henderson the Rain King bonus?

[No Rules Westbrook]

Yeah, it's important to be creative in choosing third parts, and to have some Hartian empathy for the player as to what's honestly gettable. That much we can agree on.

But, I really wish you'd make these discussions more relevant to robots like me by talking about the plight of the surface generalist! At least as often as people go venturing into their scholarly knowledge of Milton's canon to dredge up answers to third parts of bonuses, they go venturing into that murky Well of Knowledge which contains a whole matrix of words and phrases that the player has learned to mindlessly spit at the moderator because they've encountered said words and phrases previously in their quizbowl travels.

Just sayin. I mean, obviously the parts need to be on legitimately important things, and not the Edwidge Danticat of the season. I just think these discussions are often a bit stilted when they're driven exclusively by people citing their particular areas of extensive primary knowledge.

[grapesmoker]

What did you think about the Henderson the Rain King bonus?

I thought this was a great bonus. It went to Yale, but I would have 30d it from having read it; the "I want!" is pretty key, and I was glad to see it asked.

[grapesmoker]

Yeah, it's important to be creative in choosing third parts, and to have some Hartian empathy for the player as to what's honestly gettable. That much we can agree on.

But, I really wish you'd make these discussions more relevant to robots like me by talking about the plight of the surface generalist! At least as often as people go venturing into their scholarly knowledge of Milton's canon to dredge up answers to third parts of bonuses, they go venturing into that murky Well of Knowledge which contains a whole matrix of words and phrases that the player has learned to mindlessly spit at the moderator because they've encountered said words and phrases previously in their quizbowl travels.

I think that Milton example really works here, though: it's one of the most memorable lines from one of his most memorable (perhaps the most memorable) sonnets, so it's not like it's some crazy imposition on the player to be asked about that. The problem of course is that in a lot of situations you have to be careful with what you ask about so that you don't venture into obscurata. It's a tough line to walk and I'm personally inclined toward lenience when judging questions with regard to this.

[Sima Guang Hater]

Eric, I think Selene and I can offer more science feedback after the set is posted.

That offer still good

[setht]

Eric, I think Selene and I can offer more science feedback after the set is posted.

That offer still good

Sure, here's some more detailed commentary on physics/astro/math/earth science, going packet by packet. Actually, I'll give some general comments, then start in on the nauseating detail.

First off, as Jerry said, the physical science in the set was good as a whole. I think the main suggestions I would offer for making future physical science sets even better would be: 1) rein in some TU answer selection (I think this was more of a problem in bio and chem than in my areas, but I'm not sure; in any case, there were some tossups in my areas that probably fell pretty flat in most/all matches); 2) work on evening out the bonus difficulty--getting all the easy parts roughly in line with each other (e.g., is bremsstrahlung without German/braking radiation as easy as convection off of "heat transport by fluid motion"?), similarly for medium parts (do you want medium parts to look like Rayleigh number or Hohmann transfer orbit [or Mid-Atlantic Ridge]?) and hard parts (goinometer or Boussinesq approximation [or drift velocity]?); 3) work on clarity of language and root out Wikipedia-isms with fire and sword--I caught a lot of these while writing up comments. In step 2, my vote would be to aim in all three parts for the lower end of the difficulty range exhibited in the MO science: I think there are just too many teams without a science player with deep knowledge of subject X for any science field X (or no science player at all), even at an open event like MO, to justify the alternative.



vacuum polarization: too hard for a TU

complete graph/four/Turan's theorem: Turan seems like it may be an overly-hard hard part

Wigner-Eckhart theorem/angular momentum/Levi-Civita symbol: good; maybe a bit less awkward if it was reordered to ask for ang. mom. first, then Wigner-Eckhart


tokamak: not sure what the second-to-last sentence really means

dust: porosity and fluffiness sometimes refer to different things (effective density in a 3D grain structure with lots of voids vs. a fractal structure with dimension < 3); I wouldn't bother with stuff like "aforementioned zodiacal cloud"--if someone isn't buzzing on zodiacal light they're probably not buzzing on zodiacal cloud, and quizbowl could use less of this stuff (anaphoric referents, Wikipedia tells me).

renormalization/infrared divergence/bremsstrahlung: I don't think this has a true easy part and infrared divergence feels possibly hard for a hard part; lead-in mashes together two unrelated clauses


Hamiltonian: maybe switch second and third sentences; maybe break up first sentence; but really, this is a solid physics tossup

Liouville: writing out formulas that aren't super-famous without defining terms usually feels like a waste of time--in the second sentence I would just give some quick description of the irrational algebraic number result and then cut to Roth's theorem (or just cut this sentence: this questions feels like it has two sentence's worth of lead-in material). In the third sentence I would either define terms in the description of Sturm-Liouville methods or try to explain it a bit in words.

SDSS/CCD/great wall: all great stuff, but I'm not sure this has a true easy part

Boussinesq/convection/Rayleigh number: looks good


superstring theory: I generally dislike these tossups; I often figure out pretty quickly that we're in non-Standard Model land ("dilatons") and then can't decide if I should buzz with string theory, superstring theory, or something even more specific. Perhaps there are enough players out there who can actually distinguish amongst all the options while a tossup is being read; I can't and wouldn't mind seeing these questions relegated to the occasional bonus part at most.

Feit-Thompson: too hard for a TU

GR/metric/parallel transport: all good, solid stuff, but I wonder if this is too hard for many teams to get 20 on; on the other hand, I think the hard part is not impossible, which is always nice


density matrix: feels hard for a TU

QHO/coherent/annihilation: possibly hard to 20/30


mica: it's always mica; some pronoun confusion in second sentence ("it" referring to phlogopite rather than mica)

grad: solid

waveguide/tir/Imbert-Fedorov: I-F seems overly hard--I think Goos-Hanchen is kind of pushing it but may get a pass because it's in Jackson; I-F doesn't even have that going for it.

cosets/Lagrange/normal: this looks really good to me--all stuff people would learn in an undergrad class, and a nice non-algebra secondary clue on Lagrange to make it a true easy part


H atom: the first sentence does nothing for me, but aside from that this looks good (minor nitpicks on late clues: it should really be 13.6 eV, not just 13.6; I'm not sure "Coulomb potential" really helps, and I don't think "it is able to reproduce the Rydberg formula" makes sense--maybe something like "line emission from this system follows the Rydberg formula for wavelengths" would work)

galaxies: again, the first sentence doesn't really ring any bells for me, but I'm woefully ignorant on galaxy modeling. After that, the dark halo bit is more useful to me than the following sentence on ram-pressure stripping; I guess that sentence also says they have gas, but "they have dark matter" still seems more useful to me.

B/flavor/neutrino: seems solid to me, except that the clue about hypothesizing neutrinos seems a bit off to me--as far as I know Pauli proposed the existence of neutrinos to explain how various conservation laws wouldn't be violated by standard beta decay, not annihilation events

KAM/torus/quasiperiodic: feels a bit hard--I think the attempt to give away torus doesn't quite go far enough, and KAM and quasiperiodic both feel hard for medium parts.


deep inelastic scattering: probably hard for a TU

central limit theorem: solid

escape velocity/Hohmann/hypergolic: feels like one easy and two hard parts (or maybe one hard part and one impossible part)

two-stream/radiative transfer/Eddington: all good stuff, but probably too hard given how few people learn radiative transfer


precession: define terms

open: I wouldn't say the Hausdorff condition "requires" two open sets--I'd say it "involves" them or something

basalt/Mid-Atlantic Ridge/guyots: this feels noticeably easier than pretty much all other science bonuses; I'm not even sure which of the first two parts is the medium part, and guyots feels easy for a hard part, but may be just fine.

double-slit/aberration/speckle: feels a bit easy to 20 and a bit hard to 30. It's probably fine.


star formation: I don't know any of the stuff in the first sentence cold; I guess there are some suggestive clues there, but this is almost certainly more lead-in (and much harder lead-in) than is necessary. After that, the initial mass function and ambipolar diffusion clues seem more important and well-known to me than Bonnor-Ebert and possibly Herbig-Haro, but my views may be skewed.

magnetic impurities: possibly hard for a tossup, and I'm not sure the semiconductor clue works with this question's construction--semiconductor dopants are not generally magnetic impurities.

path integral/Feynman/loop quantum gravity: I was going to carp about lqg, but let's chalk that up to a defect in my personality and move on


eutectic: this feels a bit like superstring theory--I can figure out very quickly what kind of answer we're going for, but I can't be sure it's not peritectic or eutectoid or some other ridiculous thing. Well, maybe it's a bit less annoying than superstring theory since it feels more like stuff that people might learn in some undergrad class, plus the possible alternative answers seem more far-fetched.

quadrupole: good stuff; probably more lead-in than is needed, but at least every clue feels useful

Hubble constant: solid; I don't believe I've ever heard the phrase "Malmquist bias" before (everyone just says "selection bias")

weak/neutral current/CKM: good stuff; I appreciate the effort to give away neutral current, but I wonder if enough people have even heard of that to make it an acceptable medium part


atmosphere of Venus: The first 3 sentences don't really do anything for me, and I have no idea if they're really unique to Venus's atmosphere.

Clausius-Mossoti relation: probably too hard for a tossup

Miller/Bragg/goinometer: Miller indices feels slightly hard for a medium part and goinometer feels very hard for a hard part (but hilarious to say, so that's something)

geochemistry/Goldschmidt/pyroxenes: as amusing as it is to contemplate a future in which we write questions on "atomic physics" and "zoology," I'm not a big fan of the first part. Goldschmidt seems important but may be too hard; pyroxenes are great.


fractional quantum hall effect: we're definitely wandering into "stuff I encountered only through quizbowl"-land.

termination shock/RTGs/IBEX: this feels kind of hard in the easy, medium and hard parts.

Newtonian/incompressible/shear thinning: all good stuff, but no true easy part--presumably it would be easy to make Newtonian into an easy part by adding some non-fluid clues.


momentum: looks good

metallicity: possibly not widely-known enough for a tossup

saltation: this has three sentences of lead-in level material, then "it happens on dunes," then some stuff about Bagnold's equation (which I've never heard of). The rest seems generally useful. I'd probably move the bit about dunes to just before FTP, cut most of the lead-in, and then look for some more middle material.

K40/inner core/remanence: looks generally good to me, except that the inner core doesn't generate the geomagnetic field.

mean free path/cross-section/drift velocity: good stuff, if noticeably easier than many other science bonuses (actually, drift velocity seems easier than some of the medium parts in other bonuses); drift speed should be acceptable in the third part.


power: good stuff--lead-in clues that are useful, always a plus.

Lagrange multipliers: I like this question and wish there were more questions on "super-useful math techniques that are used everywhere"

geostrophic/isobars/baroclinic: good stuff--isobars for the earth science-impaired, and baroclinic is a nicely gettable (I think) and important hard part.

cyclotron frequency/effective mass/semiconductors: kind of annoying that the first part wasn't written to allow both cyclotron and gyro/Larmor frequency, but oh well. I don't know anything about effective mass but I trust Jerry/Eric that it's fine as the hard part.


-Seth

[Auroni]

atmosphere of Venus: The first 3 sentences don't really do anything for me, and I have no idea if they're really unique to Venus's atmosphere.

I wrote this tossup since I thought it would be a cool idea that hasn't really been explored in questions well. I took the early clues from instrumental findings of the atmosphere from fairly recent studies. Maybe other atmospheres have hydroxyl radicals, excited oxygen singlets, and S shaped bodies connecting polar vortices, but I thought that mentioning the instrument used for some calculations and mentioning the exact numbers would make it unique. I'm genuinely curious as to how you'd improve the question, though, since I want to get better at writing good astro questions on important subjects.

[Sima Guang Hater]

Post of the year. Let me give some insight into some of my editorial decisions here.

The too difficult TUs were obviously a problem everywhere and will be reined in for future iterations of this tournament. I think I took the opportunity to write a hard tournament way too far. On top of that, I'll freely acknowledge that magnetic impurities and vacuum polarization were outright bad ideas. I somehow got the impression that vacuum polarization was something that you would always know if you had taken a class in QED, and having not taken said class I don't know where I got that impression.

I wholly agree with the selection of easy and middle parts in an empathetic way; it might be worth talking about how to select and normalize medium and hard parts that reward people when you're looking in outside of a subject, because I can't figure it out. To me the Rayleigh number and Hohmann transfer are equally difficult (I've seen one in a class and another in a textbook), for example, and Turan's theorem is something I was taught for contest math. Its hard to get a perspective.

The cosmic dust tossup was both James Lasker and I really trying to stretch a subject out into a reasonable number of clues; I think that influenced the way its written. I would from now on just write the 6-line tossup even in a sea of 10 line monsters. Sort of the inverse of this problem happened in the star formation tossup, where James' textbook had a whole chapter on it and we wanted to pack in as much as possible.

The eutectic tossup was submitted by Michael Hausinger for chemistry, and I believe I kept it mostly as is because its something I trust that he knows well (I did fact-check it).

The star formation tossup, galaxies tossup, and bonus part on drift speed actually illustrate something that I wanted to refine for this tournament. The star formation tossup purposefully put clues about the initial mass function before Herbig-Haro objects (which is where several people buzzed), and the galaxies tossup put a clue about the dark matter halo before named things about galaxy rotation curves on purpose, even if those things might be easier to people who know those topics from a class. This was done on purpose; I tried to differentiate between buzzes based on named objects/clue absorption from packets by occasionally reversing those clues with things more well-known in the field. I know this isn't revolutionary, but I think its something worth doing more.

All in all I think the thing to take away is to aim way easier than you think you're aiming for.

[setht]

atmosphere of Venus: The first 3 sentences don't really do anything for me, and I have no idea if they're really unique to Venus's atmosphere.

I wrote this tossup since I thought it would be a cool idea that hasn't really been explored in questions well. I took the early clues from instrumental findings of the atmosphere from fairly recent studies. Maybe other atmospheres have hydroxyl radicals, excited oxygen singlets, and S shaped bodies connecting polar vortices, but I thought that mentioning the instrument used for some calculations and mentioning the exact numbers would make it unique. I'm genuinely curious as to how you'd improve the question, though, since I want to get better at writing good astro questions on important subjects.

I agree that this is a cool topic that hasn't really been explored in questions; it certainly feels a lot more worthwhile and accessible than another tossup on the 15th moon of Uranus or whatever. I think the problem for me is that I just don't know that many deep things about the Venusian atmosphere. I don't know if that's just my problem and this played fine for other people, or if no one really had a shot at the first three sentences; I would guess the latter, since I imagine I've taken more planetary science classes than anyone else playing quizbowl, but I could be wrong. One way to check would be to look up syllabi, I guess; I know my own classes devoted very little or no time to talking specifically about Venus's atmosphere. If deep knowledge of Venus's atmosphere is super-rare (or nonexistent in quizbowl) then I think the way to approach this question is either to write it on Venus and sprinkle in some clues about the atmosphere, or keep it on Venus's atmosphere but cut (say) two of the first three sentences and devote more space to signaling "I'm talking about Venus here"--I think you've got a decent number of clues indicating that you're talking about some body's atmosphere, so the buzz-limiting step is figuring out which body's atmosphere is being discussed.

The clues that mean something more to me than "some body's atmosphere" are: the transit/Lomonosov clues, eruption, no magnetic field and solar wind, super-high pressure, runaway greenhouse, sulfuric acid clouds, second planet. Of those, to me the transit and Lomonosov clues point more definitely to Venus than the subsequent eruption and no magnetic field/solar wind clues, but I suspect the clue order is fine since I'm guessing more people can get something useful (if perhaps not 100% definitive) out of the latter clues. High pressure is good; I happen not to know (or remember) that Venus's atmosphere is so dominated by carbon dioxide, but that's probably also fine. You could add Venus's loss of water to the runaway greenhouse clue in place of the rather vague "thought to be responsible for the current state of this entity" (I feel somewhat similarly about all the non-eruption stuff in the sentence that mentions an eruption: the fact that there was an eruption seems useful to me, but mentioning UV radiation, oxidation and its upper portion doesn't really do anything for me beyond sort of signaling "atmosphere"). The later clues all seem fine. If you wanted to cut some lead-in and add some more middle/late clues, you could do stuff like talk about craters on Venus--I know the atmosphere is very effective at destroying the smallest impactors, and from looking around a bit apparently Venus has a bunch of misshapen craters that are probably caused by the atmosphere breaking up incoming bodies. Basically anything that hints at high pressures/high temperatures and/or a rocky planet underneath could be useful--e.g. you could put in something like "its pressure drops to less than 50 bars at the peak of Maxwell Montes." For earlier clues, maybe talk about the Magellan spacecraft and how it was designed to map the surface despite the thick atmosphere.

As I look over the question again, I think it actually has more clues that could be used (in combination) to deduce Venus than I originally thought, but I think it might be good to err on the side of giving even more such clues: the topic is not one that comes up almost at all in quizbowl, so even players with some knowledge won't be accustomed to assembling clues to arrive at "I should buzz with something about Venus."

My team wound up not hearing this packet, so I can't be sure how this question would have played out, but those are my very long-winded suggestions for making it more likely that someone with my background would buzz early.

-Seth

[setht]

I somehow got the impression that vacuum polarization was something that you would always know if you had taken a class in QED, and having not taken said class I don't know where I got that impression.

This might actually be true--I also don't know, since I haven't taken a class in QED--but I think it still wouldn't make vacuum polarization a reasonable TU topic, since I don't think a class in QED is part of any physics undergrad curriculum (it's certainly not at Berkeley, Chicago, Wisconsin or Brown). Actually, I think it's not only not a standard part of the undergraduate curriculum, I think it's also generally not offered at all at an undergraduate level--I didn't find any undergrad elective courses in QED while checking some department websites just now.

I wholly agree with the selection of easy and middle parts in an empathetic way; it might be worth talking about how to select and normalize medium and hard parts that reward people when you're looking in outside of a subject, because I can't figure it out. To me the Rayleigh number and Hohmann transfer are equally difficult (I've seen one in a class and another in a textbook), for example, and Turan's theorem is something I was taught for contest math. Its hard to get a perspective.

It certainly is hard to get perspective, and I know it's very hard to compare or figure out difficulty levels in subjects where one has no personal course experience. For something like Rayleigh number vs. Hohmann transfer, I would imagine that Rayleigh number shows up in many more textbooks (and across more [and more common] disciplines--geophysics, astro, physics, engineering), and that when it shows up it generally gets more space than Hohmann transfer does; similarly I imagine it shows up in more course syllabi. Turan's theorem is probably fine; I don't know much about graph theory, and that's really on me.

The star formation tossup, galaxies tossup, and bonus part on drift speed actually illustrate something that I wanted to refine for this tournament. The star formation tossup purposefully put clues about the initial mass function before Herbig-Haro objects (which is where several people buzzed), and the galaxies tossup put a clue about the dark matter halo before named things about galaxy rotation curves on purpose, even if those things might be easier to people who know those topics from a class. This was done on purpose; I tried to differentiate between buzzes based on named objects/clue absorption from packets by occasionally reversing those clues with things more well-known in the field. I know this isn't revolutionary, but I think its something worth doing more.

I very much support this approach; if people at the Penn site were buzzing more on the named stuff, then I think you got the clue ordering right. If drift speed/velocity achieved approximately the correct conversion rate for a hard part by having lots of scientists get it and lots of non-scientists stare blankly then that's also good. We could probably do with more "hard" science bonus parts that are actually not so hard for science people.

-Seth

[grapesmoker]

I somehow got the impression that vacuum polarization was something that you would always know if you had taken a class in QED, and having not taken said class I don't know where I got that impression.

This might actually be true--I also don't know, since I haven't taken a class in QED--but I think it still wouldn't make vacuum polarization a reasonable TU topic, since I don't think a class in QED is part of any physics undergrad curriculum (it's certainly not at Berkeley, Chicago, Wisconsin or Brown). Actually, I think it's not only not a standard part of the undergraduate curriculum, I think it's also generally not offered at all at an undergraduate level--I didn't find any undergrad elective courses in QED while checking some department websites just now.

Neither Berkeley nor Brown have QED in the undergrad curriculum.

I wholly agree with the selection of easy and middle parts in an empathetic way; it might be worth talking about how to select and normalize medium and hard parts that reward people when you're looking in outside of a subject, because I can't figure it out. To me the Rayleigh number and Hohmann transfer are equally difficult (I've seen one in a class and another in a textbook), for example, and Turan's theorem is something I was taught for contest math. Its hard to get a perspective.

Hohman transfer and Rayleight number are not anywhere close to the same degree of difficulty. The former is much, much harder and was not mentioned in any of my dynamics classes, while the Rayleigh number featured in an undergrad mech. eng. class I took. Just saying.

The star formation tossup, galaxies tossup, and bonus part on drift speed actually illustrate something that I wanted to refine for this tournament. The star formation tossup purposefully put clues about the initial mass function before Herbig-Haro objects (which is where several people buzzed), and the galaxies tossup put a clue about the dark matter halo before named things about galaxy rotation curves on purpose, even if those things might be easier to people who know those topics from a class. This was done on purpose; I tried to differentiate between buzzes based on named objects/clue absorption from packets by occasionally reversing those clues with things more well-known in the field. I know this isn't revolutionary, but I think its something worth doing more.

I very much support this approach; if people at the Penn site were buzzing more on the named stuff, then I think you got the clue ordering right. If drift speed/velocity achieved approximately the correct conversion rate for a hard part by having lots of scientists get it and lots of non-scientists stare blankly then that's also good. We could probably do with more "hard" science bonus parts that are actually not so hard for science people.

I'm not a star guy, but is "initial mass function" unique to stars? I swear I encountered this term when studying large-scale structure formation, but this time I wisely delayed buzzing until further clues were given. Anyway, I thought it was a fine questions, but am curious about the terminology employed for my edification.

[theMoMA]

Reading through the tossup on metallicity (which was an interesting idea), it seems like the kind of thing that's hard to toss up in a way that isn't pretty figure-out-able for anyone who knows what the definition is.

[setht]

I'm not a star guy, but is "initial mass function" unique to stars? I swear I encountered this term when studying large-scale structure formation, but this time I wisely delayed buzzing until further clues were given. Anyway, I thought it was a fine questions, but am curious about the terminology employed for my edification.

I think initial mass function is pretty much always used to refer to stars; I suppose it may have been co-opted for something else, but I think it's safe to say that if you walk up to an astrophysicist and start talking about initial mass functions they're going to assume you're talking about stars. I think structure formation people usually talk about density perturbations/fluctuations, or wavenumbers of Fourier modes of density perturbations, not initial masses.

-Seth

[DumbJaques]

PSA: There is nothing wrong with having a tossup be 6 lines, rather than 10, if that's the ideal form of the tossup. Spending 40 minutes looking for another leadin to add just so you can make it longer is not a good idea will only bring misery to all parties involved (and this is speaking as someone who has done this a number of times in the past).

On a related note, there are certainly a lot of tossups in this tournament that did not need to be anywhere near as long as they were. I generally don't actively mind that kind of thing - and didn't much when I was playing MO - but players less accustomed to it certainly do and really, you do take quite an appreciative note when a tournament feels like VCU Open has the last two years.

I know I still find myself surprised when people don't buzz as early as I thought they would when reading my questions at a tournament, but really, I shouldn't be - quizbowl is hard. I have a feeling that we'd all observe a similar phenomenon while reading our 10-line tossups on hard-ass crap; how many times have you looked at a question and thought "yeah, I really could have made that a lot shorter?" It's going to absolutely dwarf the amount of times you've had the opposite reaction (good writers very rarely produce clunker buzzerace tossups, and even then it's almost NEVER that the tossup doesn't have enough info, but rather that a clue is misplaced). Of course, when we make errors that lead to clunker questions and ensuing buzzer slamming/profanity/chair mass ejections/etc., we understandably end up with a huge context bias.

Couple all that with the urge to put everything you've just spent time learning about into a question, and you end up with 10-14 line tossups saturating your tournament. But it's never necessary, and in fact it's often a big drawback. Aside from the obvious negatives, stacking clue after clue often leads people to worry less about smoothing out difficulty transitions; this actually makes difficulty cliffs and buzzeraces MORE likely, not less.

So yeah, I get why we write 14 line questions, and it's understandable; we still need to stop.

[Auroni]

Yeah, Chris is right. Additionally, I think that the main reason these tournaments are so draining is that there'sa million rounds of 9-10 line long tossups on average, requiring an intense amount of focus to process all of the clues. MAGNI's line cap was pretty well-liked.. maybe it's time to enforce that MO or other hard tournaments.

[Magister Ludi]

I'm on the other side of the line limit debate. I don't think we need 10 lines to differentiate who knows the most about Henry Green, but I think having more lines is correlated with having quality questions on easy topics in hard tournaments. You actually could use 10 lines to differentiate all levels of knowledge on a tossup on a character from Hamlet because players are actually likely to have many different levels of knowledge about that topic. Moreover, people should remember that the distinctions between the best players within a category are significant at nationals and hard open tournaments. It's ok if an expert occasionally loses a buzzer race on a topic they know better than the person who buzzed at Penn Bowl because it's a regular difficulty tournament and it's unwise to devote multiple lines to making a distinction between the two best chemistry players in the country. But that kind of distinction really does matter at the major open tournaments. The physics tossup does need to differentiate whether Jerry or Seth or Sorice knows more about that topic. Especially for nationals tournaments like ACF Nationals (and perhaps Chicago Open) the most important thing is not to trim the line limit to meet someone's arbitrary standards, but to make sure that the most knowledgeable players answer the questions. I think controlling line length is more important for tournaments like Fall or Regionals that are actively involved in recruiting new teams, but the people coming to a tournament like MO, CO, or VCU Open know what they're getting into and won't be repulsed a few extra lines.

[magin]

I disagree with Ted; I think 7-8 lines is fine for easy answer lines at hard tournaments as long as you choose appropriate answer lines and clues. To my mind, Ted's argument goes astray by assuming that knowledge is mostly linear: Person A has 5 units of knowledge about Garcia Marquez, and Person B has 3 units of knowledge, so Person A should always get a Garcia Marquez tossup before Person B. (Ted can correct me if my assumption is wrong, but that's the sense I have from his argument.) I don't think "knowing more" works that way; in quizbowl, "knowing more" is about the probability of getting questions.

For instance, I have a really low probability at beating Eric Mukherjee to a chemistry tossup, since he knows far, far, far more chemistry than I do. Even though he knows far more, if I take the time to learn a little chemistry and hear a clue I know, or if he doesn't play well, I have a chance of getting a chemistry tossup against him. In the long run, he's going to answer the vast majority of chemistry questions against pretty much anyone, which is how it should be. But in the short run, knowledge is more like an infinite wall whose sections are constructed piece by piece than a straight line. No matter how vast your wall is, you'll always have some gaps. If you have more pieces than someone else, you'll get more tossups in the long run against them, but they may well have an individual piece you don't.

[Auks Ran Ova]

Not to take anything away from MAGNI, god-king of tournaments, born under a double rainbow on Baekdu Mountain, but I'm not sure that MO has an imperative to strive for the same sort of goals that regular-difficulty tournaments do. The audiences of open tournaments like MO are self-selecting to a large degree--it's not really necessary to make lots of changes to appeal to a much wider audience. There's nothing inherently wrong with shorter tossups, but when I'm not writing something designed to be difficulty-capped, I tend to be rather prolix (partly) because of the very freedom I have to be so. The intended audience of a difficult open tournament can handle slightly longer questions, and there's no real harm in working for a finer gradation of knowledge or tossing in an extra clue you find cool when writing for such an audience. I do agree that 14-line monstrosities are inadvisable, but most of this tournament's tossups (the vast majority of the ones I wrote or edited, at least) were generally south of ten lines, and that's not even taking into account the fact that I picked a big font for the packets that throws off at-a-glance length estimation based on normal standards.

Essentially, there's nothing at all wrong with a hard tournament limiting itself to a seven- or eight-line cap, but I don't think MO--or any given difficult open tournament--has any sort of duty to do so.

[Cheynem]

Almost all of the questions I wrote were eight lines, sometimes veering into nine if it required an especially wordy giveaway (or after powermarking). I think open tournaments are reasonable places to bust out (in moderation) a few ten-line questions on especially interesting or quixotic topics which require more clues or finer gradation. I'm sympathetic to shorter questions and line caps, but one thing I disagree with is the concept that there are some hard and fast rules that all tournaments and questions should obey--I don't really see quizbowl as that scientifically uniform.

[Skepticism and Animal Feed]

My view is that there is nothing inherently wrong with 7-line, or even 6-line, tossups at hard open tournaments. I think that a six line tossup, if properly written, can still distinguish between top teams.

If people want to add 3-4 more lines of good, interesting clues, then I am not opposed. If people want to add confusing clues, or pish-posh like "One man said of him, [obscure quote you found in an obscure source]. That quote came from the servant of this man's brother", then I condemn this.

[Magister Ludi]

I disagree with Ted; I think 7-8 lines is fine for easy answer lines at hard tournaments as long as you choose appropriate answer lines and clues. To my mind, Ted's argument goes astray by assuming that knowledge is mostly linear: Person A has 5 units of knowledge about Garcia Marquez, and Person B has 3 units of knowledge, so Person A should always get a Garcia Marquez tossup before Person B. (Ted can correct me if my assumption is wrong, but that's the sense I have from his argument.) I don't think "knowing more" works that way; in quizbowl, "knowing more" is about the probability of getting questions.

For instance, I have a really low probability at beating Eric Mukherjee to a chemistry tossup, since he knows far, far, far more chemistry than I do. Even though he knows far more, if I take the time to learn a little chemistry and hear a clue I know, or if he doesn't play well, I have a chance of getting a chemistry tossup against him. In the long run, he's going to answer the vast majority of chemistry questions against pretty much anyone, which is how it should be. But in the short run, knowledge is more like an infinite wall whose sections are constructed piece by piece than a straight line. No matter how vast your wall is, you'll always have some gaps. If you have more pieces than someone else, you'll get more tossups in the long run against them, but they may well have an individual piece you don't.

I think you're using the wrong example. A better example would be if I'm playing you on a literature question where two people do know a lot about a subject but one person knows decidedly more. For example, Jonathan definitely knows more than I do about Tristram Shandy, but I've written a couple questions on the book and know a decent amount of clues. If we cut off the first couple sentences of the Tristram Shandy tossup in this set then I would be buzzer racing with Jonathan on the first clue. I agree with Jonathan that there is no absolute way to ensure the the person who knows more will always answer the question, but I think we should use extra lines at hard tournaments to increase the probability that the person who knows the most will get the question. A further example, I beat Jonathan to more lit tossups at VCU Open than I ever have at any other tournament I've played and in those three games I only got one tossup on a book I've read, so I wasn't getting lucky as much as there was something about that format which objectively blurred the distinction between the players with the highest levels of knowledge. I'm not saying it's necessarily bad if writers decide to write slightly shorter tossups, but we shouldn't pretend that shorter is inherently better because it's an equally valid editorial decision to write longer questions. The legitimate use of extra lines is to improve a question's ability to determine who knows the most about a topic.

[Auroni]

For example, Jonathan definitely knows more than I do about Tristram Shandy, but I've written a couple questions on the book and know a decent amount of clues. If we cut off the first couple sentences of the Tristram Shandy tossup in this set then I would be buzzer racing with Jonathan on the first clue.

First of all, trimming a tossup down to 7 or 8 lines doesn't have to be (and, in fact, probably won't be in 99% of all cases) "cut[ting] off the first couple sentences." You can cut off one early clue and one middle clue to bring it down a line and a half.

Second of all, even granting you that the first two lines were cut, it's one thing to say you'll be buzzer racing on the first clue and it's another thing entirely for reality to bear that out. Buzzer races on first clues even between two great players at hard tournaments are so rare they are effectively negligible in this argument.

[Auks Ran Ova]

Buzzer races on first clues even between two great players at hard tournaments are so rare they are effectively negligible in this argument.

To play devil's advocate for a second: One could argue, as Ted is doing, that this is true at least in part because hard tournaments in the past have generally featured the longer tossups Ted is advocating, thus providing the exact effect he's arguing for--namely, that first-clue buzzer races don't happen.

edit: This does sort of ignore the fact that making tossups shorter obviously doesn't necessarily exactly correspond to "removing the hardest, 'most-differentiating'" clues", of course. I tend to sympathize with Ted's argument for slightly different reasons--I'd support longer tossups on easier answers in hard tournaments because they provide more harder-clue buzzpoints and thus provide more opportunities for Tristram Shandy experts, or whatever, to distinguish their knowledge before the question starts rolling into more generally familiar material. I don't really want to see leadin-heavy monstrosities either, but more clues, used properly, can indeed allow for a finer gradation of knowledge.