Common Link Discussion II: Electric Boogaloo

[DumbJaques]

This weekend's (otherwise exemplary) Minnesota Open set had, as has been noted in the general MO discussion thread, a number of common link tossups that seem to have resulted in extraordinary (and particularly frustrating) amounts of prompt-negs. Off hand, I can recall the trying to kill the Roosevelts, Japanese Fascists, Boers who made the Great Trek, talking heads, and glaciers questions, but there were several others. The issue with these question, simply put, revolved around the fact that people ended up getting punished for knowing things and/or had to (as has been noted in the MO thread) read the mind of the questions writer. Perhaps we should split off some of the MO discussion here, but I think this is a significant enough issue that it's worthy of its own thread.

I think asking questions on obscure talking heads from mythology or the League of Blood Incident rules, and I don't dispute the many positive uses of common link tossups. But when the answers become so nuanced or fail to take into account the reasonably complete knowledge people will use to buzz in on various clues, theses tossups become the most maddening type, because only people who know things about the League of Blood or whatever will be buzzing in off those clues, and they're going to end up getting shafted. If the entire premise of the Roosevelts question is that "at the beginning, it mentioned two people were involved, so I feel fine negging you for only knowing about one assassination," that's just lame. The entire purpose of pyramidal clues is because people who don't know the early ones will know the later ones. In particular, I bet tons of people don't even know people tried to kill FDR, but I think it's dumb to punish them for knowing (maybe tremendous amounts) about Teddy's shooting. For this kind of a question, the answer is simple: Do not pick this kind of thing as an answer. It's gimmicky, it's going to lead people to buzz in off a clue they probably know really well, and just get screwed for the very fact that they knew more about that clue than their opponents. Sure, they didn't know as much about it as someone who would have recognized both the FDR and TR clues, but again, that person has probably already powered the question and the point is moot. Stop writing tossups where the answer is "assassinating a president whose last name is Roosevelt, prompt on but don't accept just one of the Roosevelts" or "literary characters who are lying face down in the mud." Just don't do it. There really isn't even a reason to do that kind of a thing with these questions (FDR/TR are going to be easy answer selection, at least before you did that, and making the answer "face down in the mud" actually made that question light years harder than tossup on A Very Old Man with Enormous Wings or whatever the other giveaways were).

I don't know very much about glaciers so I can't really comment on that kind of question, but I would really again just caution against doing this. It seems like you're going to invite prompt-negs, and people really have no way of knowing that their buzzes are ill-advised, unless there's some kind of quaint preface saying "Note: your answer must be a specific subset of another kind of thing," and I don't think we want to do stuff like that. At the very least I'd think you'd create a kind of problematic slippery slope. For example, a tossup on Neutron Stars in our game against Illinois A saw Mike Sorice buzz in really early with Magnetars, which he obviously knows very many things about. According to Mike, everything that had been said applied to Magnetars (and presumably, if I followed the logic, to pulsars and whatever other weird type of neutron starts there are, I have no idea). So, he really was correct (and fortunately got the points), but at what point would you draw the line with that? The point I'm trying to make is that tossups where you'd either need to accept a very significant range of things or punish someone for their knowledge are bad, and should be avoided if at all possible.

Finally, on questions like the Japanese Fascists question - I don't want to say that I don't think these questions should be written, but I think it's hard for someone (even someone with as in-depth history knowledge as Charles, who I think wrote it) to really write the tossup in a way that points to a single, clearly-defined answer. I buzzed in early in that questions "groups who attempted to overthrow the Japanese government," or something along those lines, which from what I can tell is not only perfectly correct at the time of my buzz, but also even more specifically correct than "Japanese Fascists" as there were presumably some fascists who didn't try to overthrow the government. The second issue is that, assuming that I'm correctly concluding clue number one refers to Okawa Shumei, the factuality of the answer is at the very least a point of very inconclusive opinion. I'll go ahead and put forth that Okawa wasn't a fascist (and isn't thought of as such by most scholarship), and a few minutes of googling seems to support that conclusion. Without digressing into boring Asian history lecture mode, the organization Okawa belonged too which I think that clue is referencing was co-founded by someone Okawa is often cited as way less pro-fascist than, and who himself is centrally understood to escape definition as simply "fascist" as tons of the ideas promoted by that group were extraordinarily Marxist. Of course, this whole thing might not even matter, because the clue basically says "Okawa belonged to an organization of this type" but says nothing about the organization, so who knows if it's even the one I'm thinking of or if it's some local library group Okawa might have enjoyed dropping in on? The point of this long, detailed analysis is that common link tossups that involve any kind of subjective conclusion, refer to things that can be universally classified as falling under multiple specific categories, are at times non-specific about the relationship of the clue to the link, or refer to things that have multiple categories should really be avoided (and tossups that do all of those things, even more so). Find other answers for these tossups ("Japanese coups," for example, seems an unambiguous way to ask basically an identical question), or just put the clues in bonuses.

[Skepticism and Animal Feed]

One thing that concerns me is the way that common-link social science tossups are heading. See this example:

The “geography” and “future” of this concept were both the topics of two books by Benjamin Cohen. In his 1967 essay “On the Concept of Influence,” Talcott Parsons referred to it as a “symbolic embodiment,” and called for a sociology of this concept in the 1971 paper “Higher Education as a Theoretical Focus.” Karl Polanyi distinguished between the “all-purpose” and the “special-purpose” varieties of it, while Max Weber referred to it as “the most abstract and impersonal element... in human life.” Karl Marx used the term “fraternization of impossibilities” to refer to the way in which it “exchange every property for every other.” For 10 points, identify this substance which along with “employment,” and “interest,” is the subject of the best-known work of John Maynard Keynes.

This tossup, and many other recent tossups of its kind, are overwhelmingly made up of two types of clues:

(1) titles of works
(2) direct quotes from works

Only the Parsons and Polanyi clues are really substantive social science. I would prefer more clues that tell me what arguments people made about a given concept, and fewer clues about what the titles of works. I also don't think that direct quotes are necessarily all that great; they can be confusing outside of the broader context of the words that surround them.

[grapesmoker]

Only the Parsons and Polanyi clues are really substantive social science. I would prefer more clues that tell me what arguments people made about a given concept, and fewer clues about what the titles of works. I also don't think that direct quotes are necessarily all that great; they can be confusing outside of the broader context of the words that surround them.

The Marx clue is completely legitimate social science, and useful if you take the time to process it; Jonathan Magin in fact told me explicitly that this was the clue on which he buzzed. There also used to be a clue about Georg Simmel in there, but it was edited out due to the overlap with the Simmel tossup.

I think Chris hits on some good points, but I just want to reiterate something I've said before: you need to have titles in order to make sure your clues are unambiguous. It's fine to have arguments as clues as well, but they can often be misleading without some titles to accompany them.

[theMoMA]

I'd like to interject a somewhat opposing view to some of this. A lot of this seems to be predicated on the idea that what is being punished is "knowledge." Most of the time, when we're talking about answers that aren't actually ill-conceived, or that aren't imperfectly underlined, what is actually being punished is (correctly) a lack thereof.

If someone buzzes in with "glaciers" on a tossup whose clues are entirely about alpine glaciers, gets prompted, and doesn't know what alpine glaciers are, guess what...they do not deserve points! Similarly, if someone buzzes in with "Boers" but doesn't know that a very important subset of those Boers, a group to which all of the clues apply, is the Voortrekkers, they similarly deserve no points. If you neg on questions like this, the solution is not to blame the way quizbowl questions are written. The solution is to stop buzzing when you don't know the answer!

Look, we can all agree that tossups on "trying to assassinate a Roosevelt" or "lying face-down in the mud" are ill-conceived. Those were flawed questions, and I apologize for them. That's why we have tournament discussion threads and try to become better writers. We can also all agree that when Mike Sorice says "magnetars" on a clue about magnetic whiskers on the surfaces of neutron stars, he should not be negged, because magnetars are in fact neutron stars with large magnetic fields that often display said whiskers. That's why we have protests, because people make mistakes, people forget to underline things, shit happens. But to say that things like alpine glaciers or Voortrekkers make for flawed tossup answers because some people might not know enough about them to distinguish them from their general classifications? This seems to go against what I view to be a basic principle of quizbowl; namely, that if a player does not know the answer, a player should not get the tossup.

[DumbJaques]

That's why we have protests, because people make mistakes, people forget to underline things, shit happens.

Well, yeah, but since protests are only resolved if they matter, it's still a very negative and aggravating thing if there are multiple examples of these kind of questions in the set. If we can state categorically that these questions are ill-conceived, it's seems ludicrous to me to say that it's less of a problem to write them because you can protest. 90% of the time these don't even get resolved, and even when they do, there are tangibly negative effects on a tournament that has to deal with that kind of stuff. At the very least, I think that on tossups like these (if people make the mistake of choosing to write them), readers should at least err on the side of acceptance and let the protest go the other way, if it matters. If that were the case, I'd see less of a problem with your point (although the whole issue with "if these tossups are ill-conceived, there just isn't a reason to write them" still applies). If I end up having any common link tossups that seem marginally iffy to me in the TIT set that I just really can't bear to part with, I know I'll include several encouragements for moderator leniency. I'd say that the entire argument that "you can protest if it's not accepted, so in some way the point is less valid" is sort of a fallacy and something that I've seen used to defend bad quizbowl before; while you'd have to be insane to accuse Andrew Hart of defending bad quizbowl, we oughtn't ignore a valid conclusion about why doing something has negative impacts on the game just because there are occasionally paths for recourse if it really ends up making a difference in the final score.

Your other point I find much more valid, but I do still disagree. I don't know anything about glaciers and as I said, I don't really feel able to discuss that question. Similarly, I suppose an answer of "voortrekkers" is just fine because that group has a specific name and identity and is something you should reasonably be able to get off of a prompt if you know what you're talking about, so I won't dispute that. I don't think your assertion that people should "stop buzzing when they don't know the answer" is very valid at all, since this is really more of a case of "people needing to read the mind of the question writer" in the instances we're discussing. Even if we set aside the several factual issues with the Japanese Fascists question, for example, it's still extremely non-specific and neg-baiting. We aren't talking about people who hear a word that sounds Afrikaans and go "time to buzz in with Boers," we're talking about people who have very good knowledge of something (perhaps more than the question writer/editor) and either don't buzz because the question is problematic, or worse, buzz in with an entirely correct guess and end up getting negged because they didn't really know what variety of answer the question wants them to give. I don't think anyone here is arguing that we should award tossups when people don't know what they're talking about, but I really don't think that ends up being the case in most of the situations identified. It's basically like Jerry's point about titles being necessary to create a firm anchor point for more substantive common link clues - situations where people can have an abundantly satisfactory knowledge of a clue to answer a tossup on a work or an author should not end up sitting there confused or negging while a less-knowledgeable player picks it up later because of the nature of the common link tossup. It's not as if I'm proposing some radical thing. Just be lenient on prompt instructions and acceptances for common links, err on the side of protests the other way, and stop writing these aggravating tossups on "Chinese Muslims," "face down in the mud," "japanese fascists," or even "talking/decapitated/whatever heads," particularly when you can use basically identical clues and create far less ambiguous answer lines.

[Mechanical Beasts]

The issue with having extremely nuanced common link tossups, moreover, is that it turns into a complicated mind-game. To properly play such a common link tossup, I would have to think of the most specific category that explains the previous items mentioned, and no more, or else the prompts may never end. So after the first clue, I should really be buzzing with, like, "people who were born on DATE and who lived LIFETIME years and who wrote books entitled X, Y, Z, and..." because there may indeed be multiple of them. After the second, maybe I have someone new who wrote different books and had a different birthday, but at least they're both from Ecuador and both voted for Dukakis, so my answer turns into "people who were born on DATE and are from ECUADOR and voted for DUKAKIS," and eventually, after enough people are added to stretch out the set, so to speak, I get "political dissidents whose maiden names rhyme with dork" or something.

This has never happened and probably will never, but there's an important difference between acceptable common links and borderline ones. Is the category meaningful and can it be defined in an academically meaningful and interesting way besides through listing examples? People have studied Chinese Muslims and fascism (and probably fascists!) in Japan, so those aren't so bad, if the factual issues with the latter may have been a problem. But unless I missed a lot of lit criticism about lying face-down in the mud, I don't know what the big deal is. I just don't want "Japanese fascists" to be prompted, three years from now, until I give "male Japanese fascists who never played baseball," and I think that even these, while probably fair in theory, feel a little lame for aesthetic reasons.

I thought the money tossup was cool, if a little forced. I had no objections.

[Irreligion in Bangladesh]

It seems to me that common links such as "Japanese fascists" or themes like "the assassinations of the Roosevelts" may make decent skeletons for bonuses.

[DumbJaques]

People have studied Chinese Muslims and fascism (and probably fascists!) in Japan, so those aren't so bad, if the factual issues with the latter may have been a problem.

Notably, people who studied those things in an in-depth probably didn't write those two tossups, as each was rife with factual errors and had tremendously flawed premises. There is not a correlation between what is studied in a somewhat categorical way (or has been studied that way once by someone) and what you can make the answer to a quizbowl question. I suppose one could write tossups on "Japanese fascists" or "Chinese Muslims" that are not bad (particularly the latter), but honestly I think you'd run into so much cross-identity and acceptable alternatives that it would be an unnecessarily bad idea. Make the answer something different, use the same clues, and avoid writing tossups to which things like "nationalism" or numerous other things would be just as (if not more) of an acceptable answer. Chinese Muslims in Harvard's ACF packet, for what it's worth, also had serious factual errors. I mean it's possible to write a Chinese Muslim tossup, but evidently it's difficult enough to grasp the nuances that a team composed entirely of very good writers managed to author a question that attempted to assert that the the Hui ethnic group was completely synonymous with "Chinese Muslims" (among other issues). Instead, write tossups on actual things from Chinese history. These tournaments have very difficult answer selections - I don't see the argument for writing these tossups at all.

[Mechanical Beasts]

I think you're absolutely right in principle, and it's possible that the Chinese Muslims tossup is worse than I know; I had no hand in its making. My argument is this: common links such as the ones we're currently really concerned with aren't just bad quizbowl, they're also nonsensical, can't be fun, aren't studyable, etc. There's a reason that the 1912 birthdays Wikipedia article--surely one with fewer errors than most Wiki articles--isn't a good question writing resource: no one cares to write a tossup on 1912 birthdays, not just because it's almanac knowledge, but because it's not meaningful. Similarly, we don't write science tossups on dilithium except one year's SM that had like 1/1 science fiction or something.

I should try to be concise. Common link tossups should be common links on things of actual import or interest, a reasonable litmus test being an existence of a clue about the set of things involved, not just the elements. Failing that, they should be pretty unambiguously easy to figure out the answer line. (That is to say: if people have really studied "alkynes that you can drink more than an ounce of before dying" then a common link is okay. If not, then make the answer line, like, "alkynes," because the fact that you happened to give as your four examples of alkynes four that are fairly potable doesn't mean that anyone's going to know what you're going for and shouldn't be denied points for it.)

I do not support drinking any alkynes at all, for the record.

[Mike Bentley]

Maybe I'll expound on this later, but I'm starting to buy more into the argument Chris Frankel made in the chat a few months ago that common link tossups play differently than regular tossups and that change maybe isn't for the better. I think perhaps there is an (understanable) desire for people to write creative and funny tossups, and that common link is the best way for them to go about them. But I think these efforts might be better served by creating bonuses on "Umbrellas" or "Money" rather than tossups in at least some cases.

I very much liked MO, but in my opinion there were probably about 50% too many common link tossups, especially in areas like Social Science and Philosophy. Common link tossups are okay in moderation, but this tournament and others recently have taken it to something of a ridiculous extreme.

[naturalistic phallacy]

I very much liked MO, but in my opinion there were probably about 50% too many common link tossups, especially in areas like Social Science and Philosophy. Common link tossups are okay in moderation, but this tournament and others recently have taken it to something of a ridiculous extreme.

I would agree with you in that common links in Social Science and Philosophy can often go awry, especially when they are on terms or ideas. Those type of common link questions often devolve into "Can you fill in the blank?" and reward those with list and title knowledge far more often than those with knowledge of a works contents. However, I do feel that there is a place for the common link question in Social Science or Philosophy if the answer only concerns two (maybe three at the most) works. Writing tossups on works that share a common title doesn't have to be a bad thing; if one limits the tossup to a couple things, there is space for in-depth clues and true pyramidality.

[Kyle]

Chinese Muslims in Harvard's ACF packet, for what it's worth, also had serious factual errors. I mean it's possible to write a Chinese Muslim tossup, but evidently it's difficult enough to grasp the nuances that a team composed entirely of very good writers managed to author a question that attempted to assert that the the Hui ethnic group was completely synonymous with "Chinese Muslims" (among other issues). Instead, write tossups on actual things from Chinese history. These tournaments have very difficult answer selections - I don't see the argument for writing these tossups at all.

I didn't write a tossup on Chinese Muslims. I wrote a tossup on the members of the Hui ethnicity. You'll have to take up the question of whether Chinese Muslims should have been an acceptable answer (in my opinion it shouldn't have) with the editors who added it. Every person mentioned in that question, from Ma Zhongying down to Zheng He, was a Hui Muslim. Ma Zhongying, who was the lead-in, was important historically precisely because he was a Hui Muslim and not any other kind of Chinese Muslim (he put down a rebellion of Uighurs who unsuccessfully tried to appeal to him as a co-religionist).

[Mechanical Beasts]

It's a good thing that I didn't try to defend Kyle's tossup out of school pride when I've only ever seen the edited form. And also when I recognize one name or term from that explanation, and that is Uighurs and that is, I believe, because of Age of Empires II.

[Skepticism and Animal Feed]

Kyle did in fact write a Hui ethnicity tossup. Because I thought that "Hui ethnicity" was too hard of an answer choice, I changed the answer line to "Chinese Muslims" before submitting the question to Kwartler. My thinking was that for people who knew who the Hui were, it would be a Hui tossup, and for everyone else (the majority of quizbowlers, I bet) it would at least be gettable at the end.

I would prefer, however, that this thread be a "big picture" level discussion of grand flaws in the way that common-link tossups are written these days. I don't want this thread to degenerate into a discussion about details of specific questions.

[Skepticism and Animal Feed]

Kyle did in fact write a Hui ethnicity tossup. Because I thought that "Hui ethnicity" was too hard of an answer choice, I changed the answer line to "Chinese Muslims" before submitting the question to Kwartler. My thinking was that for people who knew who the Hui were, it would be a Hui tossup, and for everyone else (the majority of quizbowlers, I bet) it would at least be gettable at the end.

I would prefer, however, that this thread be a "big picture" level discussion of grand flaws in the way that common-link tossups are written these days. I don't want this thread to degenerate into a discussion about details of specific questions.

My inspiration for doing this, BTW, came from something that Kyle did when writing 2007 EFT. Somebody wanted to write an Italo Calvino TU, but Kyle thought that it was too hard, so he ended up writing a literature tossup on "Italy", the early clues of which were entirely about Italo Calvino. For people who knew who Calvino was, it was a tossup on him, but this indulgence of a select group of high school players did not ruin it for everyone else, who was able to pick it up in the end.

I think this is a perfectly good strategy, even if Kyle has now abandoned it.

[Matt Weiner]

My inspiration for doing this, BTW, came from something that Kyle did when writing 2007 EFT. Somebody wanted to write an Italo Calvino TU, but Kyle thought that it was too hard, so he ended up writing a literature tossup on "Italy", the early clues of which were entirely about Italo Calvino. For people who knew who Calvino was, it was a tossup on him, but this indulgence of a select group of high school players did not ruin it for everyone else, who was able to pick it up in the end.

I think this is a perfectly good strategy, even if Kyle has now abandoned it.

I think this is where common link tossups have a very valuable role. These questions were originally promoted within ACF around 2004-2005 as a way to make things easier--for example, instead of trying to force that tossup on the Aswin Twins into a tournament where most teams won't know it, why not use the same clues to start a tossup on "gods of medicine" that can go on to mention Asclepius and get converted more? I wouldn't want to see that extremely useful type of question be discouraged in low-to-regular difficulty events. At the harder end, maybe it would make more sense to just write on the Aswin Twins. If there is nothing in the question that is a plausible tossup answer even at a difficult tournament, as in "Japanese facsists", then perhaps that tossup shouldn't be written, on the grounds that, if people don't know the actual clues, they will get it only on the giveaway, through fraud, or not at all. A question which does not produce early buzzes for more knowledgeable teams is by definition a bad question.

[Mechanical Beasts]

When you're running common links that way, they're terrific. They are, essentially, author tossups, because author tossups (save ones that include biography clues, which are objectively bad) are isomorphic to common link tossups on "works by x."

[DumbJaques]

I didn't write a tossup on Chinese Muslims. I wrote a tossup on the members of the Hui ethnicity. You'll have to take up the question of whether Chinese Muslims should have been an acceptable answer (in my opinion it shouldn't have) with the editors who added it. Every person mentioned in that question, from Ma Zhongying down to Zheng He, was a Hui Muslim. Ma Zhongying, who was the lead-in, was important historically precisely because he was a Hui Muslim and not any other kind of Chinese Muslim (he put down a rebellion of Uighurs who unsuccessfully tried to appeal to him as a co-religionist).

Kyle did in fact write a Hui ethnicity tossup. Because I thought that "Hui ethnicity" was too hard of an answer choice, I changed the answer line to "Chinese Muslims" before submitting the question to Kwartler. My thinking was that for people who knew who the Hui were, it would be a Hui tossup, and for everyone else (the majority of quizbowlers, I bet) it would at least be gettable at the end.

Ah, this makes much more sense now! It's unfortunate that happened, as keeping the question the same and making the answer "Chinese Muslims" is, as Kyle points out, factually incorrect. Changing the clues a bit in the question to also take about Uighurs or others would have made that jive better, but in the vein of Bruce's call for the big picture, I still wonder what happens when someone knows, say, a hypothetical second clue because of deep knowledge of Uighur culture and then ends up getting a prompt and not realizing it wants a religious group, resulting in a neg. I think Matt hits the nail on the head when he says that having tossups on more difficult things at ACF Nationals and the like is just fine when you're faced with a possibly iffy common link situation.

In the vein of ignoring Bruce's noble call for generalities, there was a factual error in the Chinese Muslims tossup, and I bring it up because I think it's relevant to one of the issues I identified with common link tossups. The clue about tracing origins back to a people who showed up in the wake of the An Shi revolt *might* refer to the Hui, as sources are kind of sketchy about that and I mean, it's probably undeniable that during the Tang dynasty, people came to China and later on children of children of those people ended up birthing some Huis or whatever. However, that clue makes it sound an awful lot like it means the Huihe people, who are famously linked with the An Shi revolt and unambiguously are part of the ancestry of the Uighurs. In fact I buzzed in off that clue and became utterly confused knowing that the first clues weren't uighurs at all, and said something amusingly incorrect. My point is that a lot of times in these common link tossups that try to get creative in such a fashion, at least one clue usually ends up ambiguous at best and outright wrong at worst. I think this further points to a solution of making sure it's a damned specific thing (word, title, country, person's name, whatever) when choosing the answer for a common link tossup.

[Skepticism and Animal Feed]

Ah, this makes much more sense now! It's unfortunate that happened, as keeping the question the same and making the answer "Chinese Muslims" is, as Kyle points out, factually incorrect.

I don't see how it makes the answer incorrect. Assuming that every single clue in there applies to Hui who are also Muslims (as Kyle says is the case), then both "Hui" and "Chinese Muslims" are correct, as the former falls entirely within the latter. This would make it essentially equivalent to the HFT "Italy" tossup, whose underlying concept has been praised by Matt Weiner himself.

ACF Nationals 2008 was very much a rejection of the "hard questions on easy things" school. I admit that one of the reasons I changed the answer line was because I thought "hard questions on easy things" was still ascendant among the ACF community. Perhaps if I had known beforehand that this was no longer the case, I would have left Hui in. But probably not, as it's extraordinarily difficult when you look at the very underdeveloped state of the Asian history canon.

[theMoMA]

The problem I have with this discussion is that we're lumping plenty of well-written tossups in with ones that have serious issues. The tossups on assassinating Roosevelts and Japanese fascists were probably the two most ill-advised tossup answers in the tournament. They are nothing like the tossups on alpine glaciers and Voortrekkers, which were both well-written and pointed to one answer; if you negged those with glaciers or Boers and didn't know more, then indeed, you shouldn't have buzzed. The tossup on decapitated heads was a fine tossup, and the answer line specified that moderators should accept anything that specified that the answer was heads that were no longer attached to their bodies. If moderators didn't follow this, I'm sorry, but it was still a good question with the right answer line.

I'm not really sure what the point of your discussion on protests is. I'm admitting that writers make mistakes, either in underlining or in writing on ill-advised answers. If you look at the discussion thread, there are maybe five or six fairly egregious examples of this throughout the tournament (which, I would like to mention, contained 357 tossups). If you really want to scold us for screwing up one out of every 70 or so questions, thus giving you -5 instead of 10 or 15 points in a situation where it didn't matter to the game, by all means do that. But realize that while this is certainly a problem, there is a means of fixing it in the rules of the game. I know it's aggravating that writers sometimes make mistakes that cause you to lose points you think you deserve, but honestly, this happens at every tournament ever, and it's not going to stop any time soon.

Also, the idea of erring on the side of correctness for protests seems wrongheaded. There's a reason that every major quizbowl entity specifies in its rules to count the questionable answer wrong first, and possibly to overrule it if the protest matters: this is the fairest way to do things, because it eliminates unevenness in moderation as much as possible. What problem does it solve, exactly, to do this the other way around? Getting another 15 or 20 tossup points in a game where those points have no bearing on the outcome seems to me not the kind of thing worth crusading for.

[grapesmoker]

The problem I have with this discussion is that we're lumping plenty of well-written tossups in with ones that have serious issues. The tossups on assassinating Roosevelts and Japanese fascists were probably the two most ill-advised tossup answers in the tournament. They are nothing like the tossups on alpine glaciers and Voortrekkers, which were both well-written and pointed to one answer; if you negged those with glaciers or Boers and didn't know more, then indeed, you shouldn't have buzzed.

I agree with this...

The tossup on decapitated heads was a fine tossup, and the answer line specified that moderators should accept anything that specified that the answer was heads that were no longer attached to their bodies.

...but not with this. Seth has already explained the problem of this kind of reasoning. As I've said before, there is no possible reason I should need to guess that the answer must be "severed" heads, simply because, by virtue of the way the question is written, you are referring to heads that are already obviously detached from the body. I don't think anyone is seriously going to contend that a decent player isn't going to know that Mimir's head or Bran's head were severed from their possessors bodies, so I don't see what requiring one to say this adds to the question, other than to result in negs when the person giving the answer can't read the writer's mind.

[DumbJaques]

I don't see how it makes the answer incorrect. Assuming that every single clue in there applies to Hui who are also Muslims (as Kyle says is the case), then both "Hui" and "Chinese Muslims" are correct, as the former falls entirely within the latter. This would make it essentially equivalent to the HFT "Italy" tossup, whose underlying concept has been praised by Matt Weiner himself.

It makes a number of the lines the question incorrect. The first community of Chinese Muslims is not the same as the first community of Hui (I mean, almost certainly not, I'm not going to do in-depth research on this); "Muslims" do not enjoy autonomous rule from Ningxia; notably, "Chinese Muslims" are not distinguished from Uighurs or Kyrgyz, as most of those people are in fact Muslims in China. The whole basic assumption that Hui and Chinese Muslim is the same is what's wrong with the question; it's not at all equivalent to the "Italy" thing. Rather, it's analogous to writing a Calvino tossup and accepting "Italian authors."

he problem I have with this discussion is that we're lumping plenty of well-written tossups in with ones that have serious issues. The tossups on assassinating Roosevelts and Japanese fascists were probably the two most ill-advised tossup answers in the tournament. They are nothing like the tossups on alpine glaciers and Voortrekkers, which were both well-written and pointed to one answer; if you negged those with glaciers or Boers and didn't know more, then indeed, you shouldn't have buzzed. The tossup on decapitated heads was a fine tossup, and the answer line specified that moderators should accept anything that specified that the answer was heads that were no longer attached to their bodies. If moderators didn't follow this, I'm sorry, but it was still a good question with the right answer line.

Yeah, I also agree with the first part, as I said earlier. The Voortrekkers and Alpine glaciers are very specific things that happen to be subsets of other things, and they're not what I mean. I'm with you there. Like Jerry, I also think the whole decapitated thing falls under needless underlying, which is why one of my original points was that when writing common link (well-conceived or otherwise), being lenient with what you accept is important.

I'm not really sure what the point of your discussion on protests is. I'm admitting that writers make mistakes, either in underlining or in writing on ill-advised answers. If you look at the discussion thread, there are maybe five or six fairly egregious examples of this throughout the tournament (which, I would like to mention, contained 357 tossups). If you really want to scold us for screwing up one out of every 70 or so questions, thus giving you -5 instead of 10 or 15 points in a situation where it didn't matter to the game, by all means do that. But realize that while this is certainly a problem, there is a means of fixing it in the rules of the game. I know it's aggravating that writers sometimes make mistakes that cause you to lose points you think you deserve, but honestly, this happens at every tournament ever, and it's not going to stop any time soon.

You seem to be getting extraordinarily defensive about this dude. I started a separate thread for this precisely because I didn't want to conflate a discussion about the present state of common link tossups and the 2008 Minnesota Open. The things are utterly distinct and in fact the last 6 posts or so were centered on discussing a tossup from a different event last year. I'm not sure why you feel the need to point out that your egregious:good ratio is 5:357 - the tournament was phenomenal, I and everyone I've spoken to and seen post loved it, and I wouldn't even put the ratio anywhere near that high. In fact I just made a post in the MO discussion thread criticizing someone for conflating some kind of commentary on the philosophy of pronunciation guides with your set. There's not really a need for you or anyone else to feel attacked or respond with references to the holistic nature of MO (again, noted excellent tournament), particularly when what we're attacking is the perceived current community standard of common links. I mean, even if we conclude that they are the devil's spawn and must be banned from all qb, this is the first real thread we've had that's bringing negative opinions out, so nobody could or should be blamed for writing them. Nothing about this thread is scolding you, which is why it bothers me that it seems impossible for me to state my opinion on the pitfalls of these tossups without you experiencing it as criticism of you for not giving me 15 points or whatever. That's just not the issue and even if it was, it wouldn't change the more important implications of the discussion.

Sometimes writers make mistakes, and sometimes you don't get points you deserve. I continue to not understand why stating this fact in any way invalidates the argument that "people should really put a lot of thought into common link tossups that move into the territory of ambiguity, because it really can invite hoses or factual issues, or the issue of needing to read someone's mind." I'm not talking about writers overlooking things, I'm talking about the decisions to write certain tossups. I mean, if we're judging something as ill-conceived, shouldn't we avoid whether or not there's a way to sometimes protest it, but usually not?

Also, the idea of erring on the side of correctness for protests seems wrongheaded. There's a reason that every major quizbowl entity specifies in its rules to count the questionable answer wrong first, and possibly to overrule it if the protest matters: this is the fairest way to do things, because it eliminates unevenness in moderation as much as possible. What problem does it solve, exactly, to do this the other way around? Getting another 15 or 20 tossup points in a game where those points have no bearing on the outcome seems to me not the kind of thing worth crusading for.

I think I misstated my opinion on this. I don't think moderators should keep it in their minds to be more lenient on common link, as that causes exactly the kind of problem you're identifying (and, also, would affect only the moderators who read this thread). I'm saying that when someone is writing an answer line, putting in underlines, and noting prompts and also accepts, they as a writer should err on the side of leniency so as to prevent punishing knowledge.

[Captain Sinico]

Most common-link questions are bad questions by the standards we hold other questions to because their clues do not unambiguously refer to their answers, even to any degree of approximation. It's long been the case that if one can't write a question wherein at least very nearly every clue at least very nearly unambiguously refers to its answer, then one is writing a bad question; I don't see why the same standard shouldn't be applied to common-link questions.
Therefore, what I'd like to see is a litmus test for common link questions along the lines of the above. A writer ought to ask themself: "Do my question's clues unambiguously refer to my question's answer?" If the answer is "No," then that question ought to be fixed or scraped. However, this is really nothing new; it is rather a call to restore long-held, reasonable standards of good quizbowl that this somewhat novel type of question have, until now, been exempted from by otherwise good writers and editors for reasons unknown.
Let's examine some examples from MO. The way I see it, the tossup on "attempting to assassinate a President/Presidential candidate... named Roosevelt" fails this test. No amount of knowledge of just one of the clues of that tossup is going to let a player get that question; one has either to know more than one of the clues, or to make a leap of reasoning not entirely justified by the known facts. The "severed heads of mythology" question, on the other hand, seems to succeed by this if the answer line is generous enough (so, like, if I go in with "You're talking about the head of Orpheus..." it ought to be accepted outright; I've said the relevant part of the answer.)
That brings me to my last point. Frequently, these questions are written in ways that don't promote the understanding of hearers and they are prone to overunderlining. The parenthetical comment above suffices, I think, to communicate what I wish to about the latter. As for the former, I think that, even for questions that fail the above litmus test, i.e. that don't/can't have clues unambiguously referring to their answers, writing in the proper way can at least let people realize that they don't know the answer and should guess something more general or not buzz.
While I don't mean to pick on any specific question there, I think it's instructive to examine the same examples as above. The "assassinating a Roosevelt" question several times uses the phrase "this act" as the stand-in to its answer. However, normal usage would divide the answer into two separate and distinct acts, so saying "this act" is confusing; it implies that a common-sense single act is the answer. On the other hand, the "talking heads" tossup scrupulously uses stand-in phrases like "One of these," "The first," "a third example of these," etc. to make it unambiguously clear at every point in the question that the answer is a group of several distinct items that fall under some common rubric. That, I think, is the only way to go.

MaS

PS: I'd also meant to add in this post that I think many common link questions wind up functioning as stealth ways to re-introduce "title bowl," which I don't think is a good thing. I mean here questions where the only credible way to get them is to fill in blanks in a series of titles; at base, the knowledge being tested there is binary author-title knowledge, which I think is poor. Questions of this kind also tend to be transparent, though not without exception. For example, the MO "money" tossup wasn't transparent to me, though it is a very good exemplar of what I'm talking about above. On the other hand, the "the Author" question was both very "title bowl" and also fairly transparent (at least to me.)

[Captain Sinico]

Also, I want to note along with Andrew (or, perhaps, a forteriroi) that some of the questions under discussion here shouldn't be. "Alpine glaciers" wasn't a common link question; rather, it was a question with what some consider a too-specific answer for its clues, which is an issue related to some of the issues with common link questions. The "Chinese muslims" question was apparently meant as one on the Hui that had its answer line changed to make it wrong or misleading (with good intentions, given, but poor results.) As much as I love to hear complaints about any bad question over the past several years (which is to say "not in the least") discussion of these doesn't belong in this thread.
I also want to disagree with Andrew about the "Roosevelt" question. I agree that it was a poor choice; however, the point is that it was a poor choice precisely because it was a common link question. That's precisely the type of question we ought to be talking about here.

MaS

[theMoMA]

My only purpose in posting in this thread is to try to separate the discussion of good tossups, good tossups with bad answer lines, and bad tossups. I think the original post and some of the secondary posts seem to confuse those three, and I'm only trying to show that the three are different.

Also, you are somewhat confusing what I said about protests, because I don't think I made this very clear in the first place. My point is that protests are the player's recourse if there are good tossups with bad answer lines (such as that tossup on neutron stars that didn't accept magnetars, or that tossup on severed heads that should have, but didn't, accept just "heads"). If the writers write confusing or bad tossups, like the ones on assassinating Roosevelts or Japanese fascists, the only real recourse is for us to stop doing that.