On Exactitude in Science Writing

This forum is for discussing tournament formats, question styles, strategy, and such.

On Exactitude in Science Writing

Postby Amizda Calyx » Fri Nov 25, 2016 6:48 pm

This is somewhat prompted by recent discussions, but I started writing this a couple months ago when I first came up with the title for this thread. So here are some observations and tips regarding science writing.

New writers and people writing for a science subject with which they are unfamiliar, especially those who don't have a science background, frequently fall into the trap of using inappropriate layperson terms in their paraphrasing of scientific concepts. These manifest in two ways: (A) Using a pronoun that appears to be descriptive of the answerline but in fact has either a separate meaning in science, or is just inappropriate for that topic. (B) Using verbs that are inaccurately applied to a concept and/or change the meaning of the clue. Don't simplify a clue without verifying that the paraphrasing you use represents what you're trying to say in a precise and lossless manner. This becomes especially important in math and CS, where layperson terms like "property" or "condition"* or "construct" or "proved" or "scenario"** can have completely different meanings or be non-applicable. Mathematical statements are precise for a reason!

Several different writing problems can arise for people with science backgrounds, with the most pervasive one being complacency. This happens when writers feel so comfortable with a topic that they trust their memory and neglect to fact-check clues. Although it substantially speeds up question-writing, depending on where their knowledge came from it can have negative consequences. Here are a couple other traps:
  • Personal class notes. These are especially pernicious because they give the illusion of accuracy and unambiguousness. Unfortunately, even at the graduate and professional level coursework is not thorough enough to justify relying on just class notes or even lecture slides. A good example of this is using a symptom or sign that your med school courses really hammer home as being an indication of a particular disease. You might feel that this means it's specific, but a lot of the time a less common or linked disorder can also manifest with that phenotype. A less widespread issue occurs when your notes are just wrong. It could be from inaccurate transcription, a blatant error from the professor, or outdated information. Here are some examples from professors in my graduate courses:
    • the "lysine zipper" present in c-fos.
    • KLH is derived from "some sponge".
    • an excited claim that ubiquitinated Notch is uniquely translocated to the nucleolus – this nearly made it into a NASAT tossup I wrote but was discarded when further research revealed that this was at best a mischaracterization of the process.
    • Furosemide is "calcium-wasting" – while this is certainly true, it is not the only ion furosemide "wastes", and it's actually way better-known for wasting potassium.
  • Trusting packets. Oh god please do not do this. Quizbowl is rife with errors and ambiguity, especially in the sciences. Just look at the selection of questions pointed out here that will stay unchanged in the database. A quick look through Quinterest at recent college-level bio tossups reveals things like:
    • ACF regionals 2014 wrote:"These cells present the major histocompatibility complex."
      TU on T cells. *All* nucleated cells present MHC I, but T cells are way better known for having antigens presented to them.
    • NASAT 2014 wrote:"One enzyme found in these cells converts H2O and CO2 to H2CO3."
      TU on RBCs. Literally a five-second look at the carbonic anhydrase wiki page would have shown this to be non-unique.
    • Literally every clue in the ACF regionals 2016 tossup on "organ transplant rejection".
    • Penn Bowl 2015 wrote:"In the biosynthesis of this compound, Met Adenosyltransferase converts methionine to SAM, which is then converted to ACC."
      TU on ethylene. Everything except the last word in this sentence can apply to multiple things!

As a general rule, I confirm clues through multiple sources before using them, which I think is a pretty common approach. What seems to be somewhat less common is using the same phrasing you'd use to google a specific clue but without the answerline in the search terms. Doing this would very quickly reveal that the claim "furosemide wastes this ion" is ambiguous, for example.

Somewhat tangential to all of this is answerline choice: it can be really difficult for someone who isn't actually a scientist to identify topics that are actually important, let alone clues. Unlike in many other categories where secondary and tertiary literature is fine to learn from and pop culture distillations can still be at least somewhat useful for learning clues, math and science really aren't going to be represented by non-academic sources in a way that provides adequate information. They provide surface-level material on things that aren't covered in science courses, and they neglect the information that one would learn in even an introductory class on the subject. Another issue I've noticed with layperson articles and books is that they focus on the history of a subject more than the actual science of it. Authors tend to romanticize STEM things when writing for a general audience – just look at all the books on the Fibonacci sequence*** or golden ratio or HeLa cells or string theory that relate their respective focuses to exciting ideas like black holes and time travel and a cure for cancer without addressing, for example, what Einstein's field equations look like or how a derangement in a c-myc pathway might potentiate unchecked cell division. Could these provide a good introduction to a subject? Sure! But when it comes to writing questions for a higher difficulty, make sure the items discussed in a popsci work aren't inflated in importance to fit a literary narrative. Just because IFLS or NDT or Michio Kaku talk about something doesn't mean it's necessarily actively studied or widely important, or that they've even accurately characterized it.

*Eg: to describe max flow (how does condition fit for even a layperson's understanding?).
**Eg: to describe the halting problem.
***Popsci authors LOVE this because it's got cool historicity and the sequence is easily visualized and it has all these romantic connections to nature -- but it's not important in the real world to real mathematicians. And why would it be? Most mathematicians don't study the history of math, or try to connect their research to nautilus shells.
Joelle Smart
Ellensburg High School, 2006–10
University of Washington, 2010–14
Rutgers University, 2015–20??
President of Quizbowl at UW, 2012–2014
PACE
HSAPQ biology editor

It is a mistake to think you can solve any major problems just with potatoes.
User avatar
Amizda Calyx
Forums Staff: Moderator
 
Posts: 236
Joined: Tue May 18, 2010 9:46 pm
Location: Seattle, WA

Re: On Exactitude in Science Writing

Postby Marble-faced Bristle Tyrant » Fri Nov 25, 2016 7:39 pm

Good post.
Amizda Calyx wrote:Personal class notes. These are especially pernicious because they give the illusion of accuracy and unambiguousness. Unfortunately, even at the graduate and professional level coursework is not thorough enough to justify relying on just class notes or even lecture slides. A good example of this is using a symptom or sign that your med school courses really hammer home as being an indication of a particular disease. You might feel that this means it's specific, but a lot of the time a less common or linked disorder can also manifest with that phenotype.

I actually sometimes have a hard time in class because I'm too busy thinking of this very thing!
Farrah Bilimoria
Formerly of Georgia Tech and Central High School (Macon)
User avatar
Marble-faced Bristle Tyrant
Yuna
 
Posts: 846
Joined: Fri Feb 08, 2008 11:05 pm
Location: Evanston, IL

Re: On Exactitude in Science Writing

Postby Amizda Calyx » Fri Nov 25, 2016 10:46 pm

Another thing I wanted to note, and which was mentioned (with different examples) by Adam in the Terrapin general discussion: Players can gather context from clues outside of the specific content you are cluing. For example, in a tossup on yeast, describing an obscure metabolic pathway characterized by XXX researcher could reward someone for knowing that that researcher is prominent in yeasty things rather than for knowing about the specific pathway. This can be a problem if your intent was to stratify players based on their biochemistry knowledge rather than superficial name association. A more specific example would be the leadin to the 2013 Chicago Open tossup on "continuous": "A functor F from categories C to D has this property if it commutes with limits..." This is presumably trying to reward pretty deep category theory knowledge, but anyone who knows that a definition of continuous functions is that they commute with limits (a fundamental definition in courses far simpler than category theory) could reasonably first-line this. Definitely not worthy of a CO leadin!
Joelle Smart
Ellensburg High School, 2006–10
University of Washington, 2010–14
Rutgers University, 2015–20??
President of Quizbowl at UW, 2012–2014
PACE
HSAPQ biology editor

It is a mistake to think you can solve any major problems just with potatoes.
User avatar
Amizda Calyx
Forums Staff: Moderator
 
Posts: 236
Joined: Tue May 18, 2010 9:46 pm
Location: Seattle, WA

Re: On Exactitude in Science Writing

Postby adamsil » Sat Nov 26, 2016 12:08 am

Amizda Calyx wrote:Another thing I wanted to note, and which was mentioned (with different examples) by Adam in the Terrapin general discussion: Players can gather context from clues outside of the specific content you are cluing. For example, in a tossup on yeast, describing an obscure metabolic pathway characterized by XXX researcher could reward someone for knowing that that researcher is prominent in yeasty things rather than for knowing about the specific pathway. This can be a problem if your intent was to stratify players based on their biochemistry knowledge rather than superficial name association. A more specific example would be the leadin to the 2013 Chicago Open tossup on "continuous": "A functor F from categories C to D has this property if it commutes with limits..." This is presumably trying to reward pretty deep category theory knowledge, but anyone who knows that a definition of continuous functions is that they commute with limits (a fundamental definition in courses far simpler than category theory) could reasonably first-line this. Definitely not worthy of a CO leadin!

This is an important point in science especially, although not necessarily just for this context. When editing CO, I made an effort to clue publications from contemporary scientists in their fields to give a hint to people who know what's going on (e.g., the tossup on phage display mentioned George Georgiou's name early on, because he is an extremely important protein engineer and anybody who has read primary literature in that area should be able to start sniffing out what was going on, without actually being familiar with the paper in question.) This is no different to me than cluing formulas with the explicit purpose of letting players figure out the units of a physical quantity--if you've got that sort of first-hand knowledge, then it's not bad to write a tossup rewarding other people with that knowledge, provided that it doesn't make it too obvious.

In response to Joelle's original point here--I don't actually agree with all of its implications. I think anybody who has written biology questions in quizbowl is guilty of imperfectly uniquely-identifying clues. I've done it a lot. Biology is finicky, pathways interact, the primary literature is enormous, and it's often hard to find uniquely identifying clues without adding a bunch of clarifiers. I'm not endorsing writing bad questions with vague clues at any stretch (I remember writing "LITERALLY ANY AUTOIMMUNE DISORDER" in my notebook during that rejection tossup, too). But just like Ike said in the earlier CS thread, it's unhelpful to nitpick questions when there's clearly one obvious answer and a bunch of technicalities that could make other answers correct. These are quizbowl tossups, not PhD dissertations.

Let's say I wrote a hypothetical tossup on RBCs that says "Band 3 proteins are found on these cells' surfaces." Yes, they're in the kidneys too. And I'm certain that every single person who learned about Band 3 proteins did it in the context of a red blood cell, and it's unrealistic for most teams to be able to know (or even figure out) these minor details--you're trusting either the editor or writer to do a lot of reverse-clue lookup in this case. Normally, in this case, I would do the reverse-clue lookup, figure out this is non-unique, and compensate with things like "Band 3 proteins were discovered on these cells' surfaces". But I'm guessing that causes teams to play chicken if they know what Band 3 is but don't know that specific fact. You could also go with the quizbowl-ese construction "They're not basolateral kidney cells, but Band 3 proteins are...", which is jarring to hear and wastes valuable space in the tossup for actual clues. (For what it's worth, I'd probably have been fine with playing that carbonic anhydrase clue too. Yes, it's vague, and I'd have known that buzzing in. But you can bet I'd still say RBCs because that's certainly where carbonic anhydrase is most famous. Same with MHC/T cells.)

As another example: I once wrote a Regs tossup on cholesterol that led in something like "this compound is added post-translationally to the N-terminus of sonic hedgehog protein". That's a cool, unusual fact about a protein that lots of people have heard of. As I learned from a recent Googling, palmitic acid is also added post-translationally near the N-terminus. I don't want to ditch my cool clue or destroy the flow of the sentence just because it's potentially gonna cause people to buzz at Regionals with "palmitic acid".

NOTE: I'm not saying I would write these questions wrong deliberately! Obviously if you catch an imprecision you're responsible for dealing with it! But I don't think that haranguing editors for this sort of an oversight is fair unless it caused you to neg and the protest wasn't upheld, or something.

Exactitude is important, but evocative and clear cluing is more important when you're splitting hairs. If I had to read the primary literature for every single clue I put into a biology tossup to make sure that something is fundamentally unique... yikes. It's super important to be able to tell stuff that is obviously not uniquely identifying from stuff that is, but eventually you have to draw the line.
Adam Silverman
Georgia Tech 2012-2016
Northwestern 2016-
adamsil
Wakka
 
Posts: 195
Joined: Wed Feb 09, 2011 6:20 pm

Re: On Exactitude in Science Writing

Postby Amizda Calyx » Sat Nov 26, 2016 3:02 am

I absolutely endorse purposefully cluing contemporary researchers to reward people with deep knowledge -- but it's not so great when a name is dropped early on as part of an intended hard clue that players with only cursory knowledge of a topic can buzz on. It's kind of like having a leadin that uses a Latin term any classics player could easily translate and buzz off of.

adamsil wrote:Let's say I wrote a hypothetical tossup on RBCs that says "Band 3 proteins are found on these cells' surfaces." Yes, they're in the kidneys too. And I'm certain that every single person who learned about Band 3 proteins did it in the context of a red blood cell, and it's unrealistic for most teams to be able to know (or even figure out) these minor details--you're trusting either the editor or writer to do a lot of reverse-clue lookup in this case. Normally, in this case, I would do the reverse-clue lookup, figure out this is non-unique, and compensate with things like "Band 3 proteins were discovered on these cells' surfaces". But I'm guessing that causes teams to play chicken if they know what Band 3 is but don't know that specific fact. You could also go with the quizbowl-ese construction "They're not basolateral kidney cells, but Band 3 proteins are...", which is jarring to hear and wastes valuable space in the tossup for actual clues. (For what it's worth, I'd probably have been fine with playing that carbonic anhydrase clue too. Yes, it's vague, and I'd have known that buzzing in. But you can bet I'd still say RBCs because that's certainly where carbonic anhydrase is most famous. Same with MHC/T cells.)

See, my issue with this is that it automatically assumes all players are learning about certain topics under the same circumstances -- namely, that anyone who has heard of band 3 or carbonic anhydrase MUST have only been exposed to it in the context of red blood cells. But, using these exact same examples, I know this is not the case: I learned about both in my undergrad intro physiology class in the context of kidneys. Carbonic anhydrase is anywhere that regulates acid-base stuff, and under the lens of physiology it is often heavily discussed with respect to renal function (acidosis etc.). I spent four years as a physiology undergrad being deeply frustrated by information that I had learned in my classes showing up in ways that not only didn't reward my knowledge but seemed to actively penalize it. And while I have made (numerous) protests on such negs, there have only been like five instances where my team has lost by a margin that made my protests relevant -- the dozens of other instances just left me feeling bad for letting my team down, disappointed that I missed the one question a round that I had a chance on (I'm not by any means competent in other categories at higher levels), or deeply annoyed with the quality of the question. Sure, packet-studying would have been useful for at least getting "canonical" clues down, but do we really want to encourage that at the expense of real knowledge?

I'm not advocating people reverse-look up every clue exhaustively before submitting a tossup; my primary objective with this post was to raise writers' awareness of issues they might not have considered before. I also understand that writing science is a real burden for a lot of teams, and I'm not expecting any changes to occur at that layer of participation. But for people who WANT to write/edit science, and for those of us who already do, I think it's a good idea to keep these points in mind. I also don't believe that cool clues that have some ambiguity can only be discarded or hopelessly embulkified -- in my opinion there's a substantial difference between a standalone statement like "Band 3 is found on these cells' surfaces", and a sentence such as "A band 3 mutation in these cells can manifest as spherocytosis". I also don't find it too difficult or clunky to include qualifying information, eg "Palmitic acid and this molecule are added to the N-terminus of sonic hedgehog to localize its signaling" (which, btw, is a clue I would have loved to have heard after going over cholesterol's role in Shh gradient control fairly extensively in my first molecular bio course).

And, I'll add that the MHC clue is ONLY correct because of a technicality -- T cells ARE capable of presenting MHC, as is every other nucleated cell, but it's exactly the reverse of what the clue should be saying: that antigen-presenting cells present the MHC to T cells!
Joelle Smart
Ellensburg High School, 2006–10
University of Washington, 2010–14
Rutgers University, 2015–20??
President of Quizbowl at UW, 2012–2014
PACE
HSAPQ biology editor

It is a mistake to think you can solve any major problems just with potatoes.
User avatar
Amizda Calyx
Forums Staff: Moderator
 
Posts: 236
Joined: Tue May 18, 2010 9:46 pm
Location: Seattle, WA

Re: On Exactitude in Science Writing

Postby adamsil » Sat Nov 26, 2016 10:32 am

Well, your experiences trump my claims, so I'll concede these specific examples. :) More than a lot of writers, I value concise clues when writing--I find simple sentences to be the most effective in quizbowl settings. So I stylistically oppose clarifying or flowery language if it's at all avoidable. These are hard things to reconcile, but certainly I intend to keep them in mind when writing for things in the future.

To save face, I'll add a few other things that go along with Joelle's earlier points that may be helpful for non-scientist science writers:
-When you write a tossup on an element, if you mention an ionic compound of that element, you're almost certainly going to give away the element's oxidation state. If you say something like "This element's hexafluoride"--it's something that goes +6, which is not that common. You can get around this by saying something like "A fluoride salt of this element", which just reveals that it's a metal. Similarly, if you write that "this element is covalently bonded to three LIGAND in NAMED ORGANIC CATALYST", the answer is going to be nitrogen or phosphorus 90% of the time. Keep in mind those bonding rules you learned in high school.
-If you write a tossup on a "class of organic reactions", you should not be including the end result of the reaction in the early clues. I've edited questions before that lead in like "A reaction of this type, which converts a ketone to an alcohol, uses a chiral catalyst with boron attached to a proline derivative." The CBS catalyst is way less famous than the fact that ketone-->alcohol is a reduction.
-Signalling pathways are heavily shared. If you're writing a tossup on, like, epinephrine, saying that it binds a G protein that dissociates into Gs, which activates adenylyl cyclase, producing cAMP, which activates protein kinase A--just stop and delete all those clues, because anything that activates the PKA cascade does the same thing. The fact that it's a small molecule that activates PKA is okay for a half-sentence, non-uniquely identifying clue that points you in the right direction, but it needs clarification in the rest of the sentence (i.e., "The GPCR that activates a PKA cascade in response to this compound is blocked by propranolol", or something.)
-I would personally love it if Joelle, Eric, or somebody else with a medical background could talk about how to write symptoms clues for diseases, because I have yet to find ways that are uniquely identifying without being "named things", and as a result, these questions take me a longer time to write.

Also, a few selfish notes about using primary literature that I've been meaning to post for discussion for a while now:

I think clues from the literature are great, if you use them appropriately (READ: not ever in HS, not ever at ACF Fall, very sparingly if at all at Regionals). It tests who is regularly engaging with the academic side of a discipline, rather than reading packets. It can be a decent way to ground your tossup to make it uniquely identifying and can add context too. But even at CO, we shouldn't be asking about papers from minor journals that have fewer than 100 citations and have been around for years, just because it was the first thing you found when you typed it into Google Scholar. A more appropriate way to do this is to read an interesting, front-line paper from a high-impact journal and use this as the basis for the leadin to your tossup. Don't use paper clues for 2-3 sentences, please. The number of people who do read any particular journal is going to be pretty small and heavily biased toward graduate students.

But if you do get a clue from a paper, please include a citation to it (or even better, attach a pdf to your submission to the editors.) There were a few times that I had to ditch a clue because I couldn't for the life of me figure out where it had been drawn from--especially when I was operating without a host institution for a few weeks and didn't have access to journal archives. I'm sure that this is true in the humanities as well, but there are hundreds and hundreds of biology journals out there, plus journals of medicine, many of dubious quality, and there is no chance that the editor is going to have access to all of them, so if you're not sure, be a good person and attach the dang article to your packet submission.

Also, when you're citing biological literature, keep in mind that the last author on the paper is probably going to be the most helpful to your audience. If I write the following tossup on codons (sorry, this is off the top of my head from stuff I read recently...better examples probably exist)

"Ostrov et. al. propose eliminating seven of these things in a 2016 Science paper"

it is much less buzzable than if I say

"George Church's group proposed eliminating seven of these things in a 2016 Science paper"

because George Church is the principal investigator and is very, very well-known for doing synthetic biology/genome engineering, whereas Ostrov is some poor post-doc that did all the work. Some papers are known for the principal author on them (e.g., Gibson assembly), but to give your audience more context, mention the last author or two instead. I don't know how this works outside the biological literature, so if humanities/hard sciences people want to step in and talk about how they'd like to hear primary literature clues be used, that'd be appreciated.
Adam Silverman
Georgia Tech 2012-2016
Northwestern 2016-
adamsil
Wakka
 
Posts: 195
Joined: Wed Feb 09, 2011 6:20 pm

Re: On Exactitude in Science Writing

Postby Amizda Calyx » Sat Nov 26, 2016 1:22 pm

This is exactly the type of advice I was hoping would show up in this thread, and is a lot more applicable for writers with not much science experience than my vague high-standards editing guidelines. If more seasoned science people could jump in with examples from other fields, that would be great.

As for symptoms, Eric and Selene would be a lot more helpful since they're actual medical people, but in my experience most symptom clues just really don't work except at the ends of questions (where my fastidiousness for clue uniqueness admittedly starts to break down). This can be true for drugs too, although one way I try to get around that is by saying stuff like "this disease is the main indication for X treatment" (or, "drug X is mainly used to treat this disease").

One thing that really helped me starting out was to watch as an experienced writer edited other people's questions. It made me aware of bad tendencies in questions without putting me on the defensive. I may have my teammate Sam use my rough draft of our CO math tossup as an example of the kinds of problems people unfamiliar with a topic can run into.
Joelle Smart
Ellensburg High School, 2006–10
University of Washington, 2010–14
Rutgers University, 2015–20??
President of Quizbowl at UW, 2012–2014
PACE
HSAPQ biology editor

It is a mistake to think you can solve any major problems just with potatoes.
User avatar
Amizda Calyx
Forums Staff: Moderator
 
Posts: 236
Joined: Tue May 18, 2010 9:46 pm
Location: Seattle, WA

Re: On Exactitude in Science Writing

Postby Sima Guang Hater » Sat Nov 26, 2016 10:48 pm

RE: symptoms. This is actually easier than it looks, but it's a two-tier system.

The first tier is, there are some symptoms that are absolutely pathognomonic for a particular condition. These clues can be used unambiguously. Because I haven't read packets in a few months and just did psych, the only one that comes to mind is rotatory nystagmus for PCP intoxication. There are probably other conditions that cause rotatory nystagmus, but its association with PCP intoxication is super strong.

As you've correctly pointed out, though, a lot of symptoms are common between different diseases. This is harder to deal with, but some of these symptoms have a clear association with a disease even though it shows up in more than one thing. The only (unfortunately unhelpful) example I can think of is akinetic mutism for CJD, which is part of the diagnostic criteria for the disease but shows up in things like thalamic strokes and meningitis (but it would be stupid to toss up "stroke" or "meningitis" using that as a clue). Another example that comes to mind is the cherry red spot in the retina for Tay-Sachs, which shows up in retinal artery occlusion (I think), but that would again be a dumb thing to toss up.
Eric Anshi
Penn Imperial Civil Service Exam 1042
User avatar
Sima Guang Hater
Auron
 
Posts: 1711
Joined: Mon Feb 05, 2007 1:43 pm
Location: Kaifeng

Re: On Exactitude in Science Writing

Postby Ike » Tue Nov 29, 2016 3:44 pm

I don't know about biology in specific, but if you're writing submissions for the categories of science that I do, I would much, much, much, much, prefer that you use class notes and lectures slides over something like wikipedia. One of the problems with this post is that it's not obvious who it's geared toward, if anyone at all. By which, I mean, if you're new, seeking advice on how to write science questions, you should be writing stuff out of lecture notes over wikipedia, because you will produce a better, but imperfect question. And no one should ever be writing out of old packets.

I'm in agreement with Silverman on this, no one is going to be tossing up palmitic acid, and by and large when you play quizbowl, you should be giving the "best" answer. Even Eric, when he adds a note about symptoms implies that you have to have some amount of common sense when playing quizbowl, (but it would be stupid to toss up "stroke" or "meningitis" using that as a clue), or that it's impossible to nail things down precisely (There are probably other conditions that cause rotatory nystagmus, but its association with PCP intoxication is super strong). Editors have a finite amount of time and resources, and since you're citing tossups by some of the best editors in the game with problems, it doesn't seem right to me continuously ask more from them.

Also, Joelle, I'm going to take this opportunity to call you out on complaining about science without thanking editors for their hard work. You, more than any other poster in recent memory, complain about the small picture details of every writer's science, no matter how good, in every discussion thread. I haven't been added to the Terrapin discussion thread, but I'm willing to bet you have a bunch of nitpicks about individual questions there, even though I'm going to guess that the science was for the most part (very) good. While I don't think any writer is immune from criticism, I don't ever see you thanking anyone for their work, or even directing criticism in a way that's meant to be constructive, as opposed to say it's pure complaining. In fact it's been common for me to tell budding writers "ignore the tone in Joelle's comments" since it is demoralizing.
Ike
UIUC 13
User avatar
Ike
Yuna
 
Posts: 838
Joined: Sat Jul 26, 2008 5:01 pm

Re: On Exactitude in Science Writing

Postby Amizda Calyx » Fri Dec 02, 2016 3:08 pm

Ike wrote:I don't know about biology in specific, but if you're writing submissions for the categories of science that I do, I would much, much, much, much, prefer that you use class notes and lectures slides over something like wikipedia. One of the problems with this post is that it's not obvious who it's geared toward, if anyone at all. By which, I mean, if you're new, seeking advice on how to write science questions, you should be writing stuff out of lecture notes over wikipedia, because you will produce a better, but imperfect question. And no one should ever be writing out of old packets.

I never advocated writing any questions using just Wikipedia, but perhaps a warning should be added about that. I am standing by my position that using only personal class notes is a really bad idea -- clues should definitely be verified with multiple sources. This post was vague about its target audience mostly because I've seen issues with clues from all levels of writing experience. But I think primarily I aimed it at new editors and editors for categories that aren't their specialty, with the hope that seasoned writers/editors would chime in (as some have) with their own suggestions. I think aspiring science writers should also be cognizant of these issues. And of course people shouldn't be writing from old packets, but that doesn't mean a large swathe of players aren't heavily informed by packet-studying. I personally don't do this at all because I'm terrible at remembering isolated clues, but people who don't have this handicap may get a lot of their knowledge of a subject just from exposure to old questions.

Ike wrote:I'm in agreement with Silverman on this, no one is going to be tossing up palmitic acid, and by and large when you play quizbowl, you should be giving the "best" answer. Even Eric, when he adds a note about symptoms implies that you have to have some amount of common sense when playing quizbowl, (but it would be stupid to toss up "stroke" or "meningitis" using that as a clue), or that it's impossible to nail things down precisely (There are probably other conditions that cause rotatory nystagmus, but its association with PCP intoxication is super strong). Editors have a finite amount of time and resources, and since you're citing tossups by some of the best editors in the game with problems, it doesn't seem right to me continuously ask more from them.

Clue ambiguity was a fairly minor point in my post; I think reverse-clue-lookup is a pretty standard practice that people should be doing anyway. I agree that "common sense" should be used to narrow down an answer space, but as I pointed out in my response to Adam, editors/writers can be very biased with what they consider this to be. Sure, it's extremely unlikely that anyone would immediately think "palmitic acid" for that example from Adam, and I wouldn't actually expect this to be accounted for in a tossup. My rephrasing of that clue was to illustrate that it's not necessarily clunky to disambiguate things. I cited the carbonic anhydrase clue specifically because I remember hesitating on buzzing there due to my education. Like I said before, in my intro physiology course, CA was taught in the context of kidney cells, particularly wrt to acid-base regulation and hypertension treatment mechanisms (eg acetazolamide). I know it's usually a late clue, but I'm also not really that good of a bio player, especially in undergrad, so back in 2014 when I'd never taken a biochem or cell bio or molecular bio class CA would be one of the first things I'd recognize. Even (especially?) experienced editors will make assumptions about their audience based on their own academic track, and I think it's helpful to point out where biases can affect gameplay.

Except for the ethylene question, the other examples I use are, in my opinion, indefensible. If we go solely off what people learn in classes at pretty much any level, the MHC clue is downright wrong. And the transplant rejection question was awful. Just because they may have been edited by seasoned science players doesn't mean that they are exempt from critique! This was a big motivation for that whole section on complacency. As for the ethylene question, I honestly don't find it that objectionable -- I included it mainly because I wanted an example of a clue structure that I think can be troublesome. Namely, that it's the second line on a tossup with a fairly difficult answerline whose first clue requires pretty deep plant hormone signal transduction knowledge. It's pretty unlikely that anyone will be primed by knowing intracellular ethylene receptors that have never come up before, so the last word of the SAM clue being its sentence's only unique identifier means a full 2.5 lines are virtually unbuzzable except for the very best plant bio players (which I think is problematic for a regular-difficulty tournament).

Ike wrote:Also, Joelle, I'm going to take this opportunity to call you out on complaining about science without thanking editors for their hard work. You, more than any other poster in recent memory, complain about the small picture details of every writer's science, no matter how good, in every discussion thread. I haven't been added to the Terrapin discussion thread, but I'm willing to bet you have a bunch of nitpicks about individual questions there, even though I'm going to guess that the science was for the most part (very) good. While I don't think any writer is immune from criticism, I don't ever see you thanking anyone for their work, or even directing criticism in a way that's meant to be constructive, as opposed to say it's pure complaining. In fact it's been common for me to tell budding writers "ignore the tone in Joelle's comments" since it is demoralizing.

I know that I'm a harsh critic and have been unfairly judgmental at many points, including to you, and I apologize for the vitriolic tone in a lot of my comments. I have been trying to correct this, and I think I did a reasonably good job of commending the science in Terrapin. I complain about things a lot partly because I'm not very confident in my own knowledge, so I feel less qualified to endorse a good question, and anyway the questions I like most are typically the ones I first-line so it feels uncomfortable to say I enjoyed them without having seen the whole thing. But the bigger drive for this tendency is that it's pretty easy for me to identify issues with questions -- probably because in a lot of tournaments my rather "non-canonical" knowledge will trip me up and I'll neg with something (in)arguably correct that no editors anticipated, and it's extremely frustrating when this happens over and over again in the same tournament.
Hopefully this explanation (admission?) is adequate because I really don't think it's relevant or appropriate to drag out my personal foibles in this thread.
Joelle Smart
Ellensburg High School, 2006–10
University of Washington, 2010–14
Rutgers University, 2015–20??
President of Quizbowl at UW, 2012–2014
PACE
HSAPQ biology editor

It is a mistake to think you can solve any major problems just with potatoes.
User avatar
Amizda Calyx
Forums Staff: Moderator
 
Posts: 236
Joined: Tue May 18, 2010 9:46 pm
Location: Seattle, WA

Re: On Exactitude in Science Writing

Postby Mewto55555 » Fri Dec 02, 2016 10:24 pm

One of most people’s least favorite things in all of quizbowl (behind, of course, being beaten to a math tossup by Chris Ray) is getting taken to Screwed City by a question when it turns out they were approximately right/knew what a specific clue referred to but were ruled technically incorrect for whatever reason. It should come as no surprise, then, that another thing which is really annoying is when someone keeps complaining about questions that were technically incorrect, but confused literally no one!

Unfortunately, Joelle, it seems like this empathy you expect editors to have for you and your immense pools of Real Knowledge TM doesn’t extend too much the other way around. A lot of your posts exhibiting this phenomenon recently are about sets that aren’t clear, so this makes it a little hard to link it all together. I’ll post in one or two of them and reference the rest in this post in a way that doesn’t give away any information.

While many of the points you bring up in this thread are great (make sure the things you took notes on in class are factual! Don’t just copy from old packets! Be wary of what pop scientists think are important! Choose good pronouns!), there’s a lot in it that reads like a bunch of sour grapes. While it’d certainly be ideal to have every sentence fragment point on its own to a unique answerline, that’s impossible for a large number of clues that are nonetheless worthy of inclusion in a tossup. Much like the high school literature player, whom we expect to keep in mind enough context clues to realize that we’re in Russia and to not neg Dostoevsky with Saramago, the top-notch science player should be expected to pay enough attention and not complain when they neg with something very unlikely to be correct, given the preceding parts of the question (and, as someone whose knowledge is entirely real, it should be far easier for you to parse science clues for context than almost anyone else!).

In every question where there could be ambiguity, there’s a balancing act that needs to be done between adding enough information to make things completely unique and keeping things sufficiently terse that tossup length and understandability don’t spiral out of control. If we add an extra line to every single science tossup in order to avoid three instances per tournament of people overburdened by knowledge maybe getting screwed out of points that they can protest their way back to, then that’s an extra half-packet’s worth of questions that every single other team has to listen to, or a half-packet fewer worth of easier clues we could slot in for weaker players to buzz on. Sure, adding “It’s not X, but” to the leadin of a couple tossups isn’t that difficult (though who hasn’t laughed when sitting through an ACF Fall tossup that began “It’s not palmitic acid”), but the more complicated unnecessary fixes you’re asking for some of these nitpicks would leave far less space for useful cluing.

in my opinion there's a substantial difference between a standalone statement like "Band 3 is found on these cells' surfaces", and a sentence such as "A band 3 mutation in these cells can manifest as spherocytosis".

Yeah, the difference is that in the second example you’re causing a buzzer race thanks to a cliff from a moderately hard clue about red blood cells (which, according to you, can’t even be buzzed on uniquely until the end of the sentence) to “do you know what another name for spherocytes is???”…
because in a lot of tournaments my rather "non-canonical" knowledge will trip me up and I'll neg with something (in)arguably correct that no editors anticipated, and it's extremely frustrating when this happens over and over again in the same tournament.

There is a piece of good news, though! If you’re participating in a game where you're close enough to winning for a single tossup to matter in determining the outcome, you can actually protest the question and get the result of the game changed if it turns out you’re right! More details about this little-known loophole can be found in section H of the ACF rules, or section J of NAQT's. Unfortunately, it doesn’t come into play if all the protestor cares about is their PPG instead of how many games their team wins, but maybe one day we can get that rule changed.

Ike’s totally right, too. Editing at the college/open level is a pretty crappy job; editors put in a lot of work, are way underpaid compared to the effort they put in, and then have to deal with a ton of complaints afterwards. I’m certainly not in a position to sit here and talk about this too much (given that I’m done editing and never did that much college stuff to begin with, and certainly was never a top-tier editor) but I know a lot of the motivation is to give back to the community, some members of which sadly can be super-ungrateful. I think a lot of the most prolific ones would be pretty hurt or pissed off if you were actually insinuating that problems in their work were due to “complacency,” as opposed to honest mistakes caused by them working their asses off to provide you a good product. Hopefully, though, your first foray into college science editing will end up being error-free and we can all enjoy a great set!

Also, I get it. A lot of the science in ACF Regionals 2016 wasn’t very good. I know well now that I bit off far more than I should have tried to chew by stepping up to edit all ofthe bio/chem at the last minute, and I should have reached out for far more outside help earlier in the process. I received a lot of critiques of my work (include an every two-or-three-round live update from Ike about which questions he hated reading most!). I know most of it came from a good place (I rib Ike when I don’t like one of his tossups, and it’s more than reasonable that he gets to poke fun of me too), and the venting of people frustrated by being boned out of points can be helpful too. I even appreciate that you keep trying to provide the same feedback on the transplant rejection tossup that I already know to be subpar – if you want, I can help you set up a script to just post about it once every few months for the next couple years so you can save yourself the trouble (though I can’t help automate verbal complaints about it in person, sadly).

In concluding, I think it’s important to think about what one is trying to do when one makes a post in a discussion thread. Are you trying to vent because you’re pissed you got screwed by a question? Are you pointing out an error so that it can be fixed for a future mirror? Are you pointing out a general class of errors because you think it will help the editors in their future endeavors? Are you trying to show off and signal how much you know about a particular subfield? Are you just roasting someone because you feel their manner of posting is pretty unhelpful, and you hope they’ll improve it in the future? All of these are pretty common and reasonable (I've made at least one post of each type in my career!), but it’s usually unlikely one is helpful for a purpose other than that for which it was written -– it’s always incongruous, for example, when someone makes a dozen “ :capybara: you for screwing me out of these points” posts after every tournament and then turns around and tries to say they were merely trying to provide general advice for inexperienced science editors like Eric M. to be better. Hopefully people (me included!) try to keep this in mind when providing feedback on future tournaments.

Tangentially, I think “condition” is a pretty reasonable pronoun in a max flow tossup, since a max flow is just an assignment of numbers to edges that satisfies certain constraints, which seems to fit with a layperson understanding of what a condition is, though I’d have to see the tossup to comment further. But, unlike Joelle commenting on a tournament’s science, I don’t want to nitpick a small error instead of looking at the sum total of this thread, so I won’t belabor this point further.
Max
formerly of Ladue, Chicago
User avatar
Mewto55555
Forums Staff: Administrator
 
Posts: 707
Joined: Sat Mar 13, 2010 9:27 pm

Re: On Exactitude in Science Writing

Postby Amizda Calyx » Sat Dec 03, 2016 4:10 am

Well, Max, I'm sorry I've evidently caused you so much anguish with my criticism of a question you apparently wrote that you, Adam, and I all agree was poorly executed.

Anyway, as I said upthread, let's keep personal "roasts" to a minimum and return to the kind of helpful advice for writers/editors provided by Adam and Eric.
Joelle Smart
Ellensburg High School, 2006–10
University of Washington, 2010–14
Rutgers University, 2015–20??
President of Quizbowl at UW, 2012–2014
PACE
HSAPQ biology editor

It is a mistake to think you can solve any major problems just with potatoes.
User avatar
Amizda Calyx
Forums Staff: Moderator
 
Posts: 236
Joined: Tue May 18, 2010 9:46 pm
Location: Seattle, WA

Re: On Exactitude in Science Writing

Postby Bloodwych » Mon Dec 05, 2016 8:18 pm

Mewto55555 wrote:there’s a lot in it that reads like a bunch of sour grapes


This is a pretty pathetic way of delegitimizing criticism of a question in any context. "Be a better player" or "learn to buzz like a top-notch scientist" is a not a great riposte to someone who buzzes in with legitimate knowledge in a deep area and gives a right but "wrong" answer. As you have no doubt learned from Chris Ray by now, a lot of the game is learning to play by instinct and learning to listen to the voice in your head when it counts. That's not a good excuse to discount the preposterous idea that someone might know something technically correct in a question. Obviously, as you pointed out, the protest rules don't really account for rewarding knowledge when it doesn't matter in a game, but that doesn't mean that it shouldn't be talked about or taken seriously.

It's really a no-brainer that players shouldn't expect every clue in a question to be perfect, but it is not a useless exercise to point out things in a question that trip people up because it can only make whoever wrote it that much better at writing in the future. In fact, I don't think it's necessarily "nitpicky" at all as long as it lets them improve. That is, after all, is the point of this thread. This is not a defense of "complaining is a good form of criticism," but more a call to realize that it's not useless or petty to point out things in a question that misled you as long as it helps editors to improve.
Last edited by Bloodwych on Tue Dec 06, 2016 12:31 am, edited 1 time in total.
Chris
Quince Orchard HS '11
Maryland - College Park '15

I think I'll miss you forever
Like the stars miss the sun in the morning skies
Late is better than never
Even if you're gone I'm gonna drive, drive

I've got a tight grip on reality,
But I can't let go of what's in front of me here.
I know you're leaving in the morning when you wake up.
Leave me with some kind of proof it's not a dream.
User avatar
Bloodwych
Wakka
 
Posts: 144
Joined: Tue Jun 01, 2010 3:12 pm
Location: not College Park anymore

Re: On Exactitude in Science Writing

Postby Ike » Mon Dec 05, 2016 11:37 pm

I never advocated writing any questions using just Wikipedia, but perhaps a warning should be added about that. I am standing by my position that using only personal class notes is a really bad idea -- clues should definitely be verified with multiple sources. This post was vague about its target audience mostly because I've seen issues with clues from all levels of writing experience. But I think primarily I aimed it at new editors and editors for categories that aren't their specialty, with the hope that seasoned writers/editors would chime in (as some have) with their own suggestions. I think aspiring science writers should also be cognizant of these issues. And of course people shouldn't be writing from old packets, but that doesn't mean a large swathe of players aren't heavily informed by packet-studying. I personally don't do this at all because I'm terrible at remembering isolated clues, but people who don't have this handicap may get a lot of their knowledge of a subject just from exposure to old questions.


The fact that you don't realize that the bigger problem with newer writers is "people writing out of Wikipedia" instead of "people writing out of only personal class notes" suggests to me that you don't understand how the quizbowl ecosystem works. So let's talk about this a little bit. Threads like these aren't produced in a vacuum. It's OK to create threads to aggregate useful tips for writing. But when you want such a thread to exist, and begin by making a thread whose first post consists of shooting off like a rocket at questions produced by some of the game's best science writers, your suggestion that this thread contains useful advice is obviously just backpedaling since you come off as whiny in your original post.

Here's the larger ecosystem of the organ transplant tossup. As you may know, Max Schindler stepped up to handle biology and chemistry for ACF Regionals 2016 after Ben Zhang flaked on it. I don't think the science in the set was particularly good, and I probably would have still griped about Max's stuff even if Ben covered the biology and chemistry. So obviously, Max is going to produce some suboptimal questions under the circumstances. I think it's perfectly fine to call these questions out for being bad: in fact I did so to Max in person and if you check the discussion, you'll see that you and others did call out the organ transplant rejection for sucking. But it's been over six months from that fateful day in which the organ transplant tossup ruined the lives of everyone it touched, Max is no longer editing any science on the collegiate level, and hapless editors are not inspired by that tossup to write their science. So what is the point of continuously bringing it up to either Max or the community at large over and over again?

There's nothing that says an ACF or a housewrite editor has to listen to feedback or even read criticism. I'll be honest, when you post mathematics or computer science complaints I just skip it over. Other editors are starting to do the same for your complaints in other categories. And you know what - when your tone is passive-aggressive sarcasm over a question that was written 8 months ago in a situation where an editor was covering for someone else's delinquence, that's entirely OK. An editor and question writer is basically volunteering to produce their set. They do not have to make everyone happy, though they obviously want to. But if they don't hear a single word of praise when they do something right, and only get a mouthful for the shittiest question in the set they produced, it shouldn't come as a surprise that the QB community is getting annoyed with your posts.

Joelle, whether you intend it or not, your style of criticism drives people away from writing quizbowl science. When I volunteered to write NASAT's math because no one else was writing it, you attacked me in an unprofessional and hostile manner, even suggesting that "Ike Jose should not be allowed to write mathematics at any level." Aside from the fact that you have no say in what math I do edit, you've won half of the battle I guess: I refuse to write any biology or chemistry outside of my comfort zone for HSAPQ, even if the company needs to fill a set, and I know how to write it competently if not perfectly. When I recommend budding science writers to work for a company, it's always NAQT over HSAPQ since the communication and feedback is always courteous and professional. I don't know if that doesn't bother you, but hopefully what I'm saying is kind of a wake-up call to stop being such an ass to writers. If not, then it should be for anyone who chooses to edit with you, because your style of criticism is extremely toxic to nurturing and retaining talented science writers.
Ike
UIUC 13
User avatar
Ike
Yuna
 
Posts: 838
Joined: Sat Jul 26, 2008 5:01 pm

Re: On Exactitude in Science Writing

Postby Cheynem » Mon Dec 05, 2016 11:52 pm

For the record, Joelle has nothing to do with any science other than biology in HSAPQ, and people should actually write science questions for our company.
Mike Cheyne
The Ayatollah of Rock and Rollah
User avatar
Cheynem
Auron
 
Posts: 5979
Joined: Tue May 11, 2004 11:19 am
Location: Grand Rapids, Michigan

Re: On Exactitude in Science Writing

Postby RexSueciae » Mon Dec 05, 2016 11:56 pm

Ike wrote:There's nothing that says an ACF or a housewrite editor has to listen to feedback or even read criticism. I'll be honest, when you post mathematics or computer science complaints I just skip it over. Other editors are starting to do the same for your complaints in other categories. And you know what - when your tone is passive-aggressive sarcasm over a question that was written 8 months ago in a situation where an editor was covering for someone else's delinquence, that's entirely OK. An editor and question writer is basically volunteering to produce their set. They do not have to make everyone happy, though they obviously want to. But if they don't hear a single word of praise when they do something right, and only get a mouthful for the shittiest question in the set they produced, it shouldn't come as a surprise that the QB community is getting annoyed with your posts.


Jesus, dude.

When someone tells me that I have made an error, I generally thank them and continue on my merry way, older and wiser than before. People have sometimes told me things in very stern tones, or otherwise been less than friendly, and -- unless they're being actively hypocritical in their provided advice -- I don't take it personally to the degree that you sound like you've been taking it.

I'm trying to learn very specific subfields of science and one of the ways I learn things is by writing. I'd rather have someone who'd tell me exactly what I was doing wrong than someone who tiptoes around the point because they think my feelings would get hurt.

--

Science writing feels, to me, a very special category of quizbowl writing (the closest one could get would probably be the big debates over music that I dimly remember), and there isn't any sort of institutional backing to encourage people to write -- while people who know history or lit or myth or whatever don't face much of a hurdle in converting their casual knowledge into questions, one thing I notice about science (specifically, CS, because that's what I'm trying to learn) is that more often the body of canonical knowledge is divorced from things you'd learn in class or a useful hobby. If any quizbowl organization starts doing writer's workshops (have these been a thing at any point?) and they specifically highlight science as a Thing You Will Learn To Write, I'd sign up in a heartbeat.
Vasa Clarke
Maggie Walker '14
Virginia '18
RexSueciae
Rikku
 
Posts: 261
Joined: Sun Dec 29, 2013 12:24 am

Re: On Exactitude in Science Writing

Postby Marble-faced Bristle Tyrant » Tue Dec 06, 2016 12:19 am

The original post touched on some things I've noticed in my own editing assignments, and that's pretty much all I have to say.
Farrah Bilimoria
Formerly of Georgia Tech and Central High School (Macon)
User avatar
Marble-faced Bristle Tyrant
Yuna
 
Posts: 846
Joined: Fri Feb 08, 2008 11:05 pm
Location: Evanston, IL

Re: On Exactitude in Science Writing

Postby Amizda Calyx » Tue Dec 06, 2016 2:44 am

Ike wrote:
This post was vague about its target audience mostly because I've seen issues with clues from all levels of writing experience. But I think primarily I aimed it at new editors and editors for categories that aren't their specialty, with the hope that seasoned writers/editors would chime in (as some have) with their own suggestions. I think aspiring science writers should also be cognizant of these issues.

Joelle, whether you intend it or not, your style of criticism drives people away from writing quizbowl science. When I volunteered to write NASAT's math because no one else was writing it, you attacked me in an unprofessional and hostile manner, even suggesting that "Ike Jose should not be allowed to write mathematics at any level." Aside from the fact that you have no say in what math I do edit, you've won half of the battle I guess: I refuse to write any biology or chemistry outside of my comfort zone for HSAPQ, even if the company needs to fill a set, and I know how to write it competently if not perfectly. When I recommend budding science writers to work for a company, it's always NAQT over HSAPQ since the communication and feedback is always courteous and professional. I don't know if that doesn't bother you, but hopefully what I'm saying is kind of a wake-up call to stop being such an ass to writers. If not, then it should be for anyone who chooses to edit with you, because your style of criticism is extremely toxic to nurturing and retaining talented science writers.


I want to reiterate what Mike said about my holding no position in HSAPQ other than as biology editor; I'm not on the board and to my knowledge I no more represent the company than any other full writer. Additionally, I might ask why you are publicly posting about my behavior during editing last year's NASAT when I was under the impression that you had agreed you wouldn't do so if you accepted a formal apology from me regarding that incident. I had thought this issue resolved after you accepted my apology. Any forums critiques of your work since then have been strictly outside of my capacity in HSAPQ, so I am uncertain why HSAPQ is now being being brought up when as far as I know I have not been involved in any further official communication between you and the company. I would like to apologize to you again for my atrocious commentary during that time, but I would also like to apologize to HSAPQ for unknowingly having brought disparagement onto it by bringing up problems I had with a couple of your questions in recent sets. I do not recall when or where I would have "suggested" what is in the given quotation, but if you are paraphrasing something second-hand from some non-HSAPQ discussion you were not involved in I believe it is unfair to attribute it to the company.

I feel I have addressed some of your issues with my style of criticism in my first response to you. I won't comment on the other aspects of your post other than highlighting that I primarily aimed the post at new editors, who presumably have enough experience to know not to write straight from wikipedia.
Joelle Smart
Ellensburg High School, 2006–10
University of Washington, 2010–14
Rutgers University, 2015–20??
President of Quizbowl at UW, 2012–2014
PACE
HSAPQ biology editor

It is a mistake to think you can solve any major problems just with potatoes.
User avatar
Amizda Calyx
Forums Staff: Moderator
 
Posts: 236
Joined: Tue May 18, 2010 9:46 pm
Location: Seattle, WA

Re: On Exactitude in Science Writing

Postby Cody » Tue Dec 06, 2016 10:49 am

Amizda Calyx wrote:I want to reiterate what Mike said about my holding no position in HSAPQ other than as biology editor; I'm not on the board and to my knowledge I no more represent the company than any other full writer. Additionally, I might ask why you are publicly posting about my behavior during editing last year's NASAT when I was under the impression that you had agreed you wouldn't do so if you accepted a formal apology from me regarding that incident. I had thought this issue resolved after you accepted my apology. Any forums critiques of your work since then have been strictly outside of my capacity in HSAPQ, so I am uncertain why HSAPQ is now being being brought up when as far as I know I have not been involved in any further official communication between you and the company. I would like to apologize to you again for my atrocious commentary during that time, but I would also like to apologize to HSAPQ for unknowingly having brought disparagement onto it by bringing up problems I had with a couple of your questions in recent sets. I do not recall when or where I would have "suggested" what is in the given quotation, but if you are paraphrasing something second-hand from some non-HSAPQ discussion you were not involved in I believe it is unfair to attribute it to the company.
I would note this either way, but I hope it carries some special weight given that I was HSAPQ's Chairman during the production of last year's NASAT: Ike (and indeed, anyone in HSAPQ) is well within his "rights" to discuss his experience in HSAPQ in any forum, public or private. Obviously, I would hope that anyone with a grievance would bring it to the attention of HSAPQ's higher-ups first, so that they can at least attempt to resolve any issues. And I believe we reached a somewhat satisfactory resolution in this case—at the very least, Ike is (very graciously) still writing for us.

To be clear, I no longer speak for the company, but I have no doubts that Mike Cheyne feels the same way. HSAPQ is not a company that requires you to sign a non-disclosure agreement about its operations, and I think that's a good thing.
Cody Voight – CBGS ‘09, VCU ‘14. I write lots of science and am an electrical engineer.
VCU Tournament Director ‘13-pres. HSAPQ President ‘15-16.
Hero of Socialist Quizbowl Labor (NSC ‘14). “esteemed colleague” of Snap Wexley, ca. 2016. Stats Hero (Nats ‘16).
Quizbowl at VCU
User avatar
Cody
2008-09 Male Athlete of the Year
 
Posts: 2007
Joined: Sun Nov 15, 2009 12:57 am
Location: Richmond

Re: On Exactitude in Science Writing

Postby touchpack » Tue Dec 06, 2016 11:33 pm

I fully agree with Ike that the subtext of this thread makes writing science seem impossible and intimidating and is discouraging to new writers. These types of nitpicks tend to range from "minor" to "extremely minor" in terms of how much they actually affect gameplay, while being able to write questions that avoid these problems requires enormous amounts of background knowledge and effort. When talking to budding science writers, I try to avoid talking about nitpicks on this level unless it's something really blatant, (like a clue that doesn't even come close to picking out a single answer) for this reason: it's difficult for them to avoid and is less important to constructing a good tossup. In my mind, by FAR the most important skill when writing science questions (or questions in any category, really), is being able to find good question topics with clues that people who study the subject (either academically or out of personal interest) care about, and I tend to emphasize that over all else when giving feedback.

A couple specific theoretical points I dispute:

Regarding uniqueness of individual clues: there are scenarios where a clue can be technically not unique, but that's not a problem. I can't think of any great examples off the top of my head, but consider this scenario. Clue A is canonically taught in undergraduate classes as something that applies to answer X. However, there's a 20 citation journal article written in 2013 that shows it could also apply to answer Y. (fun fact, in a recent HSNCT, there was a game-deciding protest regarding a scenario very similar to this one) This is not a problem, because 1) the odds of someone actually being familiar with that journal article are vanishingly small (in the HSNCT scenario I reference, the player wasn't actually familiar with the article--he was just guessing). 2) Even if a player is familiar with the journal article, a player that has that much high-level knowledge is, 99% of the time, going to have the quizbowl skill to realize they just want X. This is what Max was trying to get at in his post--He's not just "delegitimizing criticism" by saying "git gud scrub." (although, when you post something along the lines of "this cholesterol tu is problematic because a clue that also applies to palmitic acid", my natural response IS "git gud scrub", because any reasonable player will know that palmitic acid is never going to be tossed up) So according to the premise of this thread, in this situation you would have to add "It's not Y, but clue A". I would posit that in this scenario, you're actually making the question worse by wasting valuable clue space you could be using to construct a better pyramid. In the 0.0000001% chance someone actually buzzes with answer Y, they can protest if they really need their points. I'm not going to waste everyone's time with "it's not palmitic acid" when it's fully unnecessary.

Now, it's not always quite so cut-and-dry as "taught in all undergraduate classes" vs "single 20 citation journal article"--there is a gray area where a clue could potentially be slightly problematic, but is very unlikely to be. This thread comes off as very antagonistic to writers that accidentally intrude onto that gray area, but in the greater scheme of things this is a highly technical point about how to take your writing from A to A+ material, while this thread makes it seem like it's some widespread major issue throughout quizbowl or something.

In addition, there's places where it's fully on the other side of the gray area, where it's definitely impossible for a reasonable player to distinguish whether you want answer X or Y based on clue A. These are the types of criticisms that are useful for new writers, and I highly encourage any budding writer reading this thread to just read Adam Silverman's posts and ignore all of the other posts.

Regarding sentence uniqueness, it is often perfectly okay for a sentence to not be unique until the end. As long as you aren't baiting a player into buzzing with something different (it's hard to give any general advice to whether or not a given clue will do this; it's kinda just a case-by-case thing), having "non-unique" clues in the beginning and middle parts of a sentence can provide vital context clues to the player and/or let the player perform what I call an "anticipatory buzz".

An anticipatory buzz is when a player uses context clues to formulate a guess, then buzzes on a sentence fragment successfully because they anticipated what the rest of the sentence would say. I personally do this all the time, (a recent example that readily comes to mind is the commutator tossup at last CO, where I buzzed on obviously not-unique sentence fragment "For the Pauli matrices") and I consider it to be one of the hallmarks of a good quizbowl player. Now, let's look at the example you picked out:

"In the biosynthesis of this compound, Met Adenosyltransferase converts methionine to SAM, which is then converted to ACC."]


Now, I see what the problem here is, the wording is slightly suboptimal. The issue is that SAM serves as a cofactor for methyl group transfers in tons of different biochemical reactions. However, that's not what's going on here--in this case SAM is serving as the substrate itself. I personally would have worded it something like "In the biosynthesis of this compound, it is formed from a methionine precursor which is converted to SAM, which is then converted to ACC." This is still technically not unique until the last word of the sentence, but it provides some "evocative" material that would allow for a player to contextualize what's going on and possibly make an anticipatory buzz. This clue is fine.

Oh, and yeah, when you repeatedly bring up one tossup that was hastily written by Max last-minute because his actual biology editor :capybara: ed out on him leaving him to write all of the biology and chemistry for an important collegiate set last-minute, then yeah, you can expect that Max is going to be "caused anguish" when you call it out as an example of a systemic problem (which does not actually exist).

tl:dr this thread is bad (except for Adam's posts) and new writers should ignore it.
Billy Busse
Illinois '14
Member, ACF
Writer/Subject Editor/Set Editor, NAQT
touchpack
Wakka
 
Posts: 244
Joined: Tue Sep 20, 2011 12:25 am

Re: On Exactitude in Science Writing

Postby sbraunfeld » Wed Dec 07, 2016 1:46 am

People seem to have missed the fact that, despite the title of this thread, the content of the initial post is not "You must read every journal article to check for potential ambiguities in clues, and if you find any, the way to correct your clue is with the following construction: 'It's not [list of every other conceivable answer], but …'". Reverse-clue lookup is only passingly referenced, the palmitic acid clue that people are repeatedly bringing up doesn't appear, nor is anything like writing "It's not palmitic acid, but …" suggested. If anything, I find the primary issue with the original post is that it is thoroughly uncontroversial, its contents have been laid out many times before, and any editor ought to already know these things.

The main points seem to be: (1) don't use bad pronouns (2) sometimes words have technical meanings that differ from their ordinary usage, and change the meaning of clues (3) If using clues from class notes or packets, check with a reliable source (4) Do reverse clue lookup sometimes, such as when unsure about material from personal notes, as in the given furosemide example, to ensure clues don't point to multiple reasonable answerlines (5) pop science distorts and simplifies, and is a bad source for determining importance of answerlines (Addendum) clues give context beyond the specific fact you're referring to, and this can cause pyramidality issues.

Some specific examples of ambiguities are mentioned, but the context is "Here are some ambiguous/incorrect clues that are now in the archives. Be wary of being informed by packets", rather than "Here are some examples of our widespread ambiguity crisis. You should do exhaustive reverse-clue lookup, so you don't write like these awful writers". If you disagree with the particular examples, then pick some others (e.g. from Billy's post in the ACF Fall thread). There may not be a plague of ambiguous and incorrect clues, but there are enough trusting the information in packets can easily create problems.

If people find Joelle's posts in discussion forums to be unnecessary and unhelpful, they may have a valid point (and perhaps there ought to be a separate thread about how to post constructively in tournament discussions). But that has no bearing on the contents of the initial post.
Sam Braunfeld
Berkeley '13
Rutgers
sbraunfeld
Lulu
 
Posts: 54
Joined: Wed Jan 06, 2010 2:09 am

Re: On Exactitude in Science Writing

Postby sbraunfeld » Wed Dec 07, 2016 2:56 am

By the way, for an example of a non-science discussion unmarred by irrelevant personal attacks following a post actually about exactitude in clues, see here. (Note: I am not necessarily endorsing the methods suggested here.)
Sam Braunfeld
Berkeley '13
Rutgers
sbraunfeld
Lulu
 
Posts: 54
Joined: Wed Jan 06, 2010 2:09 am

Re: On Exactitude in Science Writing

Postby Amizda Calyx » Wed Dec 07, 2016 6:15 am

Ok, I'm getting pretty frustrated that people seem to be latching onto this idea that my entire post is about "clue fragment ambiguity" and that I'm forcing nitpickiness down the throats of "new writers". Maybe I should have changed the title to "On Exactitude in Science Editing" or just dropped the Borges pun altogether, since apparently that has led some people to interpret my whole post as being targeted at ambiguity and new writers despite the times I explicitly say this is not the case.

Re: ethylene: I also noted that I had no objection to this clue in isolation; I was just concerned about the buzzability of it when "ACC" is 2.5 lines from the start of the question and the leadin hasn't shown up before in qb.

As for palmitic acid, here is the relevant paragraph in my response to Adam:
I'm not advocating people reverse-look up every clue exhaustively before submitting a tossup; my primary objective with this post was to raise writers' awareness of issues they might not have considered before. I also understand that writing science is a real burden for a lot of teams, and I'm not expecting any changes to occur at that layer of participation. But for people who WANT to write/edit science, and for those of us who already do, I think it's a good idea to keep these points in mind. I also don't believe that cool clues that have some ambiguity can only be discarded or hopelessly embulkified -- in my opinion there's a substantial difference between a standalone statement like "Band 3 is found on these cells' surfaces", and a sentence such as "A band 3 mutation in these cells can manifest as spherocytosis". I also don't find it too difficult or clunky to include qualifying information, eg "Palmitic acid and this molecule are added to the N-terminus of sonic hedgehog to localize its signaling" (which, btw, is a clue I would have loved to have heard after going over cholesterol's role in Shh gradient control fairly extensively in my first molecular bio course).

Note that A) I was not the person to bring up this hypothetical palmitic acid/cholesterol situation; B) I offer a less bulky alternative to Adam's example of having to clunk up a good clue with clarifiers -- my rewording is supposed to be a suggestion for that *general* type of cluing problem (as is the Band 3 example that I also did not originate), not me claiming "this is exactly what I would do in this exact situation"; C) I literally say that the issues I brought up, which cover quite a bit more territory than just the sentence where I allude to reverse clue lookup, are meant to be things people who want to seriously write and edit science should be aware of.

Also, regarding reverse clue lookup, which seems to have provoked a visceral reaction among people who misinterpreted my original paragraph referencing it: Here is a relevant topic by Eric on the subject, and here is a related discussion.

As for my behavior in threads, I'd like to give a comparison between the tone of my criticism of ACF Fall bio and that of Billy's regarding its physics. What if those subjects were edited under similar circumstances to Max's or some other acceptably "reasonable" duress that wasn't publicized? Do we really expect all players to be aware of the particular context of a writer/editor's situation for every tournament before they post any criticism (I certainly wasn't!)? Or is it really that context matters iff the editor/writer in question is an established Good Science Player and Experienced Editor with a lot of friends who also fit those descriptions, and anyone else who tries to justify a bad set is just incompetent?

Here are links to the most recent things I have said in other discussions. Nitpicky? Sure, although at least for some of the items I point out an argument can be made for other very reasonable answers. I'll let others decide whether I'm overly harsh.
  • STIMPY (Note: this post was made after being chided by Chris Manners for never following with a detailed criticism promised in my (mostly praising) initial post).
  • CO 2016. I do regret not commending Adam more on the rest of the bio, which I thought was really good -- despite what some people might say about CRISPR miscellanea showing up too much, I thought it was refreshing to hear a bioengineering question on cas9! Here is another post by me in the same thread, as well as the follow-up by Tommy.
  • MYSTERIUM. Yeah, this wasn't a helpful post, especially since it was so late.
  • HSNCT 2016. Please also compare tone with the quoted text.
  • MLK
  • Regionals 2016. This response and my bringing it up again ITT are I think what primarily motivated Max's post. Yes I am really harsh here: I was aware that Ben had ditched editing, but I was not aware that editing had actually been taken over at the last minute by Max and assumed this was just an unedited submission. After the other team negged with what I think was an autoimmune disease, my teammate Stephanie buzzed in with "graft versus host disease", which is explicitly rejected in the answerline. We ended up winning that game against Brown by 5 points.

After looking over my own old (public) forums posts as well as the relevant topics by more esteemed players/editors, I'm honestly just depressed that I've inspired such personal animosity among so many people that airing scathing grievances against me overrides any possible utility in what I've said. To me, it looks like my status as a mediocre player and inexperienced college-level editor, coupled with a few rude but definitely not unusually-caustic public posts, alongside the far more acerbic things I said to Ike on a more private platform, have turned this thread into what it is now. I find it pretty troubling that a theme seems to be "[not-good players] aren't qualified to find fault in these questions written by better players" and "only people in the elite 'in-group' are allowed to use words like 'terrible' and ' :capybara: ing stupid' and ' :chip: show' in their criticisms because everyone knows they're all just good buddies ribbing each other". As for my own behavior in the past, I can only assume that the complaints are driven by my use of specific examples to illustrate my points. Well, based on my personal experience, I think one of the most beneficial things for new editors is to watch someone else point out general issues in particular questions (i.e. edit in front of them). Maybe I didn't choose the ones that most exemplified my points, and maybe it was in bad taste to pick out recent questions that writers are still sensitive about. But why should that invite the hatred that has shown up in response? Why should that delegitimize the other aspects of my post that other commenters have said are helpful and touch on recurring problems? Why does there have to be the implication that I'm a pretentious knowledge-signaling :capybara: if I mention that my knowledge of the subject I have a degree in, something not represented as much as (or as "canonical" as) the cell/molecular/biochem topics in qb that I admit I'm deficient in and have difficulty learning, sometimes trips me up and points me towards answerlines emphasized in my education but not in the writer's? Take a look at Adam's thoughtful post and the discussion it prompted: what is so problematic with this that it needs to be completely derailed into attacks on my personality and capability as a player/editor?
Joelle Smart
Ellensburg High School, 2006–10
University of Washington, 2010–14
Rutgers University, 2015–20??
President of Quizbowl at UW, 2012–2014
PACE
HSAPQ biology editor

It is a mistake to think you can solve any major problems just with potatoes.
User avatar
Amizda Calyx
Forums Staff: Moderator
 
Posts: 236
Joined: Tue May 18, 2010 9:46 pm
Location: Seattle, WA

Re: On Exactitude in Science Writing

Postby vinteuil » Wed Dec 07, 2016 11:20 am

Joelle, I think people are latching onto slightly tangential topics for the reason Sam brought up: this thread doesn't contribute much that's new. (Not that that's actually a problem. It just means people won't have much to say about it.)

sbraunfeld wrote:If anything, I find the primary issue with the original post is that it is thoroughly uncontroversial, its contents have been laid out many times before, and any editor ought to already know these things.

The main points seem to be: (1) don't use bad pronouns (2) sometimes words have technical meanings that differ from their ordinary usage, and change the meaning of clues (3) If using clues from class notes or packets, check with a reliable source (4) Do reverse clue lookup sometimes, such as when unsure about material from personal notes, as in the given furosemide example, to ensure clues don't point to multiple reasonable answerlines (5) pop science distorts and simplifies, and is a bad source for determining importance of answerlines (Addendum) clues give context beyond the specific fact you're referring to, and this can cause pyramidality issues.

I mean, John Lawrence's guide to science writing says all of these things! And it's not like he figured them out from years and years of great science editing work! Your points—"don't do the things that are stupid" and "do do the things that are not stupid"—are well taken (and sometimes it's necessary to reiterate them for categories that use technical language); but maybe it shouldn't be so surprising that people don't have much more to say in response, and are much more interested in addressing other problems in the post.
Jacob Reed
Yale '19
East Chapel Hill '13
"...distant bayings from the musicological mafia"―Denis Stevens
User avatar
vinteuil
Yuna
 
Posts: 983
Joined: Sun Oct 23, 2011 12:31 pm

Re: On Exactitude in Science Writing

Postby The Stately Rhododendron » Wed Dec 07, 2016 2:00 pm

I found your OP useful, Joelle. This (inexactitude) happens all the time with my areas of expertise and it's frustrating that it remains a problem. If you remember last year, I, too, faced plenty of grumbles for posting about my disagreements wrt Anthro, culminating in an argument with Matt Jackson in a airport. I found Ike's mansplaining about editing and posting in this thread incredibly irritating and I wish you didn't have to face all this lecturing about how to post (including the previous post, which seems to veer close towards telling people how to post!).
Oakland Mills '14 - The Most Unpredictable Team in Quiz Bowl
Yale '18 - Majoring in X Studies
"Field Commander" Isaac KD, "The Savior of the Forest"
User avatar
The Stately Rhododendron
Rikku
 
Posts: 405
Joined: Sun Oct 16, 2011 7:18 pm
Location: Heart's in the woods

Re: On Exactitude in Science Writing

Postby Auks Ran Ova » Wed Dec 07, 2016 2:31 pm

The Stately Rhododendron wrote:Ike's mansplaining

Please do not devalue this potentially useful term with such a heinous misapplication. It's a genuine concern, especially in any community with a gender breakdown like quizbowl's, but I don't want it to be used to handwave away good-faith efforts at discussion and criticism. If anyone in this (or any other) conversation is feeling inappropriately condescended to, please come to me--I am happy to help set up a clarifying conversation or whatever else might be necessary.

Additionally: experienced writers, editors, and critics debating among themselves the ways in which various types of question criticism are or are not helpful is inarguably an appropriate use of this forum, and it would absurd to crack down on it as being somehow "against the rules". "I don't think this is a particularly useful approach to criticizing questions, and here's why" is very different from "you don't know anything, don't post" or "you can't comment on my region because you're not from here". The "telling other people how to post" rule may have been applied with an inappropriately heavy hand at times in the past, but I have every confidence that the current board staff is more than capable of applying the rule fairly.
Rob Carson
University of Minnesota '11, MCTC '??
Member, ACF
Member, PACE
Writer and Editor, NAQT
User avatar
Auks Ran Ova
Forums Staff: Chief Administrator
 
Posts: 3795
Joined: Sun Apr 30, 2006 10:28 pm
Location: Minneapolis

Re: On Exactitude in Science Writing

Postby Sima Guang Hater » Wed Dec 07, 2016 8:33 pm

So this thread illustrates two useful points that I want to distill.

First, we're imperfect. People - whether they're professors, editors, or writers - are only human. The errors Joelle lists from her class notes are actually pretty easy misspeaks ("lysine zipper" for "leucine zipper") that doctors or professors or whoever else make very often. My resident today told me that a particular type of stroke happens in the thalamus instead of the internal capsule, so I had to jump in and correct him (nicely. As in "are you sure? I might be wrong but I remember reading..." Medical students develop a whole script for doing this). The MHC error is something I could make when I'm tired and cranking out questions quickly, and would ideally be caught somewhere down the line (and I have made similar errors - the clue about neuromyelitis optica in ACF Nationals 2016 comes to mind). This is made much worse when you're working outside of your area, as Max discovered when he tried to perform an organ transplant - he thought he was qualified because he knew the MELD score, but no dice.

But we have a mechanism to deal with this problem - tournament discussion threads. There's a natural equilibrium where people are screwed by a mistake, then they post about it (sometimes in excessive terms). The editor takes the criticism on the nose, tries to do better, and/or offers reasons for why the question is the way it is.

This thread represents an attempt, in my estimation, to make a larger comment on how to drive the number of mistakes even further downward. This is a noble goal. However, I don't think it offers much of anything new in how to prevent mistakes, and I think we're reaching the point of diminishing returns where all we can say is "please double check your writing", "double-check your sources", or "playtest your questions". Rather than pointing out the pervasive existence of a hitherto unknown type of systemic error and offering a way to correct it (like my thread on reverse clue lookup attempted to do), that part of the original post is a general call to be more careful about what essentially amounts to random error. This is a lesson we can all take to heart, but its not really a new lesson, and I think this is part of what's frustrating to Ike, Billy and others posting in this thread.

Second, quizbowl is a language game that has constraints. The point Joelle makes about the appropriate use of pronouns is a good one - a word like "condition" can mean something different in ordinary English than it does in science. However, most of the cases she pointed out are pretty easily discernible by an experienced player. We've all been a little befuddled by the choice of pronoun in a question, but it's worth remembering that the editors probably spent a good deal of time trying to figure out what the best pronoun is for those particular questions. You never really have to refer to the Halting problem or Max Flow is those oblique terms when you're learning about them, and part of what makes quizbowl a language game distinct from either science or ordinarily English is the fact that you have to refer to concrete things obliquily. I personally don't think "condition" is a particularly bad pronoun here, and I don't know that anyone was really thrown by that usage. I'll note also that generally, a good moderator and a generous prompt line can deal with any pronoun confusion. For example, one year at CO there was a tossup on Khomeni's fatwa against Salman Rushdie which called it a "statement"; I buzzed in and said something like "This statement? 'Kill Salman Rushdie'?!" and was prompted my way to points once I figured out what they were going for.

The bottom line is, quizbowl is different from class and ordinary English, and coming up with the appropriate pronoun is a great illustration of this. The best thing to do is to approach it on a case by case basis - if you think scenario isn't a great pronoun for "halting problem", it may be worth starting a larger conversation about what constitutes an appropriate pronoun for "halting problem". But I don't think this thread is offering any kind of general rule for how to select an appropriate pronoun, which would be very useful (though I don't think such a rule really exists).
Eric Anshi
Penn Imperial Civil Service Exam 1042
User avatar
Sima Guang Hater
Auron
 
Posts: 1711
Joined: Mon Feb 05, 2007 1:43 pm
Location: Kaifeng

Re: On Exactitude in Science Writing

Postby vinteuil » Wed Dec 07, 2016 9:41 pm

One thing in this thread (from multiple people's comments) that I think needs some explication: how do we determine when cluing was confusing?

There are two extreme cases that I think we can agree on:
  1. It's possible for any player to be confused by any wording, no matter how non-confusing that wording is (i.e. using any definition of "non-confusing"); you could have zoned out, the moderator screwed up, teammates were annoying, whatever. In other words, confusion isn't exactly an inherent property of the clue—
  2. On the other hand, it's obviously possible to word a clue in a way that confuses a large number (even all) of the people playing it, even under ideal attentional and environmental conditions (and even if they know what's being clued). So confusion can be an inherent property of the clue.

With these two facts in mind, I think we should be careful about how we muster evidence about a question's confusingness. Point 1. shows that you can't "prove" that a question is confusing by just saying "I was confused." But it also cuts both ways—you can't "prove" that a question "wasn't confusing" by saying "I don't know anybody who was confused."

The only satisfying argument you can really make is probabilistic/statistical. So, maybe we might be clearer if we posted things like: "I think that if you learned the material in X way, you would be inherently confused by the cluing. Moreover, I bet/suspect that a lot of people who {do|might do} quizbowl learned it X way and no other way." [EDIT: I'm not saying that people don't post like this already! And sometimes, the confusion can be at a more basic, linguistic level—but that's something that any conscientious editor with a reasonable amount of time should catch, so maybe it's not worth writing theoretical posts about.]

It should be possible to provide some evidence for assertions like this—maybe a widely-used introductory text? Or a sketch survey of syllabi? And if, having learned it X way, we were confused, did the research, and found out that the confusion was just an artifact of an idiosyncrasy of our schooling, we learned something too—and we no longer need to post.

In other words, we should keep in mind all of Bruce Arthur's posts about how our experiences aren't necessarily universal—but also the possibility that they were at least widely representative.

Of course, this kind of distinction/careful wording isn't just helpful if you're talking about wording. I don't mean to single out or make fun of Isaac here, but I think that if he had really tried to look up how often Ruth Benedict shows up in anthropology surveys, he might have been surprised, realized his experience wasn't universal, and nuanced his point a little bit: Ruth Benedict is studied frequently for her work on Japan, sometimes for her comparative work in Patterns of Culture, and rarely in other contexts—so don't write about her obscure stuff. (At any rate, he had this experience when a Yale anthropology course on Japan assigned her work this semester....)
Jacob Reed
Yale '19
East Chapel Hill '13
"...distant bayings from the musicological mafia"―Denis Stevens
User avatar
vinteuil
Yuna
 
Posts: 983
Joined: Sun Oct 23, 2011 12:31 pm

Re: On Exactitude in Science Writing

Postby Auks Ran Ova » Wed Dec 07, 2016 9:52 pm

vinteuil wrote:One thing in this thread (from multiple people's comments) that I think needs some explication: how do we determine when cluing was confusing?

There are two extreme cases that I think we can agree on:
  1. It's possible for any player to be confused by any wording, no matter how non-confusing that wording is (i.e. using any definition of "non-confusing"); you could have zoned out, the moderator screwed up, teammates were annoying, whatever. In other words, confusion isn't exactly an inherent property of the clue—
  2. On the other hand, it's obviously possible to word a clue in a way that confuses a large number (even all) of the people playing it, even under ideal attentional and environmental conditions (and even if they know what's being clued). So confusion can be an inherent property of the clue.

With these two facts in mind, I think we should be careful about how we muster evidence about a question's confusingness. Point 1. shows that you can't "prove" that a question is confusing by just saying "I was confused." But it also cuts both ways—you can't "prove" that a question "wasn't confusing" by saying "I don't know anybody who was confused."

The only satisfying argument you can really make is probabilistic/statistical. So, maybe we might be clearer if we posted things like: "I think that if you learned the material in X way, you would be inherently confused by the cluing. Moreover, I bet/suspect that a lot of people who {do|might do} quizbowl learned it X way and no other way." [EDIT: I'm not saying that people don't post like this already! And sometimes, the confusion can be at a more basic, linguistic level—but that's something that any conscientious editor with a reasonable amount of time should catch, so maybe it's not worth writing theoretical posts about.]

It should be possible to provide some evidence for assertions like this—maybe a widely-used introductory text? Or a sketch survey of syllabi? And if, having learned it X way, we were confused, did the research, and found out that the confusion was just an artifact of an idiosyncrasy of our schooling, we learned something too—and we no longer need to post.

In other words, we should keep in mind all of Bruce Arthur's posts about how our experiences aren't necessarily universal—but also the possibility that they were at least widely representative.

Of course, this kind of distinction/careful wording isn't just helpful if you're talking about wording. I don't mean to single out or make fun of Isaac here, but I think that if he had really tried to look up how often Ruth Benedict shows up in anthropology surveys, he might have been surprised, realized his experience wasn't universal, and nuanced his point a little bit: Ruth Benedict is studied frequently for her work on Japan, sometimes for her comparative work in Patterns of Culture, and rarely in other contexts—so don't write about her obscure stuff. (At any rate, he had this experience when a Yale anthropology course on Japan assigned her work this semester....)

I endorse this product and/or service.
Rob Carson
University of Minnesota '11, MCTC '??
Member, ACF
Member, PACE
Writer and Editor, NAQT
User avatar
Auks Ran Ova
Forums Staff: Chief Administrator
 
Posts: 3795
Joined: Sun Apr 30, 2006 10:28 pm
Location: Minneapolis

Re: On Exactitude in Science Writing

Postby rajk » Thu Dec 08, 2016 12:16 am

Personal class notes. These are especially pernicious because they give the illusion of accuracy and unambiguousness.


Another reason why this can be bad if you don't look up the clue is that professors very often make up terms for convenience, that are not "real" terms and are used only in that class. For example, my professor talked about and used the "Sunglass Lemma" extensively, but when I went to look it up, the only reference is his own notes. It's also possible that the professor is using nonstandard terminology, like QSAT instead of TQBF. Or in an algebraic geometry class, maybe if you read Hartshorne, you would say "algebraic set" instead of "variety" and no one would know what you're talking about unless they'd read Hartshorne too. Or worse, you say "a variety has this property", for a tossup on irreducibility, but this is not true for anyone else's definition of "variety."
Bala
UC Berkeley, 2018
rajk
Lulu
 
Posts: 23
Joined: Thu Oct 25, 2012 10:55 pm
Location: California

Re: On Exactitude in Science Writing

Postby Ike » Thu Dec 08, 2016 5:12 am

rajk wrote:
Personal class notes. These are especially pernicious because they give the illusion of accuracy and unambiguousness.


Another reason why this can be bad if you don't look up the clue is that professors very often make up terms for convenience, that are not "real" terms and are used only in that class. For example, my professor talked about and used the "Sunglass Lemma" extensively, but when I went to look it up, the only reference is his own notes. It's also possible that the professor is using nonstandard terminology, like QSAT instead of TQBF. Or in an algebraic geometry class, maybe if you read Hartshorne, you would say "algebraic set" instead of "variety" and no one would know what you're talking about unless they'd read Hartshorne too. Or worse, you say "a variety has this property", for a tossup on irreducibility, but this is not true for anyone else's definition of "variety."


It's not the case that editors are going out of their way to include this nonstandard terminology in their tossups. In fact, all the good editors actively remove informal shorthand and nonstandard terminology from their questions. So again, this seems kind of obvious. But if you're a new writer, do not fret about this! Writing a good tossup that has one or two minor blemishes is >> than submitting a tossup straight out of wikipedia. If you're editing a tournament for the first time, it's OK if a couple of your questions suffer from this particular blemish; if you produce a stack of questions that has all the fundamentals down, no one should be giving you crap about your good work!

Actually, that's one of the reasons why this thread's subtext is so bad, and I'm willing to go after it with fire and sword. Most of the time, when the biggest complaint about a stack of science questions is "shouldn't have used terminology from class notes," it already is a really good set of science questions! If a set's science sucks, there's many other issues to be complaining about, and complaining about this would be a very weird complaint. So if anything, you should probably be thanking the question writer for writing a stack of decent science. And that's where the tone issue kicks in: a bad tone automatically puts a seasoned question writer on the defensive, and scares away new potential talent. It would be one thing to say "Hey, I liked this set's science, can I see the tossup on X since it seemed to me to use non-standard terminology?" versus the tone presented in Joelle's most recent interaction with me outside of this thread in Penn Bowl 2016, where I redact clue content:

"The tossup on '[redacted]' was just not good. Aside from being really transparent (to the point that I and several other people expected it to be asking for something more specific and hesitated to buzz on the [redacted] clue), the answerline should include what ... buzzed with at the [redacted] clue, which was '[redacted]'."


In this particular instance, I made one change based on Chris Chiego's explanation of this comment, so it's not like this comment was contentless. But when the tone in a comment consists of hostility ("this question is bad!"), has an assertion about what the answerline should include based off of a wrong perception of what is right (as it turned out, the player had a gameplay misunderstanding of what happened, a la what Jacob Reed identifies), cites vague sources ("several other people") in a thin attempt to give credence to a perceived injustice, it automatically puts a question writer on the defensive, and they're not going to want to engage with you. I think this was a B-grade tossup, it had 85% of the components it needed to be a great tossup, so I take issue with the part of this comment that says it "was just not good," since it wasn't a complete travesty. Furthermore, I take issue with this tone when it is directed toward the Penn writing team, which largely consisted of a cadre of writers who are working on their first tournament and tried to write some good earth science -- it's incredibly off-putting!

As I mentioned earlier, the theory of this thread is OK. But my issue with the initial post and subsequent morass lies in the fact its subtext is offensive to writers and editors: what we're talking about here essentially boils down to really minor nitpicks that rarely, if ever, affect gameplay. I don't know who you are Raj aside from being that one guy who decided to attack the hard work of people who wrote computer science without thanking them a while back in that one thread. But if you actually think this usage of personal class notes is a systematic problem, it's not. And if you encounter a set of questions that happened to have this one minor blemish but are otherwise immaculate / made you happy, please thank their author for their hard work.
Ike
UIUC 13
User avatar
Ike
Yuna
 
Posts: 838
Joined: Sat Jul 26, 2008 5:01 pm

Re: On Exactitude in Science Writing

Postby rajk » Thu Dec 08, 2016 6:56 pm

Ike, you seem to like making giant strawmen out of anything anyone says.

It's not the case that editors are going out of their way to include this nonstandard terminology in their tossups. So again, this seems kind of obvious.

Yes. No one was ever suggesting this that editors are "going out of their way." And it's not really obvious. If your entire class talked about some concept with a name, a new writer might not consider that there might be other names for it without looking it up. You yourself said that this thread should be for constructive criticism, well this is it.

Writing a good tossup that has one or two minor blemishes is >> than submitting a tossup straight out of wikipedia.

No one has advocated for writing a tossup straight ouf of Wikipedia in this thread. Joelle even said so explicitly. Why are you pretending like new writers only have the mental capacity to write ONLY from Wikipedia or ONLY from notes?

However, I do agree with you on this point completely. But it's not just new writers who write straight from Wikipedia. If you look up Chicago Open mathematics, you'll notice that almost every single clue is directly from the answerline's Wikipedia page or a Wikipedia page that was linked from the answerline's page.

But if you actually think this usage of personal class notes is a systematic problem, it's not.

I said nothing like this. I noted that a few tossups that I've seen used nonstandard terminology, so I posted some examples of how this could arise and why it's important to look up key terms before using them. Stop with the strawmen, it just wastes everyone's time.

But my issue with the initial post and subsequent morass lies in the fact its subtext is offensive to writers and editors: what we're talking about here essentially boils down to really minor nitpicks that rarely, if ever, affect gameplay.

You keep saying this even though it's been pointed out multiple times for multiple tossups that these questions are either just completely incorrect, or ambiguous even at the undergrad level. I don't know about biology but this is certainly true in CS (when I asked you to point out the specific criticisms you had problems with in my other thread, you didn't respond).

Moreover, I think you're taking everything too personally. Objectively, the OP has useful advice, to not blindly trust class notes, or packets, or your own memory, all of which I'm sure you agree with. There doesn't have to be a pernicious subtext to everything. Eric and others seem to have pointed out that all this has already been said many times, in which case maybe it is redundant, but it's certainly not offensive. At the very least, Adam's reply seems to contain useful and original tips for writers.

And if you encounter a set of questions that happened to have this one minor blemish but are otherwise immaculate / made you happy, please thank their author for their hard work.

Sure. Penn Bowl CS was mostly good. Mysterium CS was excellent. Your 2015 ACF Nats bonus on complexity classes/IP/GI was good. I mentioned this in my post, but Lederberg CS and Avogadro CS were great.
Bala
UC Berkeley, 2018
rajk
Lulu
 
Posts: 23
Joined: Thu Oct 25, 2012 10:55 pm
Location: California

Re: On Exactitude in Science Writing

Postby sbraunfeld » Thu Dec 08, 2016 8:39 pm

Ike wrote: I don't know who you are Raj aside from being that one guy who decided to attack the hard work of people who wrote computer science without thanking them a while back in that one thread.


I would like to mention that pointing out that errors and ambiguities in questions, particularly in the quite level tone present in Bala's thread, hardly constitutes an "attack". Also, I'm uncertain what the relevance of the fact that you don't know him is.
Sam Braunfeld
Berkeley '13
Rutgers
sbraunfeld
Lulu
 
Posts: 54
Joined: Wed Jan 06, 2010 2:09 am

Re: On Exactitude in Science Writing

Postby sbraunfeld » Thu Dec 08, 2016 8:41 pm

Sima Guang Hater wrote:The bottom line is, quizbowl is different from class and ordinary English, and coming up with the appropriate pronoun is a great illustration of this. The best thing to do is to approach it on a case by case basis - if you think scenario isn't a great pronoun for "halting problem", it may be worth starting a larger conversation about what constitutes an appropriate pronoun for "halting problem". But I don't think this thread is offering any kind of general rule for how to select an appropriate pronoun, which would be very useful (though I don't think such a rule really exists).


As far as appropriate pronoun use, in many cases there is an obvious pronoun, and I think that's almost always the right one. There are occasions when using the obvious pronoun gives too much context, but I think these cases are fairly rare. Writers should remember that they know the answerline, and so their question seems more transparent to them than it will to players. And, of course, if you do decide to use a less specific pronoun, you should make sure it doesn't introduce ambiguity to the question.

Take the halting problem question (from 2015 Nats). (Disclosure: The use of "this scenario" led me to neg with "undecidability" on the second clue.) The obvious pronoun is "this problem". The question actually does switch to "this problem" towards the end, which makes me think that "this scenario" was chosen because of transparency concerns. First, I think that concern is unwarranted. It's not as if there's only one computer science problem that ever gets tossed up; there was a question on the knapsack problem earlier in the same tournament (which used the pronoun "this problem" throughout). Furthermore, the use of a less specific pronoun made the second clue ambiguous. That clue mentions a theorem "resulting from this scenario", and then describes the proof that the halting problem is undecidable, with the use of "scenario" leaving it unclear whether the referent is undecidability or the halting problem. (On a side note, aside from the pronoun choice, I really like this clue. I have elsewhere mentioned that I think describing proofs is a good way to reward actual knowledge of math and cs, although it takes familiarity with the subjects to execute well.)

I'll close with some pronoun recommendations for math questions. For questions about "things" (in some suitably Platonic sense), e.g. triangles, groups, vectors, I recommend "object" as a pronoun (Caveat: This might cause problems if you use a category-theoretic clue.) Other nice pronouns are "property" (e.g. independent, differentiable), "function" (e.g. polynomials, continuous functions), and "value" (e.g. dimension, determinant).
Sam Braunfeld
Berkeley '13
Rutgers
sbraunfeld
Lulu
 
Posts: 54
Joined: Wed Jan 06, 2010 2:09 am

Re: On Exactitude in Science Writing

Postby The Ununtiable Twine » Thu Dec 08, 2016 11:50 pm

rajk wrote:
Writing a good tossup that has one or two minor blemishes is >> than submitting a tossup straight out of wikipedia.

If you look up Chicago Open mathematics, you'll notice that almost every single clue is directly from the answerline's Wikipedia page or a Wikipedia page that was linked from the answerline's page.
[/quote]

Chances are I won't address you in more than this post as I am enjoying my retirement at the moment, but I will address you.

Typically, the majority of interesting and accessible information about math topics appears on Wikipedia because the pages are typically concise. It is the job of the editor to determine the cluing order and whatnot, as well as what is relevant information and what is not. Just because math appears on Wikipedia doesn't mean that the tossup was copied/pasted from Wikipedia. Just because math appears on Wikipedia doesn't mean that it is forbidden in cluing. For instance, with the matroids tossup, I read an entire article on matroids to make that tossup on an admittedly hard answerline as natural as possible (which I believe I linked that paper in the CO discussion thread). I also think someone yapped at me for using my notes to write tossups (which, as referenced earlier in the thread, is a valid way of doing things as long as clues are verified - admittedly I could have done a little better with this). I even provided samples of my notes in the thread.

I do occasionally click on a link and clue something that I find is interesting and relevant to the topic, which is fine as long as the clue is correct. Additionally, there are a lot of things that appear in Wikipedia, MathWorld and the like that also appear in textbooks. It remains unbeknownst to you where I got the information, although according to your eloquently wordsmithed post, if I clued something simple like Lagrange's theorem, you could state that "it was retrieved from Wikipedia" instead of thinking that perhaps the editor gets his information from some references which Wikipedia quotes. As it happens, I am near an internet connection way more often than I am near my textbooks, and so I use alternate sources to confirm what I already know (which, since you seem to think I copy and paste from Wikipedia, is quite a bit mind you). Theorems are theorems, lemmata lemmata - the overwhelming majority of math articles on Wikipedia are maintained by professionals, so Wikipedia is rather trustworthy when it comes to math. If Wikipedia is used to remember a theorem, there's nothing wrong with that. Lastly, if you're insinuating that I just copy and paste from Wikipedia (while *somehow* achieving the most accessible bonuses of any of the science editors in the process!), that's rather petty and I'm expecting an apology in your next post.
Jake Sundberg
Louisiana '04-'10, '14-'16
Alabama '10-'14
Associate Director, Louisiana Quiz Bowl Alliance
User avatar
The Ununtiable Twine
Yuna
 
Posts: 982
Joined: Fri Feb 02, 2007 11:09 pm
Location: Lafayette, LA

Re: On Exactitude in Science Writing

Postby Cody » Fri Dec 09, 2016 12:58 am

shockingly, a free encyclopedia, constituting the largest reference work on the Internet, might restate academic content you can find from other sources, and restate it correctly?

Raj, a number of your critiques strike me as originating from a position of naivety. This doesn't make your criticisms invalid, but it does make me question your approach. Wikipedia, for example, is a perfectly fine source for all manner of subjects as long as the clues are subsequently vetted. It is a poor place for new writers to pull ideas from, but it is a fine place for experienced writers to find clues. The fact that some (or many) CO math questions sounded like they took clues from Wikipedia is not prima facie evidence of bad clues (or questions).
Cody Voight – CBGS ‘09, VCU ‘14. I write lots of science and am an electrical engineer.
VCU Tournament Director ‘13-pres. HSAPQ President ‘15-16.
Hero of Socialist Quizbowl Labor (NSC ‘14). “esteemed colleague” of Snap Wexley, ca. 2016. Stats Hero (Nats ‘16).
Quizbowl at VCU
User avatar
Cody
2008-09 Male Athlete of the Year
 
Posts: 2007
Joined: Sun Nov 15, 2009 12:57 am
Location: Richmond

Re: On Exactitude in Science Writing

Postby Sima Guang Hater » Fri Dec 09, 2016 1:01 am

A lot of confusion in this thread is arising from the fact that Joelle begins by talking about new writers, but then says that the posts should be directed at editors rather than writers. I have to take Billy and Ike's side on the utility of this thread - if I were a new writer, I wouldn't worry too much about what's going on in this thread, focus more on writing questions from your textbooks and class notes, and let the editors, with their experience, fix up your questions. Furthermore, I think this fixation on minutiae in Joelle and Sam's post is really unhelpful and I hope people aren't turned off from writing because of that fixation. Raj's post in the other thread is a really bad example of this - none of those examples make those questions unbuzzable, and didn't really screw up anyone's game (whereas other mistakes have).

I also think Ike's questions are being unfairly singled out in this thread for criticism, which is strange to me because I've known him to be extremely careful about the facts and language in his tossups. In particular, Sam, your buzz of "undecideability" makes very little sense to me, because the leadin is about the halting probability and the second clue, which leads in with "the proof of the principal theorem resulting from this scenario", is about proving the halting problem is undecideable - you have the causality backwards if you buzz with "undecideability" there, because the result that the halting problem is undecideable clearly arises from the halting problem.

I can't tell people how to post, but I think more empathy and less personal affront (both giving and taking) would make this and any future thread about writing much easier to read. It's kind of shocking how much vitriol I see in a thread ostensibly about making quizbowl better.
Eric Anshi
Penn Imperial Civil Service Exam 1042
User avatar
Sima Guang Hater
Auron
 
Posts: 1711
Joined: Mon Feb 05, 2007 1:43 pm
Location: Kaifeng

Re: On Exactitude in Science Writing

Postby rajk » Fri Dec 09, 2016 1:32 am

The Ununtiable Twine wrote: It is the job of the editor to determine the cluing order and whatnot, as well as what is relevant information and what is not.


Raj, a number of your critiques strike me as originating from a position of naivety....It is a poor place for new writers to pull ideas from, but it is a fine place for experienced writers to find clues.


Jake and Cody, I agree with you 100%, but I think my initial wording was very unclear.

I'm not saying there's anything wrong with looking at Wikipedia at all. I'm not saying any of these tossups are bad, because I don't have the math knowledge to evaluate them in the first place. As you mention, math is math and it doesn't matter what the source is as long as it's correct. If I insinuated that you copy and paste directly, I apologize.

The only thing I wanted to note is that Ike said that new writers often use clues "straight from Wikipedia", but even in CO, where the tossups are very difficult, most of the clues are still listed on Wikipedia, which is to say that using Wikipedia can lead to both bad and good questions depending on the writers' expertise, so the combination of using things you have learned in class + Wikipedia makes for a more "robust" question.

Raj's post in the other thread is a really bad example of this - none of those examples make those questions unbuzzable, and didn't really screw up anyone's game (whereas other mistakes have).

This is going offtopic and should probably be on the other thread, but I don't understand this. Objectively, even with the most generous interpretations, the questions are wrong. If that didn't mess anyone up that's fine (for example the BB(5) inaccuracy was unimportant), although I don't see how that's possible for other questions - I sent you a note about the pumping lemma tossup in Penn Bowl where a factual error caused me to neg (not during in an actual game). As another example, I'm sure the problems in the 2011 Nats DFA question tripped some people up. My only goal in that thread was to write down mistakes that I have come across and conjecture on how to avoid them, not to point out bad questions.
Bala
UC Berkeley, 2018
rajk
Lulu
 
Posts: 23
Joined: Thu Oct 25, 2012 10:55 pm
Location: California

Re: On Exactitude in Science Writing

Postby Amizda Calyx » Fri Dec 09, 2016 2:00 am

Sima Guang Hater wrote:I have to take Billy and Ike's side on the utility of this thread - if I were a new writer, I wouldn't worry too much about what's going on in this thread, focus more on writing questions from your textbooks and class notes, and let the editors, with their experience, fix up your questions. Furthermore, I think this fixation on minutiae in Joelle and Sam's post is really unhelpful and I hope people aren't turned off from writing because of that fixation. Raj's post in the other thread is a really bad example of this - none of those examples make those questions unbuzzable, and didn't really screw up anyone's game (whereas other mistakes have).

I want to reiterate that this thread was aimed at people who want to edit/become better writers of science, although I do think Sam's tip about the types of pronouns to use in math questions is really informative for new writers as well. The halting problem "minutiae" that he brings up were a direct response to what he quotes you as saying. And, newer editors may not have seen previous threads on science writing/clueing and certainly haven't had the opportunity to participate in their discussions, which is partly why I wanted to make this topic. Plus, advice from people with strengths in particular subdisciplines like math and CS isn't that widespread, so why not let people talk about their experiences writing/editing/playing in those categories?
As for Bala's thread, if it's true that some of the questions he mentions have literal errors in them, then that is a problem! In what other categories do we permit factual errors and blatant ambiguities to (apparently regularly) show up without discussion because they "probably don't make the question unbuzzable"? I think that occasionally the same sub-ideal approaches are made in writing math/CS questions that produce inaccurate or poorly-stated clues, and, as people like Vasa have implied, dissecting such inappropriate methodologies could be valuable for improving writing. Because not a lot of players are fluent enough in those categories to identify most errors, many of them may not have been noticed at the time they were played/during the discussion period -- unlike the more major areas such as lit or history where inaccuracies can be addressed immediately. Just because these questions are old doesn't mean they should be immune from (reasonable) critique from the people who are most knowledgeable of/affected by them.
Joelle Smart
Ellensburg High School, 2006–10
University of Washington, 2010–14
Rutgers University, 2015–20??
President of Quizbowl at UW, 2012–2014
PACE
HSAPQ biology editor

It is a mistake to think you can solve any major problems just with potatoes.
User avatar
Amizda Calyx
Forums Staff: Moderator
 
Posts: 236
Joined: Tue May 18, 2010 9:46 pm
Location: Seattle, WA

Re: On Exactitude in Science Writing

Postby sbraunfeld » Fri Dec 09, 2016 2:14 am

Sima Guang Hater wrote: Furthermore, I think this fixation on minutiae in Joelle and Sam's post is really unhelpful and I hope people aren't turned off from writing because of that fixation.


I only bothered to post about this at your suggestion, and I tried to make sure there was a general point to the post. While my issue with the halting problem question is on the one hand minutia (which is why I haven't bothered posting about it before), I think it nicely illustrates the general, and new-writer/editor-friendly, point that one usually shouldn't spend time trying to think up obscurantist pronouns, but just go with the obvious choice.

Sima Guang Hater wrote:I also think Ike's questions are being unfairly singled out in this thread for criticism, which is strange to me because I've known him to be extremely careful about the facts and language in his tossups. In particular, Sam, your buzz of "undecideability" makes very little sense to me, because the leadin is about the halting probability and the second clue, which leads in with "the proof of the principal theorem resulting from this scenario", is about proving the halting problem is undecideable - you have the causality backwards if you buzz with "undecideability" there, because the result that the halting problem is undecideable clearly arises from the halting problem.


Well, time to continue arguing about minutiae. I'm pretty sure the argument "You should have known the lead-in; then the ambiguous second clue wouldn't have made you neg" is not widely accepted. Your next claim also makes very little sense to me. The halting problem and the fact that it is undecidable both "exist" (to put on my Platonist hat) prior to the theorem, and the theorem "results" (whatever that may mean) from the one no more than from the other.

Sima Guang Hater wrote:Raj's post in the other thread is a really bad example of this - none of those examples make those questions unbuzzable


I'll note that I negged the 2011 Nats DFA question when I buzzed in on the Moore/Mealy machine clue (a distinction frequently applied to NDFAs as well) with "finite state machines", was prompted, and could think of no better answer. (I guess I buzzed, so this technically isn't an example of a question being unbuzzable, but it's probably worse.) Some of the other issues he points out are sufficiently wrong/ambiguous to make those clues unbuzzable, and some escape the unbuzzability criterion only by virtue of being bonuses. Since when is having a blatant and misleading falsehood as a bonus lead-in ("We are sure this class is not equal to PSPACE") acceptable? Is this also an issue new editors shouldn't concern themselves with?

If there's an issue with Bala's thread, it's just that there is no general lesson for most people to take away from it. But there is still some useful information more knowledgeable writers can glean, such as "when referring to an Arthur-Merlin protocol, consider whether the number of steps is important". Since Bala's thread didn't specify an intended audience of new writers/editors, this seems fine.
Sam Braunfeld
Berkeley '13
Rutgers
sbraunfeld
Lulu
 
Posts: 54
Joined: Wed Jan 06, 2010 2:09 am

Re: On Exactitude in Science Writing

Postby Cheynem » Fri Dec 09, 2016 10:36 am

I don't know anything about science, but to be honest, more questions are slightly ambiguous/vague/slightly wrong/misleading than you might suspect. Usually a couple times a tournament I will think to myself "Hmm, is that the best way of describing that history thing?" or "Well, that actually really applied to everyone in the administration." In some way, sure, these questions could be made better; on the other hand, I just want to point out that this is not unique to science questions. I cast no opinion on this thread and its potential usefulness.
Mike Cheyne
The Ayatollah of Rock and Rollah
User avatar
Cheynem
Auron
 
Posts: 5979
Joined: Tue May 11, 2004 11:19 am
Location: Grand Rapids, Michigan

Re: On Exactitude in Science Writing

Postby vinteuil » Fri Dec 09, 2016 11:56 am

Cheynem wrote:I don't know anything about science, but to be honest, more questions are slightly ambiguous/vague/slightly wrong/misleading than you might suspect. Usually a couple times a tournament I will think to myself "Hmm, is that the best way of describing that history thing?" or "Well, that actually really applied to everyone in the administration." In some way, sure, these questions could be made better; on the other hand, I just want to point out that this is not unique to science questions. I cast no opinion on this thread and its potential usefulness.


Exactly this: questions often get historical causality, or even chronology flat-out backwards, because it's hard not to have that happen if you write so many history questions. Just like it's really hard not to over-drastically simplify philosophical arguments; or to represent properly the relationships between fictional characters; or to get a really concise but still correct description of a painting. Lots of small errors happen all the time, and that's almost a given considering how little time most editors (volunteers!) are working with.

This is why most people realize that it's neither necessary nor worthwhile to post a critique of every misleading or incorrect clue they read or hear in practice.

In fact, Joelle:
Amizda Calyx wrote:As for Bala's thread, if it's true that some of the questions he mentions have literal errors in them, then that is a problem! In what other categories do we permit factual errors and blatant ambiguities to (apparently regularly) show up without discussion because they "probably don't make the question unbuzzable"?

Could it be that you only notice the errors in subjects you know things about? As I said above, this happens in literally every category. But if you want another example: have you forgotten about threads like this? (recap: John and I were complaining that clues were either wrong or literally not grammatically correct enough to be understood; since I was being kind of an asshole about it, people either said "Clearly we can't do any better, so shut up!" or "Just because you don't like the clues doesn't mean the questions are a disaster!")
Jacob Reed
Yale '19
East Chapel Hill '13
"...distant bayings from the musicological mafia"―Denis Stevens
User avatar
vinteuil
Yuna
 
Posts: 983
Joined: Sun Oct 23, 2011 12:31 pm

Re: On Exactitude in Science Writing

Postby Ike » Fri Dec 09, 2016 2:20 pm

Bala, I don't meant this to be condescending, bu you haven't edited any packet submission tournaments to know anything about what new writers do and don't do. I hope that if you're ever science editor, you receive the same tossup on packets and Code Division Multiple Access that I did, and you come back here and tell me with a straight face that this thread will be useful to those writers. That's not even talking about the actually bad questions that comprise 20% of our submissions which we chuck out quickly.

Sure. Penn Bowl CS was mostly good. Mysterium CS was excellent. Your 2015 ACF Nats bonus on complexity classes/IP/GI was good. I mentioned this in my post, but Lederberg CS and Avogadro CS were great.


No, what I'm saying is, even if they produce a flawed question, you should be thanking them for trying, especially if you want them to listen to you. A pretty good example of this was your poor wording of Sundberg's work on mathematics for Chicago Open. I'm gonna be generous and assume you accidentally got another editor irritated. My point is, people do spend a lot of time on writing questions, and unless they are being categorically unreasonable you should be more careful about this if you want them to listen to you and not just go "lol, Berkeley guy is complaining again!" Also, you only wrote up your posts in computational complexity theory. Do you really have enough knowledge to say the same for the rest of computer science? Was the anti-aliasing tossup any good from 2011 nationals, or the TU on Boltzmann Machines from 2012? So in effect you're claiming expertise in like 1/10th of computer science, and aside from the fact I think you're writing from a position of naivete like Cody Voight says, your posts do feel rather narrow, and that when you say the CS is good, you mean "the CS that I know is good**". After all this is a thread about exactitude in science writing.



I'll talk about the halting problem question in detail for a bit. The second sentence states "The proof of the principal theorem resulting from this scenario relies on [description of Rice's Theorem by halting problem reduction]." So, in this case the principal theorem is Rice's theorem. Rice's Theorem states " that all non-trivial, semantic properties of programs are undecidable." So the sentence unambiguously reads "the proof of Rice's Theorem resulting from the halting problem relies on [description of Rice's Theorem by halting problem reduction.]" It's really wrong to say Rice's Theorem results from undecidability, other than the concept of undecidability exists before hand (much in the same way numbers exist. Actually, if you do buzz in and say "numbers" in this tossup, your points will change, I assure you!) This feels to me like one of the cases outlined by Jacob Reed - misunderstanding of the English in the question. As for the usage of "this scenario", this was something that I chose after careful deliberation. I chose it because I felt that we can't have an ACF Nationals tossup whose pronoun was "this problem" and heavily discussed Turing Machines in its middle clues. In fact, this was not a choice that was made in a vacuum, I discussed it with former ACF Nats CS editor Jerry Vinokurov, and former science writer phenom Cody Voight (two people who have lots of expertise!) read through this question, and they concurred in that it was unlikely to be confusing, and that this was a good idea. I just can't even come close to seeing that this is a good-faith argument -- it's almost as bizarre as answering "F. Scott Fitzgerald" for a question that asks "This dude kills Jay Gatsby." In a weird sense, that answer isn't wrong, and if the protest committee were run by a set of tenured metaphysicians, you may get the question thrown out. But to the rest of us, it feels to me that you have made an error that Jacob Reed identifies upthread and are contorting English to try lasso your way to these imaginary points. For what it's worth, for the third straight year in a row, I'm your math and CS editor for ACF Nationals, I'm also your CS editor for ICT, and I'm writing a huge portion of the math. All of my questions / edits will get reviewed by Seth Teitler at NAQT, and there will be at least one other CS/MATH expert who I will consult as well. I just went back through my conversion data for ICT, and it's not the case that my science questions are causing anything remotely close to anomalously high neg rates that you would expect if imprecision in language were creating ambiguities like these on a massive scale.

Re: unfair criticism of Ike.

The halting problem was brought up as a footnote in the original post, and speaking personally, it does feel like you two - Joelle and Sam are targeting me, as you guys have done in the past. Apparently, I'm not being unreasonable because someone else has even brought this up! Joelle's initial post from my understanding, was a comment about the halting problem that was mentioned by Sam to her, and she doesn't really have a bone to pick with this tossup. So it really makes sense to only hear it from Sam, and I question why Joelle even included it in her manifesto. Sam and Joelle, this past summer one / both of you expressed to others something along the lines of Ike Jose should never be allowed to edit math again on any level after I wrote mathematics that you didn't like for NASAT. I do take offense that this was a sentiment that Joelle (and / or Sam) expressed to other writers, and while I doubt that it did do damage, because it was similar to a bedtime story of hamfisted intimidation, it's not something I appreciate -- after all I did spend one year of my life making a living entirely through writing QB questions. So, yes when either of you make a thread like this, it does feel you are trying to further undermine my work when two peole I barely know, treated me rudely last year, and told other folks that I should not be allowed to edit math ever again.

Sam asks
"Also, I'm uncertain what the relevance of the fact that you don't know him is."
, the relevance is what happened this past summer: I don't know you Sam at all. I have never met you before except maybe once or twice at a QB tournament, and you spoke to me only through posts from your significant other's account by posting using her username or having her write up harshly-worded attacks on my writing and vent to other people what I said earlier about me editing math. Again, you're someone who I barely know, and it's impossible for me to tell whether it's you or Joelle who is saying this stuff, so the idea that you would think I would be OK with this is ludicrous. But either way, if any person gives this acerbic feedback all the time through the guise of their significant other because they can't talk vis a vis, I want them to stay the :capybara: away from me, and stay the :capybara: away from every other writer. In fact my interaction with you two as happened over the summer, and somewhat in this thread, has been so toxic that for these two reasons, and only these two reasons, I recommend that no one take either of you two as a science editor. And I recommend that no new writer take either of your posts seriously.

But there is still some useful information more knowledgeable writers can glean, such as "when referring to an Arthur-Merlin protocol, consider whether the number of steps is important".


lol, next time someone asks for CS feedback, I'll be sure to send them this gem. Look, this is a grad level topic, it doesn't take a wizard or king to see that this advice will affect no one's writing, since this has a total of 3ish hits in the last 16 years of packet writing.

Also the empire in Borges's "On Exactitude in Science" destroyed itself by being too exacting with science, nitpicking to create a map in every detail, thereby refuting the existence of this thread. Protip: if you're going to pun on the title of a literary work, you should make sure that it itself doesn't undermine your argument.

Also, this is my last post in this thread, because I just don't care anymore. I'm going to drink some malort with Andrew Wang in the near future because of this thread. And speaking of Andrew Wang, Joelle claimed upthread that one of her posts was better in tone than Andrew Wang. That is categorically false, Andrew Wang is probably the most articulate, concise, and well-toned poster I know (how can you be all three at the same time?!?!).

** To be fair, you could be the most badass person alive and actually have implemented anti-aliasing, and solved the n=12 case of the two-state busy beaver function with your own Boltzmann Machine, possibly at the same time. So I really don't know how much CS you have expertise in and am not trying to call you a bad CS student (in fact you've demonstrated that you know more about IP than I do), but just consider how much you actually know, and don't insist that a tournament's CS was excellent when there were tossups that had serious issues.
Ike
UIUC 13
User avatar
Ike
Yuna
 
Posts: 838
Joined: Sat Jul 26, 2008 5:01 pm

Re: On Exactitude in Science Writing

Postby rajk » Fri Dec 09, 2016 5:05 pm

Bala, I don't meant this to be condescending, bu you haven't edited any packet submission tournaments to know anything about what new writers do and don't do. I hope that if you're ever science editor, you receive the same tossup on packets and Code Division Multiple Access that I did, and you come back here and tell me with a straight face that this thread will be useful to those writers. That's not even talking about the actually bad questions that comprise 20% of our submissions which we chuck out quickly.

I
As I said above, this happens in literally every category.

Ike, I don't think this is condescending at all. I haven't even written many questions so it's true that I don't know what useful or not useful advice is for writers. Part of the reason I posted was because I didn't realize that there were a lot of errors in other categories, and I wasn't sure if the stuff I pointed out was already known. But are you saying only people who have edited tournaments should post criticism? At the very least, I hope nothing I posted was incorrect, and if it doesn't help anyone practically, then it can be ignored safely.

No, what I'm saying is, even if they produce a flawed question, you should be thanking them for trying, especially if you want them to listen to you. A pretty good example of this was your poor wording of Sundberg's work on mathematics for Chicago Open.

Okay, I see what you're saying, if I post anything in the future I'll make sure to thank them, and choose my wording carefully.

that when you say the CS is good, you mean "the CS that I know is good**".

You are correct, at least for higher levels.

lol, next time someone asks for CS feedback, I'll be sure to send them this gem. Look, this is a grad level topic, it doesn't take a wizard or king to see that this advice will affect no one's writing, since this has a total of 3ish hits in the last 16 years of packet writing.

Maybe more accessible advice is "reduction to and reduction from are different, be careful" or "try to copy down theorems/lemmas exactly unless you know them very well".
Bala
UC Berkeley, 2018
rajk
Lulu
 
Posts: 23
Joined: Thu Oct 25, 2012 10:55 pm
Location: California

Re: On Exactitude in Science Writing

Postby Amizda Calyx » Fri Dec 09, 2016 5:45 pm

Ike wrote:And speaking of Andrew Wang, Joelle claimed upthread that one of her posts was better in tone than Andrew Wang. That is categorically false, Andrew Wang is probably the most articulate, concise, and well-toned poster I know (how can you be all three at the same time?!?!).

Amizda Calyx wrote:
Bubalus Period wrote:who's the :capybara: ing idiot who let in a tossup on prolactin


There was a tossup on prolactin??? With what high school/early college clues? Dopamine, anterior pit, oxytocin and...? Please tell me metamorphosis and histology stuff weren't clues...

............

Ike wrote:Re: unfair criticism of Ike.

The halting problem was brought up as a footnote in the original post, and speaking personally, it does feel like you two - Joelle and Sam are targeting me, as you guys have done in the past. Apparently, I'm not being unreasonable because someone else has even brought this up! Joelle's initial post from my understanding, was a comment about the halting problem that was mentioned by Sam to her, and she doesn't really have a bone to pick with this tossup. So it really makes sense to only hear it from Sam, and I question why Joelle even included it in her manifesto. Sam and Joelle, this past summer one / both of you expressed to others something along the lines of Ike Jose should never be allowed to edit math again on any level after I wrote mathematics that you didn't like for NASAT. I do take offense that this was a sentiment that Joelle (and / or Sam) expressed to other writers, and while I doubt that it did do damage, because it was similar to a bedtime story of hamfisted intimidation, it's not something I appreciate -- after all I did spend one year of my life making a living entirely through writing QB questions. So, yes when either of you make a thread like this, it does feel you are trying to further undermine my work when two peole I barely know, treated me rudely last year, and told other folks that I should not be allowed to edit math ever again.

Sam asks
"Also, I'm uncertain what the relevance of the fact that you don't know him is."
, the relevance is what happened this past summer: I don't know you Sam at all. I have never met you before except maybe once or twice at a QB tournament, and you spoke to me only through posts from your significant other's account by posting using her username or having her write up harshly-worded attacks on my writing and vent to other people what I said earlier about me editing math. Again, you're someone who I barely know, and it's impossible for me to tell whether it's you or Joelle who is saying this stuff, so the idea that you would think I would be OK with this is ludicrous. But either way, if any person gives this acerbic feedback all the time through the guise of their significant other because they can't talk vis a vis, I want them to stay the :capybara: away from me, and stay the :capybara: away from every other writer. In fact my interaction with you two as happened over the summer, and somewhat in this thread, has been so toxic that for these two reasons, and only these two reasons, I recommend that no one take either of you two as a science editor. And I recommend that no new writer take either of your posts seriously.


I referenced a single question of yours that my teammate was negged on due to ambiguous language, in a post that is partly about editors being more aware of ambiguous language. If you hadn't posted, in all likelihood this thread would have carried on in the same vein as Adam's comment with no mention of you or your questions. You made this about you. And, as I noted in my apology to you about my behavior at NASAT, Sam had said absolutely nothing rude to you in his critiques of your math questions. After I read him some math people had been writing in the months leading up to the tournament (he was not going to be around to play any mirrors), he noted some problems that I thought would be helpful to pass on. He posted comments through me because he did not have access to QEMS2. Many of these suggestions were implemented without any trouble. However, when he began critiquing some of your questions in the same manner, you dismissed him and insisted on ignoring any helpful advice from anyone, including Eric. Here are the comments on your "kissing number problem" tossup:
Eric: "Dude. This is ridiculous, and neg-baitey as :capybara: . Change to sphere-packing at the very least, even that's probably too hard."
Ike: "Not sure how this is neg baiting in that sphere packing must be a percentage based answer. Also, this is taught in discrete mathematics classes, is mentioned in my high school discrete math book, I think it's a great idea!"
Me: "fyi, i just read this to Sam and he negged with sphere packing around the Leech lattice clue. Also he's never heard of this."
Me: "Sam's response: That Leech lattice clue is incredibly vague, and certainly applies to sphere packing (just check the Leech lattice wikipedia page). The leadin also seems non-unique, although I don't know enough to guarantee this. Also, I've never heard of this, and I've taken a year of graduate combinatorics classes, although I won't claim they were exhaustive."
Me: "Also, you don't know that the answer isn't a percentage until "with an answer of 240", which appears halfway through the question."
Me: "Cursory googling continues to make me believe the leadin is a hose for sphere packing, as is the fact that this problem is weirdly known in dimension 8 and 24."
Eric: "Alright, let's junk this and replace it with something else."
Ike: "That paper is unverified and just came out less than 2 months ago! It has yet to be verified and cannot be treated as correct, much like that idiot who claimed to have found a proof for P != NP"
Ike: "FWIW you can read all about this in many recreatoinal mathematics books, much like you won't get knowledge of magic squares from classes."
Cody: "for what it's worth, I am pretty familiar with this concept (and have seen the name many times); I do not see a problem with the answerline, per se"
Me (not quoting Sam): "this is a bad question with bad clues!"

Again, maybe I was harsh with that last statement, although I'd argue not nearly as harsh as Eric's first comment. But as you can see, you refused to acknowledge any problems with this question, which was really frustrating.
I think the main incident you are referencing with your complaints about NASAT was the comment string regarding Cramer's rule, which Sam and I admitted we were embarrassingly wrong about (we had interpreted your comments as claiming it was taught in "10th grade linear algebra class" since neither of us had encountered it elsewhere). I said some really vitriolic things mixed in with Sam's reasoned (if misinformed) criticism, for which I have apologized repeatedly. Other than that, the only aggressive comments I've leveled at you were in response to your vector spaces question during google docs editing. Sam wrote out detailed, non-judgmental explanations on how every single clue in that tossup was flat-out wrong and how many of them pointed, sometimes uniquely, to other answers, resulting in Cody saying he would replace that tossup and rewrite it later. At no point in that discussion did anyone say anything mean to you. Your response was to state, "This tossup is fine. Also stop being dramatic" and to resolve the comment. This was immensely frustrating when coupled with your other reactions to justified criticism, so I made a rude trio of comments ("This tossup is fine?!??!!?!!? WHAT??"; "dude, literally every clue except one is incorrect!"; "and the one that isn't is poorly worded!" (Sam later confided to me that he had been "too generous" in his assessment of that clue)) and reopened the thread. Cody rightfully called me out for being a bully, and I later apologized to you for being so unprofessional.

As for the gossip: I assume you were told about some harsh things I'd said about you in a non-main-room irc channel at some point. Other people in that room also complained heavily of your behavior when faced with any amount of criticism and have occasionally brought your other science writing up unprompted by me, as have multiple people in person and in recent PMs. You have vented about me and trashed me to other editors/writers yourself; however unlike you I have not made strictly public condemnations of you at any point. This thread was prompted by various things I had been mulling over with regards to biology, and I added in more stuff about pronouns because I knew it to be an issue in other STEM subjects as well. It was not about you, and it's pretty ridiculous to imply that I built the entire first post around a single footnote regarding a 2011 Nats question just to target you. I am not going to let you use your considerable experience and respect in quizbowl to try to intimidate me into not writing or editing anymore, and I find it pretty childish that you would try to blackball me and HSAPQ over things I said in relative privacy to people I thought would have the good sense not to spread gossip.

And with respect to Bala's comments, I still don't understand why you not knowing who he is has anything to do with whether or not he can criticize your questions. That you are taking out your frustration with me from this summer on him because we share the quality of "not being known to you" is completely unreasonable.

Ike wrote:Bala, I don't meant this to be condescending, bu you haven't edited any packet submission tournaments to know anything about what new writers do and don't do. I hope that if you're ever science editor, you receive the same tossup on packets and Code Division Multiple Access that I did, and you come back here and tell me with a straight face that this thread will be useful to those writers. That's not even talking about the actually bad questions that comprise 20% of our submissions which we chuck out quickly.
Sure. Penn Bowl CS was mostly good. Mysterium CS was excellent. Your 2015 ACF Nats bonus on complexity classes/IP/GI was good. I mentioned this in my post, but Lederberg CS and Avogadro CS were great.


No, what I'm saying is, even if they produce a flawed question, you should be thanking them for trying, especially if you want them to listen to you. A pretty good example of this was your poor wording of Sundberg's work on mathematics for Chicago Open. I'm gonna be generous and assume you accidentally got another editor irritated. My point is, people do spend a lot of time on writing questions, and unless they are being categorically unreasonable you should be more careful about this if you want them to listen to you and not just go "lol, Berkeley guy is complaining again!" Also, you only wrote up your posts in computational complexity theory. Do you really have enough knowledge to say the same for the rest of computer science? Was the anti-aliasing tossup any good from 2011 nationals, or the TU on Boltzmann Machines from 2012? So in effect you're claiming expertise in like 1/10th of computer science, and aside from the fact I think you're writing from a position of naivete like Cody Voight says, your posts do feel rather narrow, and that when you say the CS is good, you mean "the CS that I know is good**". After all this is a thread about exactitude in science writing.

Ike wrote:So I really don't know how much CS you have expertise in and am not trying to call you a bad CS student (in fact you've demonstrated that you know more about IP than I do), but just consider how much you actually know, and don't insist that a tournament's CS was excellent when there were tossups that had serious issues.

This actually gets to a point that I referenced in an early post, which is that some people might not feel confident enough in their knowledge to post a glowing review of a question out of fear that something was actually wrong with it, especially if they don't know what clues were used after they buzzed -- sometimes it also just feels arrogant to praise a question that you first-lined (I have been called out for this). What you seem to be suggesting is that everyone should thank all the writers and editors for their hard work, but not claim that any particular category was good unless you have extensive knowledge of all facets of it. I guess it's understandable to want to hear people thanking your editing/writing team, but to me it means a lot more when particular categories or questions I've written are praised rather than just every single post being "thank you for writing a tournament".

cheynem wrote:I don't know anything about science, but to be honest, more questions are slightly ambiguous/vague/slightly wrong/misleading than you might suspect. Usually a couple times a tournament I will think to myself "Hmm, is that the best way of describing that history thing?" or "Well, that actually really applied to everyone in the administration." In some way, sure, these questions could be made better; on the other hand, I just want to point out that this is not unique to science questions. I cast no opinion on this thread and its potential usefulness.

From what I can tell, and from what other people have said about their personal experiences with some of those questions, the issues extended beyond being "slightly" anything and in fact were executed in a similar way to saying "Haydn wrote a concerto for this instrument" in an early clue, or [insert blatantly false clue that points to a different answer]. Now, I'm not saying that many of the examples given are that egregious; that level of absurd ambiguity is not actually pernicious in math/CS AFAIK. But a good proportion suffer from issues that very well could neg people knowledgeable in the subject. I also didn't claim that errors don't appear in other subjects -- I just noted that those things are frequently pointed out in discussions and there isn't a symmetric population of players with deep enough knowledge of math/CS to bring things up every time an issue appears.
Joelle Smart
Ellensburg High School, 2006–10
University of Washington, 2010–14
Rutgers University, 2015–20??
President of Quizbowl at UW, 2012–2014
PACE
HSAPQ biology editor

It is a mistake to think you can solve any major problems just with potatoes.
User avatar
Amizda Calyx
Forums Staff: Moderator
 
Posts: 236
Joined: Tue May 18, 2010 9:46 pm
Location: Seattle, WA

Re: On Exactitude in Science Writing

Postby Auks Ran Ova » Fri Dec 09, 2016 6:12 pm

This thread has drifted far out of the realm of productive discourse and into areas that would be much more ideally resolved through private discussion. I'm locking it, at least for now.
Rob Carson
University of Minnesota '11, MCTC '??
Member, ACF
Member, PACE
Writer and Editor, NAQT
User avatar
Auks Ran Ova
Forums Staff: Chief Administrator
 
Posts: 3795
Joined: Sun Apr 30, 2006 10:28 pm
Location: Minneapolis


Return to Theory

Who is online

Users browsing this forum: caroline, tksaleija and 2 guests