2019 ICT DI specific question discussion

[theMoMA]

This is your thread for specific question discussion of DI ICT.

[Nabonidus]

My only specific complaint is that the bonus on Miles Vorkosigan should have Miles underlined and not Vorkosigan, especially since there are lots of Vorkosigans in those books.

[excessive dismemberment]

I have some specific complaints that will probably sound more negative on the tournament as a whole than I really am:
- The clue in I think it was the coatlicue tossup about a goddess creating her son fully formed was negbait for Parvati, and could/should have been worded better.
- The clue about Anchises in the Ascanius tossup saying that he decided to flee troy sounds at game speed a bit like you were asking for Ascanius's son the way it was written. I was going to say Ascanius based on the symbols mentioned, but the word grandfather at the end of the clue threw me off. Maybe that's something that only happened to me, but I think my complaint that the clue could have been worded better is valid.
- The bonus on bhishma/karna/arjuna was way easier than any of the rest of the myth bonuses in the tournament. All 3 of them are very major characters in the Mahabharata, and they would all be fine as bonus parts even in a fall level tournament I think
- Maybe I'm wrong, but the David tossup cluing david frum seemed to be somewhat miserly powermarked (very possible I'm misremembering this, and it's a fairly minor complaint
- I found the tossup on lemnos to have utterly useless clues for 3/4 of the tossup, and honestly I was surprised that it was tossed up

[CPiGuy]

Mentioning the principle of explosion in the first line of a tossup on "contradictions" would be too easy at regular difficulty, let alone a national.

The "ocarina" tossup was kind of fraudable; I'd have used "this item" rather than "this instrument" at least in the first clue. There are only so many musical instruments in popular video games.

The question on "word order" was a cool linguistics thing.

Packet 5 had a biology tossup on islands followed by a geography tossup on the Channel Islands, which didn't affect gameplay but is kind of unfortunate.

I can't believe you tossed up "boustrophedon" but that was Extremely Fun.

Mentioning slavery in power for Mauritania is too early.

[naan/steak-holding toll]

I found the tossup on lemnos to have utterly useless clues for 3/4 of the tossup, and honestly I was surprised that it was tossed up

I didn't write this but I'm not sure why the clues were useless - there are lots of stories about Lemnos and I think plenty of them are worth knowing. The question also gave you power for knowing Hypsipyle, who appears in the Argonautica. This seems fair to me, even generous (though reasonably so).

[Wartortullian]

Despite NAQT stuff being generally outside of my wheelhouse, I greatly enjoyed this tournament. Most of the questions were extremely well executed, and the weirder answerlines tended to be on the more interesting side.

Questions that I particularly liked:

- The drifts tossup in the final did an excellent job of testing core knowledge in an area of physics that, in my incredibly biased opinion, is generally underasked. I felt the same way about the longitudinal/Debye length/electron density bonus. The reconnection tossup, though a bit on the stocky side, was also strong.

- I'm not sure how the boustrophedon tossup played overall, but it taught me something new that turned out to be really cool.

- I thought a lot of the "spicier" lit tossups on core works, like the ones on Room 101 and the circumcision of Tristram Shandy, were well-executed.

- The Blake/Giger/Ginastera bonus clued from Brain Salad Surgery may be my favorite bonus since last year's "Box & Cox." Also, the fact that this tournament had 2/1 progressive rock (with Syd Barrett and the King Crimson clues in 21st Century) warms my heart for entirely selfish reasons.

- One of you wrote a bonus in which the webcomic Questionable Content was an answerline. You rock.

Questions that I had issues with:

- The leadin for spontaneous (SPDC) was rather easy and reduced to word association.

- While I liked the idea of the tossup on spin-1/2 systems, I felt it was harder to recognize what the tossup was looking for than it was to actually answer it.

- I disliked the tossup on The Good Place for the reasons that Will elaborated on in the other thread. This is an specially egregious example, as the show in question is:
1. primarily engaged with through Netflix, which is conducive to bingeing
2. highly serialized, so people are more likely to have specialist knowledge or none at all, and
3. popular enough that it's not unreasonable to expect most teams to first-line it.

I'll probably have more comments as I read through the packets again, but all in all, this was a great set!

[Cheynem]

One of the challenges of TV questions as well is that for current TV, it's almost like current events in terms of "freshness." I wrote The Good Place tossup in 2017, for example. It wasn't very good, in my opinion, but it might have been better had it been more recent and could draw upon a greater range of material. I do think it's hard to find TV shows that aren't especially suited for binging or serialized watching, though, without asking about fairly frivolous programming.

[naan/steak-holding toll]

One of the challenges of TV questions as well is that for current TV, it's almost like current events in terms of "freshness." I wrote The Good Place tossup in 2017, for example. It wasn't very good, in my opinion, but it might have been better had it been more recent and could draw upon a greater range of material. I do think it's hard to find TV shows that aren't especially suited for binging or serialized watching, though, without asking about fairly frivolous programming.

Exactly: this is precisely why TV should be cut down. There aren't that many shows, particularly recent ones, where the field is going to have a wide variety of levels of knowledge. For people who haven't seen the show, a question on some recent Netflix series that goes dead feels like a waste of time in a way that one on, say, an important concept in psychology, or a highly influential film director, does not. I've never heard of The Good Place and that question did not really make me want to go learn about it more, nor do I think any possible question on The Good Place could make me want to go learn about it or watch it, and I suspect this is true of most people in the field who haven't watched it. I guess this applies to a lot of the more technical subjects in quizbowl as well, but at the very least those merit their place through being academic subjects - I think non-academic subjects need to be held to a higher standard.

[Cheynem]

To be clear, The Good Place is a NBC series, not a Netflix series, and is a fairly intelligent, philosophy-driven series; I just wrote the question pretty badly.

[theMoMA]

Thanks for the interesting thoughts on the current events and TV distributions. In some ways, I think the criticisms of those questions are similar, in that they boil down to a dislike of questions that seem to clue from a fairly deep, but not very wide, vein of clues. Put another way, if a player has seen the entirety of a TV show, or has followed a particular recent news story closely, then the pyramidal structure of the question might be a bit easy for them, because they are very familiar with body of facts that are being clued; on the flipside, a player who has not seen the entirety of a show, or has little knowledge of a particular news story, will not be served by any of the clues before the giveaway. This could create a more polarized distribution of buzzes than is ideal.

Perhaps it's worth thinking about the place of certain kinds of questions in the distribution, whether that's political current events or modern TV, but I also think it's worth thinking about how we write the questions (however many there are) that slot into these categories. My suspicion is that we could take a step back from asking about specific TV shows or minutiae of unfolding events and focus on things of wider significance in these categories, whether that's something like asking about an actor who has been on multiple shows, a politician who has done several significant things, or a trend or theme of some kind that underlies various shows or news stories. It would take a bit more work to identify broader answer lines like this, but I think it would be worth it, because the questions that would result would be more pyramidal for everyone and would ideally hone in on some interesting material that people who care about these topics find more engaging.

[naan/steak-holding toll]

I agree with Andrew Hart wholeheartedly. To pick on the Avenatti tossup even further though, because I think it's a good example of the CE issue, the stuff he posts on his Twitter account frankly is ephemeral and does not have an impact on many things, even if many people have indeed interacted with it. To dig something up from my area of professional expertise for comparison - a very important law like MACRA which seriously affects the country's biggest industry will basically never get talked about in the CE distribution as it exists today. It probably won't in the news either, but I think that's much more a reflection on the quality of our news media than anything else. Our distribution need not be bound by the topics du jour of the pundits, talking heads, and chattering classes. I'm more than fine with deep cuts on important issues, in the same way that deep tossups on core literature works are awesome, but that doesn't seem to be what we're doing with a lot of these tossups.

[Zealots of Stockholm]

I thought the tossup on "Rudy Tomjanovich" could have been made both more interesting and more accessible by being written on "fighting (in the NBA)." The Tomjanovich question primarily used clues from the 70s in the NBA, and a question on fighting could easily be clued entirely from the 70s and maybe 80s, as it was a large issue that has now been mostly eliminated from the game, and would have no need for a trivial lead in asking who Michigan's all time leading rebounder is. That being said, I did like the tossup aside from the lead in, I just imagine it went dead in a lot of rooms.

I also agree with what was said above about TV questions.

[naan/steak-holding toll]

I thought the tossup on the C++ Standard Library was extremely suboptimal. In my room, a player buzzed in with "C++" and got prompted and could not pull the prompt. However, the answerline instructions accepted just "C++" after a certain point in the tossup, and the other team buzzed with the exact same answer that drew a prompt and neg earlier in the tossup and received points. This seemed extremely unfair to me - is there a reason the entire question couldn't just be on C++ or the Standard Library of C++?

[Jet Fuel Can't Melt Steel Dreams]

I thought the tossup on the C++ Standard Library was extremely suboptimal. In my room, a player buzzed in with "C++" and got prompted and could not pull the prompt. However, the answerline instructions accepted just "C++" after a certain point in the tossup, and the other team buzzed with the exact same answer that drew a prompt and neg earlier in the tossup and received points. This seemed extremely unfair to me - is there a reason the entire question couldn't just be on C++ or the Standard Library of C++?

I also recall being fairly certain the answer was Boost two lines in when the tossup was talking about smart pointers before an opponent buzzed in with Boost. I think (I don't remember exactly) he later insisted that Boost was correct as an answer.

[VSCOelasticity]

I thought the tossup on the C++ Standard Library was extremely suboptimal. In my room, a player buzzed in with "C++" and got prompted and could not pull the prompt. However, the answerline instructions accepted just "C++" after a certain point in the tossup, and the other team buzzed with the exact same answer that drew a prompt and neg earlier in the tossup and received points. This seemed extremely unfair to me - is there a reason the entire question couldn't just be on C++ or the Standard Library of C++?

I also recall being fairly certain the answer was Boost two lines in when the tossup was talking about smart pointers before an opponent buzzed in with Boost. I think (I don't remember exactly) he later insisted that Boost was correct as an answer.

That was me. In hindsight, I was wrong because the TU says "in 2011", referring to c++11 changes that brought smart pointers to the standard library. I'm not sure I really like this 2 word phase being used to distinguish between the two answers, but it factually correct.

I like the TU idea. Many people care about the changes to C++ std library. I think it would've been better to steer away from stuff also implemented in boost though.

I agree with Will that the full answer should be required all the way through. There were plenty of hard tossups at ICT. There's no reason to make this one more accessible to the point that the question gets significantly easier: from knowing the c++ std library to knowing c++ is a language.

[Mike Bentley]

I thought the tossup on the C++ Standard Library was extremely suboptimal. In my room, a player buzzed in with "C++" and got prompted and could not pull the prompt. However, the answerline instructions accepted just "C++" after a certain point in the tossup, and the other team buzzed with the exact same answer that drew a prompt and neg earlier in the tossup and received points. This seemed extremely unfair to me - is there a reason the entire question couldn't just be on C++ or the Standard Library of C++?

I also recall being fairly certain the answer was Boost two lines in when the tossup was talking about smart pointers before an opponent buzzed in with Boost. I think (I don't remember exactly) he later insisted that Boost was correct as an answer.

That was me. In hindsight, I was wrong because the TU says "in 2011", referring to c++11 changes that brought smart pointers to the standard library. I'm not sure I really like this 2 word phase being used to distinguish between the two answers, but it factually correct.

I like the TU idea. Many people care about the changes to C++ std library. I think it would've been better to steer away from stuff also implemented in boost though.

I agree with Will that the full answer should be required all the way through. There were plenty of hard tossups at ICT. There's no reason to make this one more accessible to the point that the question gets significantly easier: from knowing the c++ std library to knowing c++ is a language.

Could this tossup be posted? Reading it I was similarly confused about the answer line based on the clues but would like to confirm.

[Important Bird Area]

David Musser created the hybrid introsort algorithm during the initial development of this code collection. A 2011 update to this collection introduced "smart pointers" for automated memory management. Although written in the same language, the Boost libraries are not part of this collection. This collection was modeled on the highly templated (*) STL and contains several of the STL's header files. The std namespace wraps functions and classes in—for 10 points—what language's standard library?

answer: C++ Standard Library (accept just C++ after "language's"; prompt on "C++" or "Standard Library" before "language's"; prompt on "std namespace" or "standard namespace" or "namespace std" before "std"; do not accept or prompt on "STL" or "Standard Template Library")

[CPiGuy]

I seem to remember a tossup on neutron star mergers which clued LIGO -- is my memory mistaken or did this happen? If so, LIGO's first detection of gravitational waves was definitely the result of a black hole merger.

[Wartortullian]

Agreeing with Will, Geoff, and Jonathan here. I knew it was either Boost or the C++ standard library from the first clue, but I sat until Boost was mentioned.

[Bosa of York]

I seem to remember a tossup on neutron star mergers which clued LIGO -- is my memory mistaken or did this happen? If so, LIGO's first detection of gravitational waves was definitely the result of a black hole merger.

My memory of this question is that that sentence was saying something about LIGO's first detection of gravitational waves caused by one of these events, rather than saying that LIGO's first detection of gravitational waves was caused by one of these events (though I'd appreciate it if anyone with the question in front of them could confirm or contradict this).

[Jet Fuel Can't Melt Steel Dreams]

I seem to remember a tossup on neutron star mergers which clued LIGO -- is my memory mistaken or did this happen? If so, LIGO's first detection of gravitational waves was definitely the result of a black hole merger.

I remember hearing a bonus part on binary neutron star mergers. I said kilonovas, which I'm fairly sure should have been accepted (it wasn't).

[setht]

I seem to remember a tossup on neutron star mergers which clued LIGO -- is my memory mistaken or did this happen? If so, LIGO's first detection of gravitational waves was definitely the result of a black hole merger.

My memory of this question is that that sentence was saying something about LIGO's first detection of gravitational waves caused by one of these events, rather than saying that LIGO's first detection of gravitational waves was caused by one of these events (though I'd appreciate it if anyone with the question in front of them could confirm or contradict this).

The ninth detection of gravitational waves by LIGO ["LIE-go"] was the first seen from what type of event, which also generated a gamma ray burst detected by the Fermi and INTEGRAL ["integral"] telescopes in August 2017?

answer: neutron star merger (accept answers involving neutron stars colliding or a neutron star collision; prompt on "star collision" or "stars colliding"; do not accept or prompt on answers involving "black hole(s)")

I tweaked the phrasing of this bonus part to try to make it more clear to players that they should not hear "GW detection by LIGO" and immediately say "black hole merger!" I apologize if this was still confusing.

(And in response to Geoffrey, we should have had kilonovas listed as an acceptable alternative. I apologize for missing that.)

[Jet Fuel Can't Melt Steel Dreams]

Since we are on this subject, I'll mention that I was very very very very happy that "multimessenger astronomy" came up as the third part of that bonus.

[CPiGuy]

Ah, it was a bonus part and specified that it was the ninth such detection.

I still think this bonus part was kind of hosey (requiring people to know the LIGO detections in order is kind of unreasonable IMO), but I retract my claims of factual inaccuracy. Thanks for the response!

[Cody]

Ah, it was a bonus part and specified that it was the ninth such detection.

I still think this bonus part was kind of hosey (requiring people to know the LIGO detections in order is kind of unreasonable IMO), but I retract my claims of factual inaccuracy. Thanks for the response!

LIGO has detected gravitational waves from two types of mergers; the specific combination of ninth detection + first seen from one of these events just explicitly eliminates black hole mergers* and reduces to "name the other type of merger detected by LIGO / likely to be detected by LIGO" -- in no way does it require a player to know anything extra about the order of LIGO detections. I think it's an appropriate and clever way to rule out black hole merger without saying those words outright.

* because the fact that the first detection by LIGO was a black hole merger is extremely famous, as evidenced by your prior post

[CPiGuy]

Ah, it was a bonus part and specified that it was the ninth such detection.

I still think this bonus part was kind of hosey (requiring people to know the LIGO detections in order is kind of unreasonable IMO), but I retract my claims of factual inaccuracy. Thanks for the response!

LIGO has detected gravitational waves from two types of mergers; the specific combination of ninth detection + first seen from one of these events just explicitly eliminates black hole mergers* and reduces to "name the other type of merger detected by LIGO / likely to be detected by LIGO" -- in no way does it require a player to know anything extra about the order of LIGO detections. I think it's an appropriate and clever way to rule out black hole merger without saying those words outright.

* because the fact that the first detection by LIGO was a black hole merger is extremely famous, as evidenced by your prior post

oh, I also missed the words "the first". Yeah, this is totally fine and a cool way of asking about it. I should listen to questions more closely, apparently.

[Vinjance]

I'll also add that an answer of black hole merger doesn't really make any sense, since there's no way that such an event could generate gamma rays or any form of EM radiation. The Aug 2017 detection was unique in that it was the first such example of both light and gravitational waves being detected from the same event. I thought it was good idea to test the knowledge ICT players have of this detail.

[theMoMA]

I think the discussion of the "C++ standard library" question above illustrates an interesting and little-discussed feature of giveaway theory (which itself is a neglected topic in the larger world of quizbowl theory). There is a fine distinction between acceptable and frustrating giveaways when the last line of a question:

* Discloses part of the answer, which would have been required on early buzzes (i.e. this question);

* Gives an easy clue that is related in some way to the answer but does not draw from exactly the same body of knowledge clued before (the "cross-category giveaway"); or

* Gives one of two or more alternate names for the answer, which were previously acceptable earlier in the question.

Of these, I tend to avoid the third at all costs, because I find it incredibly frustrating to play a difficult question that ends "also known as [the answer you were going to say, and would have been correct before this point in the question]." The second can be fine and even good and interesting, but occasionally it has the effect of tacking a speedcheck on a loosely related subject onto the end of an otherwise difficult question. The first is something I tend to try to avoid the more difficult a tournament's level is, on the theory that questions should be "about" a consistent answer, and that a question that needs to disclose part of the answer is probably too hard (or people who can answer it after the disclosure probably don't deserve points, even if the question is on the hard side of acceptable), but it's also something I don't always have a problem with, on a case-by-case basis.

My bottom line is that I don't mind having the giveaway of a question retain purity with the rest of the question, rather than veering sharply to increase end-of-question conversion. Some questions can be a little harder, and requiring a player to know the easiest clue for a specific answer, rather than bending over backwards to give people with less knowledge a chance at points, seems like a fine practice (and avoids the speedcheck issue laid out above). This all assumes, of course, that the question in its difficult form fits into the difficulty structure of the tournament; if it doesn't, then it should usually be rewritten in a way that makes it consistently easier, as opposed to tacking on a giveaway that is not in line with the difficulty of the rest of the question.

In this case, because computer science is not my area of expertise, and because I wasn't primed to be thinking about this potential issue more than I normally would, writing the giveaway to allow the answer of C++ seemed fine to me. In retrospect, however, I can see how it would be frustrating, and it's good to have that issue identified for further discussion.

I'd be interested to hear more thoughts on giveaway theory and would be happy to make a thread in the public section of the forum that does not disclose non-clear question content if it's something people would like to discuss.

[Wartortullian]

Could someone post the question on reconnection? Some people in the plasma department got really excited and want to see it, and I don't have our binder.

[setht]

Could someone post the question on reconnection? Some people in the plasma department got really excited and want to see it, and I don't have our binder.

This process has a slow rate proportional to one over the square root of the Lundquist number in the Sweet-Parker model. The flux freezing seen in ideal MHD prohibits this process, but nonideal effects allow the necessary flux diffusion. This process may occur when a current sheet collapses, releasing (*) stored magnetic energy while changing the field topology. Solar flares are thought to be driven by—for 10 points—what process in which oppositely directed magnetic field lines merge?

answer: magnetic reconnection (or reconnecting; prompt on answers involving the "breaking of magnetic field lines" or "realignment of magnetic field lines" before "field lines")

[naan/steak-holding toll]

I think an OK analogy to this C++ Standard Library question is a tossup on, say, Ludwig van Beethoven's Piano Sonata No. 32. Normally, in such a question, you say "Beethoven" in the giveaway and still require Piano Sonata No. 32 since that is what identifies the piece you're talking about. Similarly, a tossup on the C++ Standard Library might say "C++" in the giveaway, and only require Standard Library after that. However, this question seems analogous to only requiring Beethoven - the far easier part of the answer - after a giveaway that says "What composer's Piano Sonata No. 32?"

It's not a perfect analogy, of course, especially since "this code collection" leaves a lot more room for player error than "this piece" but I think appropriately illustrates the magnitude of the difficulty drop. Questions like this seem ridiculously unfair to me.

[touchpack]

I think the discussion of the "C++ standard library" question above illustrates an interesting and little-discussed feature of giveaway theory (which itself is a neglected topic in the larger world of quizbowl theory). There is a fine distinction between acceptable and frustrating giveaways when the last line of a question:

* Discloses part of the answer, which would have been required on early buzzes (i.e. this question);

* Gives an easy clue that is related in some way to the answer but does not draw from exactly the same body of knowledge clued before (the "cross-category giveaway"); or

* Gives one of two or more alternate names for the answer, which were previously acceptable earlier in the question.

Of these, I tend to avoid the third at all costs, because I find it incredibly frustrating to play a difficult question that ends "also known as [the answer you were going to say, and would have been correct before this point in the question]." The second can be fine and even good and interesting, but occasionally it has the effect of tacking a speedcheck on a loosely related subject onto the end of an otherwise difficult question. The first is something I tend to try to avoid the more difficult a tournament's level is, on the theory that questions should be "about" a consistent answer, and that a question that needs to disclose part of the answer is probably too hard (or people who can answer it after the disclosure probably don't deserve points, even if the question is on the hard side of acceptable), but it's also something I don't always have a problem with, on a case-by-case basis.

My bottom line is that I don't mind having the giveaway of a question retain purity with the rest of the question, rather than veering sharply to increase end-of-question conversion. Some questions can be a little harder, and requiring a player to know the easiest clue for a specific answer, rather than bending over backwards to give people with less knowledge a chance at points, seems like a fine practice (and avoids the speedcheck issue laid out above). This all assumes, of course, that the question in its difficult form fits into the difficulty structure of the tournament; if it doesn't, then it should usually be rewritten in a way that makes it consistently easier, as opposed to tacking on a giveaway that is not in line with the difficulty of the rest of the question.

In this case, because computer science is not my area of expertise, and because I wasn't primed to be thinking about this potential issue more than I normally would, writing the giveaway to allow the answer of C++ seemed fine to me. In retrospect, however, I can see how it would be frustrating, and it's good to have that issue identified for further discussion.

I'd be interested to hear more thoughts on giveaway theory and would be happy to make a thread in the public section of the forum that does not disclose non-clear question content if it's something people would like to discuss.

As one of the people responsible for looking over that question and giving it the ok, in retrospect, would it have been better had the giveaway said "what collection of code in C++?" rather than "what language's standard library"? Perhaps that would have guided players to the answer while eliminating the ability to just guess a language at the end.