The WikiGullet Project

[xxchef]

OK. I thought maybe it looked at the wiki pages I'd been visiting and tried to guess what I'd be interested in, or maybe it leaned toward most recently updated pages, or something less-than-random. Earlier, in about 30 random page tries I saw "capocola" 3 times, "Brian Polcyn" 2 times, "Char Sui Bao" 2 times and "waffle" 4 times. It was just weird!

Just tried it again and it seemed much more diverse this time (only got "waffle" twice!).

Thanks.

[Chris Hennes]

Maybe the wiki just likes waffles. I mean really, who could blame it?

[MaxH]

I had some basic, maybe philosophical, questions. I haven't spotted these addressed in the discussion or in the project's intro sections that I checked.

1) Wikipedia of course aims to be a collaborative encyclopedia too, and has food content. What's the vision here on the relationship, or contrast, of WikiGullet to Wikipedia? On strictly food topics, what's the objective argument for someone to choose to post on WikiGullet rather than Wikipedia?

2) For food topics, Wikipedia is somewhat notorious, including Here on eGullet, for misinformation. It's good at conveying what its contributors believe. Yet without violating any guidelines, people have been willing to post notions there without undertaking even basic fact-checks in easy reliable sources. That leads to misinformation that was easily avoidable, but can endure for years and even spread (when quoted) until someone else catches it. Any thoughts on avoiding that Wiki-weakness here?

[JAZ]

Good questions, Max. Thanks for bringing them up.

1) Certainly there will be crossover between Wikipedia and the WikiGullet project, but we envision WikiGullet as completely focused on the culinary aspects of its subjects. For our intended audience, Wikipedia is good at general information about ingredients, and terribly weak in providing information on how to actually cook with them. Due to Wikipedia's broad focus, someone who is looking for culinary detail often has to wade through paragraph after paragraph of less pertinent information before getting to what he or she is actually interested in. That's an area where we expect our project to excel -- on the instructional side.

2) The nature of a collaborative project like ours (or Wikipedia) makes it vulnerable to misinformation, but we see two factors that we hope will minimize the problem. First is the existence of eG Forums, which is the perfect place for advancing theories and getting feedback on all kinds of culinary topics. Our members are never shy about arguing facts and working toward the most accurate answers possible. We trust that the "back and forth" that occurs in our discussion forums will result in more factual entries in the WikiGullet project -- both because we expect that our members will discuss questions and issues before posting articles in WikiGullet, and because our sharp membership will be constantly checking articles for accuracy. Second is our stance on original research, which requires that such research be subjected to informal peer review before posting in the WikiGullet Project.

I tried to be brief here, but I hope I addressed your concerns. If not, please feel free to discuss them further.

[Mjx]

Doesn't wikipedia usually have a large note at the top of any article that lacks citations, or has questionable/inaccurate information?

I have a question about digging up a citation for what is known, at least by oneself (I'm not talking about 'I think/believe/feel', but 'I know').

For example, I know from personal experience that in Denmark, many liquorice sweets containing ammonium chloride (aka salmiak) are labelled 'for adults only'; I've seen this with my own eyes, and certainly understand enough Danish for this to be very clear to me. But if I mention this in a wiki article, is it acceptable to simply note this in passing, or should I find a reference (possibly only available in Danish) that states this (when it comes to discussing the reason for this labelling, I'd definitely include a reference; I'm just talking about mentioning the existence of this labelling)?

The former option would lack a citation, but is verifiable by anyone who wishes to do so, whether they happen to be visiting Denmark, or contact one of the candy manufacturers. The latter option, digging up a citation (essentially, data mining), would involve documentation, but would also imply that my knowledge is only second-hand (not to mention the fact that citations aren't implicitly accurate, and verifying them requires as much work as verifying a personal statement).

How does this work? I can't consider my reading food labels (or personal observation of practices and techniques) in the category of original research, so those guidelines don't quite fit.

[xxchef]

For example, I know from personal experience that in Denmark, many liquorice sweets containing ammonium chloride (aka salmiak) are labelled 'for adults only'; I've seen this with my own eyes, and certainly understand enough Danish for this to be very clear to me. But if I mention this in a wiki article, is it acceptable to simply note this in passing, or should I find a reference (possibly only available in Danish) that states this (when it comes to discussing the reason for this labelling, I'd definitely include a reference; I'm just talking about mentioning the existence of this labelling)?

<snip>

How does this work? I can't consider my reading food labels (or personal observation of practices and techniques) in the category of original research, so those guidelines don't quite fit.

I have also wondered and looked into such statements of personal knowledge or observable facts.

My understanding is that you would be fine to say, "in Denmark, many liquorice sweets containing ammonium chloride (aka salmiak) and are labelled 'for adults only" without reference. It's a readily observable fact.

This is somewhat complicated by the time factor. What if the policy changes tomorrow and it is no longer true? To be more precise, you could say "In 1998 Denmark started labeling many liquorice sweets containing ammonium chloride (aka salmiak) 'for adults only". It would be true (if it were true) at any time, but the statement would greatly benefit from a reference.

Also, taking your original statement "In Denmark, many liquorice sweets containing ammonium chloride (aka salmiak) and are labelled 'for adults only", if you went on to say "...because ammonium chloride rots the teeth of children" (or some other claim) a reference would called for.

I'm sure someone with in official capacity will correct me as necessary.

[MaxH]

Thanks JAZ. If I understand right, focus of WikiGullet is on cooking, recipes, procedures, flavors in contrast to ingredient background and history (which is more what I see on Wikipedia). If that, or some other terse distinction, fits its creators' vision, I suggest putting it right out front and obvious, such as on the WikiGullet home page, for newcomers who may well (like me) wonder how this relates to Wikipedia.

Doesn't wikipedia usually have a large note at the top of any article that lacks citations, or has questionable/inaccurate information?

Mjx, that's what happens after someone "catches" a problem as I described. But people like The Old Foodie in Australia, and me, and others I know who read widely on food history, cookbooks, etc. and have large collections of print sources, notice far too many gaffes not yet flagged, and as the Old Foodie put it, you'd need a full-time staff to clean it up. That's where Wikipedia compares poorly to reference sources by professionals, or at least people conscious of an obligation to accuracy when representing information to the public as factual.

I have a question about digging up a citation for what is known, at least by oneself (I'm not talking about 'I think/believe/feel', but 'I know').

That distinction is at the core of accurate nonfiction reporting, which implicitly asks how fundamentally do you "know?" Basically because you can show it; but many people assert (even believe) things when they can't. Citing real evidence is a small trouble compared to the payoff in credibility.

If you're already sure the Danish information is on food packaging, you're almost "there." A little searching of trade Web sites or Danish food forums, or maybe a phone call to the nearest consulate (this is part of what consulates are for) will forever distinguish your contribution from all the heartfelt online misinformation and bar talk around.

[Chris Hennes]

To some extent this is the sort of thing we're just going to have to work out as a community of people building a wiki: obviously, a statement made with a citation to a reputable source is simply more valuable that one made based on one's own experience, no matter what the statement is. At the same time, I'd hate to see the project paralyzed by people unwilling to add what they know because they don't have the time or energy or whatever to find a source. After all, someone can easily come along later and put in references, and I certainly hope they do so. The point here is that we can work together to incrementally improve the wiki: an article does not have to spring forth full-formed, and if you don't have a reference yet, I personally hope you'll still add what you know.

[Ira Dubinsky]

I've been focusing on the extremely basic articles that need to be written. Started bread but it needs a lot of love. Also started beef which could use lots more info on beef farming, cuts, preparation methods and regional takes on beef, and I did water. I somewhat enjoy the task of describing basic things in encyclopedic-like language, especially food things! Not sure if that makes me crazy or what. I've been relying on wikipedia quite heavily and just re-wording to make it more culinary.

The cheese article is excellent, btw.

Cheers

Ira

[Mjx]

. . . .

Mjx, that's what happens after someone "catches" a problem as I described. But people like The Old Foodie in Australia, and me, and others I know who read widely on food history, cookbooks, etc. and have large collections of print sources, notice far too many gaffes not yet flagged, and as the Old Foodie put it, you'd need a full-time staff to clean it up. That's where Wikipedia compares poorly to reference sources by professionals, or at least people conscious of an obligation to accuracy when representing information to the public as factual.

Oh, okay: I don't use wikipedia much for itself, but for the references listed with the articles, so your expertise on this matter exceeds mine by several orders of magnitude.

. . . .

That distinction is at the core of accurate nonfiction reporting, which implicitly asks how fundamentally do you "know?" Basically because you can show it; but many people assert (even believe) things when they can't. Citing real evidence is a small trouble compared to the payoff in credibility.

If you're already sure the Danish information is on food packaging, you're almost "there." A little searching of trade Web sites or Danish food forums, or maybe a phone call to the nearest consulate (this is part of what consulates are for) will forever distinguish your contribution from all the heartfelt online misinformation and bar talk around.

When it comes to citing what lies in the territory between common awareness (e.g. US stop signs are octagonal, most bread is made of flour of some sort) and easily checked by going to the source (e.g. HP Sauce contains vinegar), citations might actually erode credibility to some extent.

Case in point: As an undergraduate, one of the assignments for my Comparative Vertebrate Anatomy class was to write up one of the dinosaurs. Since I'd been interested in animals, including dinosaurs, since I was a kid, and had an accurate and retentive memory, I knew (i.e. had internalized to the point where I had no recollection of the original source) the bulk of what I was writing about. I wrote up my essay, and then, since I have a horror of being accused of plagiarism, looked up every piece of information I'd included, and found solid, credible references for each. My bibliography was practically an almanac, and the entire things was solid.

I got a B+ on it.

The reason was that I hadn't 'contributed any of my own knowledge', I'd (apparently) 'just looked everything up'.

And it wasn't an unreasonable assessment, even though I was miserable about it; part of writing articles and essays is bringing something of one's personal knowledge to the table (otherwise, why write about that specific topic, unless it's part of your job)

I'm careful about finding references to support my statements (depressing dino essay experience notwithstanding), and cheerfully pestered a well-known author with biographical questions, to make sure I had accurate dates (online sources were looking dodgy), but just as different forms of media demand different reference formats, so (I believe) do different types of information require different levels of formality, when it comes to reference material.

Or, taking it as a question: If one were to mention the warning to not give nuts to children under six that appears on the jars of Planter's products, would phoning and citing a trade board for this information make for greater credibility than taking a picture of the statement on the label itself? I'm only talking about information of this sort, which may be confirmed (or refuted) by anyone to whom the information is relevant (e.g. here in Denmark, everyone knows about 'grownup liquorice', and has the opportunity to see the warning; I don't think you could so easily find it in the States, and would probably have to bribe the average child substantially to eat so much as one piece ).

If in doubt, though. . . yep, I'd err on the side of caution.

To some extent this is the sort of thing we're just going to have to work out as a community of people building a wiki: obviously, a statement made with a citation to a reputable source is simply more valuable that one made based on one's own experience, no matter what the statement is. At the same time, I'd hate to see the project paralyzed by people unwilling to add what they know because they don't have the time or energy or whatever to find a source. . . .

I confess to this paralysis; I really do worry that I won't have cited the most reliable source, and have been spending hours tracking down seminal sources of information, which are often government sites, then I worry that they may be inaccurate, owing to some agenda.

So, I spend a lot of time wondering what consitutes a 'best source' for a given piece/type of information (e.g. should I find documentation to support the statement that liquorice comes in colours other than black?).

Edited February 23, 2011 by Mjx (log)

[Chris Hennes]

I've been relying on wikipedia quite heavily and just re-wording to make it more culinary.

Yeah, I've found that's a great approach for getting articles started: I take the Wikipedia article (don't forget to put "Copied from Wikipedia article XXX" in your edit summary!), and then remove all the non-culinary stuff.

I've also been striving for a sort of "inverted pyramid" article style that starts by asking "What is the most likely reason a person came to this Wiki to read this article? What question are they trying to answer?" and then arranging the article to answer the most likely question first, and so on. Of course, there are many different questions that might be asked, so the whole exercise is wildly subjective, even when just stating facts!

[MaxH]

...citations might actually erode credibility to some extent. Case in point: ... I got a B+ on it. The reason was that I hadn't 'contributed any of my own knowledge', I'd (apparently) 'just looked everything up'.

A core difference between encyclopedias and research papers bears on that example.

Wikipedia is explicitly a secondary or tertiary reference, meant to collect or summarize information already public in "reliable sources." Thus citations, i.e. pointers to the existing information, are unusually important. One of Wikipedia's basic standards is "No original research."

For that reason, writing that a professor calls unoriginal could be ideal for Wikipedia (where the writer's contributions of investigation and original wording are valued).

[MaxH]

...At the same time, I'd hate to see the project paralyzed by people unwilling to add what they know because they don't have the time or energy or whatever to find a source. After all, someone can easily come along later and put in references

It's an important point, and I've been there too. FWIW, a suggestion that's easy, yet safe enough even for hardcore scholarship or journalism, takes to heart the principle of expressing as facts only what you can back up. If your evidence is a recollection or conviction, report that fact ("I have often seen Z," "Contributor xx recalls Z") rather then just representing Z as true. Add an obvious call for a reference, and you've handled this issue as constructively as I can imagine.

Bad gaffes on Wikipedia (I can even think of whole articles, though not about food) often reflect someone with a firm notion (who maybe reads only other people sharing that notion) casting it on Wikipedia as a reality, not a belief. That's already pretentious, but the problem is when notions come from something much more nebulous than Mjx's first-hand memories of experience. Worse, when easy-to-get authoritative sources contradict those notions. So Wikipedia's misinformation that I criticize here is both easy enough to avoid, and always under the authors' control.

[Chris Hennes]

I'd be pretty surprised if that was the standard the community of contributors settled on, but I've been surprised before. Frankly, I see more value in a resource that is 99% accurate and has 100k articles than one that is 99.99% accurate and has 1000.

That said: I think it's obvious that if you (or anyone else out there!) want to focus your efforts on improving the reliability of the resource by hunting down citations I think that everyone would appreciate the work done. I'd love to see a small team of people in there doing that sort of thing. Maybe those who aren't interested in writing?

[MaxH]

I don't follow you, Chris, so maybe I expressed the last point badly. It's about as easy to write, say, I recall a book on it by James Beard as to write James Beard wrote a book on it, but it's more honest. And looks better, when the book turns out to be by Richard Olney.

Writers make this choice constantly anyway, I'm only pointing it out, and it should have no effect on the number of contributions anyone makes. (The first form above will be instinctive anyway to people who've written for fact- or libel-conscious media.)

[JAZ]

Thanks JAZ. If I understand right, focus of WikiGullet is on cooking, recipes, procedures, flavors in contrast to ingredient background and history (which is more what I see on Wikipedia). If that, or some other terse distinction, fits its creators' vision, I suggest putting it right out front and obvious, such as on the WikiGullet home page, for newcomers who may well (like me) wonder how this relates to Wikipedia.

That's certainly a part of what we're looking for, but it doesn't cover everything that's appropriate for the WikiGullet Project. For instance, it says nothing about entries on restaurants, books, authors and chefs, all of which we're interested in getting.

However, it's a good suggestion to give more detailed information on our home page -- balancing brevity and detail is always a challenge, but we'll give it a try.

[Mjx]

I'd be pretty surprised if that was the standard the community of contributors settled on, but I've been surprised before. Frankly, I see more value in a resource that is 99% accurate and has 100k articles than one that is 99.99% accurate and has 1000.

Absolutely.

That said: I think it's obvious that if you (or anyone else out there!) want to focus your efforts on improving the reliability of the resource by hunting down citations I think that everyone would appreciate the work done. I'd love to see a small team of people in there doing that sort of thing. Maybe those who aren't interested in writing?

Well, if anyone wants anything cleaned up citation-wise, point me at it, I'm game. I like writing, and I like research, but when I do both, my output tends to drop, so it would be nice to do something that makes my obsessive tendencies pay off.

Incidentally, while I understand that the wiki articles are intended for anyone interested in food and its preparation, what is the baseline of knowledge/intelligence that is presumed for the average reader? With an idea of that, it's possible to better determine the point at which details and citations only deliver an increasingly diminishing return, and contribute clutter, rather than clarity.

[Chris Hennes]

Incidentally, while I understand that the wiki articles are intended for anyone interested in food and its preparation, what is the baseline of knowledge/intelligence that is presumed for the average reader? With an idea of that, it's possible to better determine the point at which details and citations only deliver an increasingly diminishing return, and contribute clutter, rather than clarity.

Perhaps one way around this is to use inline citations only when the need seems clear, and to make an "Additional references" section to cover the rest. Does that work?

Regarding articles that need citations: the charcuterie category would be a good start, there is widespread disagreement about the definitions of all that stuff. http://wiki.egullet.org/index.php?title=Category:Charcuterie

[Chris Hennes]

I don't follow you, Chris, so maybe I expressed the last point badly. It's about as easy to write, say, I recall a book on it by James Beard as to write James Beard wrote a book on it, but it's more honest. And looks better, when the book turns out to be by Richard Olney.

I personally shy away from the notion of first-person statements in this sort of writing: in a wiki article edited by ten people, who is this "I" character anyway? Maybe it would make more sense to me if I saw it in context... want to start an article?

[MaxH]

want to start an article?

Very possibly, though I'm just getting acquainted with WikiGullet, its relationship to Wikipedia, etc.

But please don't be distracted by the particular form my offhand example took with an "I." That's distinct from my point, that when giving useful or important factoids, it's no harder to acknowledge some as tentative or from memory than to pretend they're established reality. That choice is inherent in this kind of nonfiction writing, and should not warrant any fear of 100-times drop in contributions.

[Chris Hennes]

Right, that makes sense Max: as we wrote in Rule Number One:

Sometimes the jury is still out on basic facts about food, cooking, and other culinary topics. If that's truly the case, feel free to edit articles to state exactly that.

[emannths]

Hey, quick question: any guidance for handling localization? Food in particular tends to develop regional vernacular, not to mention the usual differences in spelling and grammar between British and American English.

IMO, I'd say the guidance should be to add clarifying detail (ie, describe the use of the word 'shrimp' and 'prawn' in England and the US), but to otherwise not dictate a particular style, either grammatically or cuisine-wise.

Any thoughts?

[Chris Hennes]

Letting a solution develop organically makes sense to me, but I'll let Janet weigh in with an "official" statement on the matter.

In unrelated wiki news, however, today we crossed the 500-article mark, due in part to the amazing contributions of a few enthusiastic and fearless members hammering away day and night (do you guys sleep?!): my personal thanks (in alphabetical order) to FrogPrincesse, IraDubinsky, Liuzhou, Mjx, PedroG, and Xxchef for their outstanding efforts, and to the dozens of others helping out as well. Excellent work, folks!

Anyone else want to join them in the effort? I saw someone got a good start on the spices of the world, and Xxchef has been doing amazing things in the Cheese category. PedroG is hammering away at Sous vide, of course!

I had a start at some French sauces from the classic repertoire, but of course there's always more to add!

[Mjx]

I had some questions about stubs, and articles identified as requiring cleanup: what and who determines when their status has changed? Is it something that's simply picked up during regular sweeps of these things, or should we notify someone, if we do anything substantial to a stub or article requiring cleanup?

Any general parameters for 'stub'? A few topics can be covered quite thoroughly by a few compact sentences: are they still stubs?

[liuzhou]

If you think a 'stub' article has been adequately explained, then remove the stub note. As long as you are being sensible, there is no ultimate authority.

But, if someone else feels the article could be usefully expanded, they may reinstate the stub notice.

That is the way wikis work.