9

Another issue has come up lately that is concerning to me and I'd like to get some feedback from the community on whether this is something we need to nip in the bud or if I'm just overreacting.

General health and diet issues are off-topic. We all know that (I hope). Of course, specific health and diet issues are certainly valid to talk about if they relate somehow to cooking or food preparation. People should absolutely feel welcome to ask about low-fat or gluten-free or vegetarian substitutions. The community seemed to feel that the question asking about cheap sources of protein was appropriate. Certainly our question about whether or not microwaving destroys nutrients in food is of some relevance to culinary experts.

But I think there's a side of this that isn't properly addressed by the FAQ, and that's the presence of dubious or at least contentious information being presented in answers and comments. Sometimes a controversial assumption can also be implicit or buried in an otherwise-legitimate question.

Several examples follow:

My point is not to call out specific people for posting these types of things, or even to attempt to definitively categorize any of these claims as right or wrong. The point is, we don't know - we're not medical experts. And when it comes to nutrition, oftentimes we find that even the medical experts are wrong (very, very wrong - does anyone remember when trans fats were being promoted as a healthier alternative to saturated fats?).

I don't believe that any of us are qualified to be making health or nutrition claims without having a truckload of citations to back them up. Nutritional claims are often more politically-motivated than people realize. Even the most innocuous claim, some idea that we've lived with for so long that we think of it as common sense, can turn out to be a gris-gris in the world of health. Most of the time, we just don't know. And by claiming to know, we are doing others a disservice.

I really believe that we should - must stay out of this territory if we want our site to be a source of good information. Dubious health claims are merely dilutive, adding virtually no value to a culinary discussion if they are correct, but potentially costing much if they're wrong.

Wikipedia has a Neutral Point Of View policy. I don't think we need to, or should, be like Wikipedia in every respect; many questions here simply cannot be answered without a certain amount of opinion or original research, and that's OK because we are supposed to be a community of culinary experts/enthusiasts who will correct or at least call out errors. So in the culinary domain, we don't need to be NPOV. But, I think, in the health domain, maybe we do.

What do other people think? Should we strive to avoid making health claims beyond those relating to food safety, or are our built-in mechanisms for dealing with bad/questionable information already sufficient? If it's the latter, why aren't they working in every case?

If we do need a policy, then what should it be, and how should we handle it?


I want to clarify this because a lot of people think "neutral" means "tell everybody's side of the story", and this is simply not true. Please refer to Wikipedia's entries on due weight; the term means including all significant viewpoints - either majority (expert) viewpoints or minority viewpoints attributable to prominent experts (not quacks). It also means not taking sides, instead just explaining the viewpoints and sources: "most experts (citation) agree on X, although experts A and B disagree, claiming Y instead."

Example of NPOV vs. Non-NPOV wording

Question: I am making (name of some baked good) and want to make a healthier version. Can anyone suggest how I can reduce the amount of oil to make this recipe healthier?

Non-NPOV Answer: Oil is fat. You should cut out all that fat completely and substitute it with something like applesauce. That will be much healthier.

Still not NPOV: You can substitute applesauce for the oil, but as long as you are using an oil that's mostly mono/polyunsaturated (such as sunflower oil) and avoiding saturated or trans fats, then it's perfectly healthy already.

NPOV Answer: You can reduce the oil by substituting some of it with applesauce, but reducing overall fat intake does not have proven health benefits. A certain amount of fat is actually necessary in one's diet. Furthermore, monounsaturated and polyunsaturated fats are both known to lower LDL cholesterol (colloquially called "bad" cholesterol), and although it is widely believed that saturated fats increase the risk of heart disease, more recent research has put that into question as well. Very little is actually known for sure about the long-term effects of dietary fat intake and you should be cognizant of this when making decisions about your diet.

Also NPOV: You can reduce the oil by substituting some of it with applesauce. (Period. No health/diet-related information presented.)

  • Given that I'm one of the negative examples and I stumbled into that one out of ignorance, I'm going with yes, this would be a great policy. – justkt Nov 15 '10 at 17:05
  • @justkt: I tried to be as non-judgmental as possible; I had to point to some examples otherwise this wouldn't be a very useful discussion. ;) – Aaronut Nov 15 '10 at 17:11
  • @Aaronut - not complaining. Just saying as someone who has done the behavior that might not be helpful, I agree. I view the policy as helpful. – justkt Nov 15 '10 at 17:20
  • The way I see it, there's still way too much bad science floating around in nutrition science, and the fact that it's pretty much only communicated to masses through tabloids does not make the situation any better. This is to say that an NPOV policy might not even be of much use. I would feel more compelled to bar it completely. I really love the comparison that Michael Pollan usually makes: "As I see it, nutrition science is kind of where surgery was in the year 1650, which is to say very interesting and promising, but do you really want to get on the table yet?" – Magnus Nordlander Nov 16 '10 at 23:47
  • @Magnus: In a sense I agree with you; that's a lot of my motivation for posting this. But I think an outright "ban" would invite claims of censorship, in addition to being difficult to enforce. I don't see anything wrong with somebody mentioning that specific experts or a majority of experts in the health field support a certain theory; after all, all science is constantly changing so it's difficult to justify drawing a line in the sand at nutrition. What's important is that people qualify statements appropriately, either as opinion or as an unproven/disputed theory. – Aaronut Nov 16 '10 at 23:52
  • Of course I would prefer that people simply not talk about health at all, but this would basically be a way to "allow" it without letting it get in the way of our real mission (to improve people's cooking). – Aaronut Nov 16 '10 at 23:54
  • Maybe it would be a good idea to simply have a don't ask policy for questions, and an NPOV-policy for answers and comments? – Magnus Nordlander Nov 17 '10 at 11:14
  • @Magnus: We do have that policy for questions; they are off-topic (see "general health and diet issues"). Or is that not what you mean? My aim here was to flesh out a policy specifically for answers and comments, and also the questions that aren't really about nutrition but either mention something nutritional in passing or assume it as part of the question's premise. – Aaronut Nov 17 '10 at 15:37
  • Well, that's sort of what I meant. However, I would propose that any question asked from a nutritional perspective would be an ill-asked question, and really should be reformulated from e.g. a cooking perspective. To take the salt question as an example, at it's core, the relevant question was 'Why does it take more kosher salt than table salt to achieve the same level of saltiness', and it probably would be better to reformulate those kinds of questions into a cooking perspective... – Magnus Nordlander Nov 17 '10 at 17:47
  • As for comments and answers, unless the question specifically solicits some form of health-related answer (which it probably shouldn't), there are probably few reasonable scenarios where a health-related answer would be justified. It's not like people are going to ask for a lot of cooking advice about death caps, or anything like that. – Magnus Nordlander Nov 17 '10 at 17:49
  • 1
    @Magnus: I don't think that your reformulation of the salt question really addresses the issue being asked about. A NPOV version of the question would be "My understanding is that kosher salt can be used as a substitute for table salt but requires a larger quantity to achieve the same result. Does making this substitution mean that I am raising my overall sodium intake?" That is a neutral question which makes no claims about whether or not salt is "good" or "bad" for you, it just sticks to the facts - but it is, in essence, still health-related. – Aaronut Nov 17 '10 at 18:01
  • The problem is @Magnus, if we rephrase it so drastically (to "why does it take more kosher salt than table salt...?") then the question is just going to be duplicated over and over again over time, because that's not the question people are actually asking. So I don't think it serves anyone's best interests to erase every health reference in a gestapo-like fashion; better to ensure that health-related information, when relevant, is disseminated responsibly and without bias. – Aaronut Nov 17 '10 at 18:02
  • -1 I disagree fundementally on your assumption that NPOV is the way to go. Cooking is fundementally linked to eating, which in turn is regulated by the belief system of the individual, both eating and your beliefs are cornerstones of the "way you cook". you simply cannot realistically seperate the two. – Anonymous Type Dec 8 '10 at 22:49
  • @Anonymous Type: That is complete and utter nonsense. While there's always some subjectivity and variability in cooking, there is a ton of established science and technique around it, and SE sites survive on their ability to provide quality, verifiable answers, not whatever some guy on the internet happens to believe. This is not a site for beliefs, it's a site for facts, and if you can't separate the two then you're going to have a hard time fitting in. There is no professional cook on the planet who would agree with what you just said. – Aaronut Dec 8 '10 at 23:38
  • 1
    Haha. @Anonymous, it seems like almost everybody who has opened this thread has interpreted my intent differently. All I'm really after is to establish this community at being good at what it does, and the best way to accomplish that is to focus what we do. If somebody here feels that they have more to offer than just cooking advice, then that's awesome, but since the rest of us are probably not experts on that side topic, it's only fair if the answer is put together in a way that facilitates fact-checking. – Aaronut Dec 9 '10 at 2:27
5

In the case of the answer I posted that was referenced (again, not complaining - it's a great example and since I lived it I can remember it clearly), I was downvoted, but nothing was flagged. Comments were left; I discussed them with the commenters; I tried clarifying links. At the end of the day I saw that the commenters were likely right.

This didn't lead to a neutral point of view, though. It lead to accepting the most common scientific interpretations today. The comment discussion set that tone. If the comments had been towards NPOV, I would have edited my answer that way.

So the biggest concern I'd have with doing this is helping the newer folks to understand the idea. One thought would be to involve the SE folks and have them add a feature that does one of the pop-ups when typing an answer (as with the questions that look subjective pop-up) if someone uses "health" and "bad" or other trigger words. This would be the high tech and it scales solution. While it may trigger inadvertently on things it doesn't need to, if worded correctly it could be a good introduction/reminder to the NPOV policy.

A low-tech solution is just to pass it on by word of mouth and use comments to clarify, mixed with strategic downvotes as necessary (if the author doesn't fix things, I think it is fair to downvote).

  • Out of the two proposed solutions, I think I'd prefer the low-tech one at our current scale, and I'd add editing to the list of possible actions for privileged users, as long as it's not getting into an edit war. By the way, NPOV does not actually mean giving all viewpoints equal weight, it means giving all viewpoints their due weight according to supporting experts and evidence. So in essence, "accepting the most common scientific interpretations today" with perhaps a minor nod to alternative ideas is exactly what NPOV is. – Aaronut Nov 16 '10 at 18:46
  • @Aaronut - sorry, I originally misunderstood the idea as simply stating no opinion on health benefits or the lack thereof. – justkt Nov 16 '10 at 19:19
  • Well, that is certainly OK too. I guess to be really specific, NPOV means that if you are going to make any claim at all about an unclear subject, then it should state all relevant positions and give them each due weight. Not making the claim at all is a very easy way around that (and probably the better way in most cases). – Aaronut Nov 16 '10 at 19:22
  • +1 down vote and comment. I had this issue with someone suggesting eating scallops of dubious origin raw (and asked about it on meta). Down voting and a comment led to more down voting. The author appears to have deleted his answer, but I can't tell since I lost my 10k rep abilities :( – yossarian Nov 16 '10 at 20:19
  • 1
    @Aaronut I believe that due weight according to supporting experts and evidence would be difficult to achieve in practice due to the unusual lack of scientific consensus regarding most things nutritional. Take dietary fat for example, which ranges from very bad for you to very good for you, depending on who you ask. Both sides are able to produce many reputable nutrition scientists and studies to support their positions. This diametrical opposition is entirely typical of nutrition science. – Magnus Nordlander Nov 17 '10 at 20:34
  • 1
    @Magnus: Yes, reputable experts disagree on many things, that is the point of NPOV. I don't see how due weight is difficult; if you are going to make a claim about dietary fat, it simply means that you need to state that reputable experts disagree. – Aaronut Nov 17 '10 at 22:48
  • @Aaronut: Usually within scientific fields there is a broad scientific consensus as a base. Not so in nutritional science where theories are usually diametrically opposed. Paleontologists might be arguing about whether it was a comet, viruses or something else that killed the dinosaurs. Rarely will you find paleontologists arguing about whether the dinosaurs did in fact die. Because the statement is so utterly pointless, moderating people to say 'fat may either be very bad for you, or very good for you, experts disagree' may prove to be harder than moderating them to say nothing at all. – Magnus Nordlander Nov 17 '10 at 23:10
  • 1
    @Magnus: As I've explained to justkt, not saying anything at all is one option and generally the preferred option for NPOV. But it's also very easy to focus on areas where experts disagree without recognizing the many areas where they do agree; they agree on the basic elements of nutrition (even if they don't agree on exactly how to balance them), they agree on the importance of many vitamins and minerals and which foods contain them, they mostly agree on the basic concepts around cholesterol, antioxidants, amino acids, etc. A lot of this is just ordinary chemistry and anatomy. – Aaronut Nov 17 '10 at 23:25
1

How would you go about enforcing this? I agree that it would be best to avoid questions and answers that imply unsubstantiable opinions. Often though the undesirable implication is tangential to the actual answer. Simply downvoting or closing such an content seems inappropriate as the point of the content might be valuable. It seems like editing out undesirable content might confuse/offend people.

  • That is precisely the question I am aiming to answer: If we do need a policy, then what should it be, and how should we handle it? I'm open to suggestions; if nobody can come up with anything and/or all of the proposed solutions are too difficult to manage or prone to abuse, then I guess we don't do it. Still, I'm confident that there is a "low-tech" solution that will work at our current scale, even if it gets problematic if/when we have millions of pageviews. – Aaronut Nov 15 '10 at 21:47
  • 1
    The only solution I can think of regarding enforcement is editing. We can start with asking the OP to edit their question/answer to be in line with this policy. The simplest edit is simply to remove the parts in question. If the OP refuses then that leaves either editing it for them, or letting it stand. Personally, I tend to down-vote these types of answers until they are improved. – hobodave Nov 23 '10 at 23:20
  • @hobodave, yeah depending on what you are trying to enforce (how "doable it is) then this is the right way to implement the enforcement. Let users learn to edit questions themselves. only moderate if you find a repeat offender. – Anonymous Type Dec 9 '10 at 1:29
1

I've been thinking about it, and well, I guess a lot of it depends on what you consider 'neutral'. For instance, newspapers try to give equal time to each side of a story ... but when it's a couple nut-jobs against a well-established scientific theory that's stood for hundreds of years, I really don't think it's worth being 'neutral'.

Now, sometimes, that one outlyer has good points, and it's worth mentioning ... if the community doesn't like the response, they'll vote the person down ... but in my opinion, much of what makes a good question is 'did you answer the question?'.

Take for instance the question, 'What are some of the benefits of electric stoves versus gas stoves?'. yossarian's answer, although top rated, completely did not answer the question, and gave the exact opposite of what was asked for. All his answer really showed is that they're different, and that trying to cook on an electric stove exatly as you would a gas stove will lead to problems. I'm going to vote this sort of answer down.

Likewise, if someone asks a queston like 'how can I make (something) more juicy', and someone goes off about how you shouldn't be eating that, because it's bad for you, I'm going to do the same. If someone asks about reducing fat, answering that not all fats are bad belongs in a comment, not an answer. (in that case, it wasn't preachy enough for me to down-vote, and I probably wasn't in a bad mood when I saw it, which may affect how likely I am to down-vote things).

...

I'd say there's no real reason to need a neutral point of view on things ... if you're getting into medical/nutrition territory, it's probably worth adding a disclaimer, but there's lots of different 'healthful' (god, I hate that word) lifestyles that include lots of meat (eg, Atkins), lots of oil (many mediteranean lifestyles), msg, salt, etc. The important thing is not to overdo anything.

  • Joe - based on your first paragraph, it sounds as though you're relying on the same (incorrect) definition of neutral that a few other people are. Neutral doesn't mean you give equal time or prominence to opposing views no matter how obscure or wacky they are; it means that you give due weight to all important viewpoints. Due meaning "according to the actual proportion of experts that agree." When getting into something really general like diets, there are so many viewpoints with so few real experts that it's usually better not to bring it up at all. – Aaronut Nov 23 '10 at 15:25
  • On the flip side, I do agree with the examples you're pointing to as downvote candidates. We clearly agree on how this information should not be presented, I just think you're misinterpreting how I think it should be presented (when it's necessary or relevant at all). – Aaronut Nov 23 '10 at 16:17

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .