Property talk:P7971

Documentation

• Start a query
• Current uses
• Statistics by class
• Units
• Scale of values
• List of qualifiers
• Count

food energy
amount of chemical energy that animals (including humans) derive from this item

Represents	food energy (Q186205)
Data type	Quantity
Domain	food (Q2095), recipe (Q219239) or type of food or dish (Q19861951)
Allowed units	kilojoule (Q4989854), kilocalorie (Q26708069), kilojoule per kilogram (Q21077849), kilojoule per gram (Q106808096), kilojoule per 100 gram (Q107213614) or kilocalorie per 100 gram (Q107883587)
Example	Big Mac (Q506796) → 540 kilocalorie Whopper (Q1135327) → 660 kilocalorie Baconator (Q4839954) → 970 kilocalorie apple (Q89) → 270 kilojoule
Source	https://fdc.nal.usda.gov/
Lists	Items with the most statements of this property Count of items by number of statements (chart) Items with the most statements Items with the fewest statements Count of items by number of sitelinks (chart) Items with the most sitelinks Items with the most sitelinks, but no link to one Wikipedia (sample: Welsh Wikipedia, cywiki) Items with the fewest sitelinks Most linked wikis Items with the most identifier properties Items with no other statements Most recently created items Items with novalue claims Items with unknown value claims Usage history (total) Usage history (main statement) Minimum and maximum Items with highest numeric value Items with lowest numeric value Database reports/Constraint violations/P7971 Map Random list
Proposal discussion	Proposal discussion
Current uses	Total 39 Main statement 35 89.7% of uses Qualifier 4 10.3% of uses

[create Create a translatable help page (preferably in English) for this property to be included here]

Type “food (Q2095), recipe (Q219239), type of food or dish (Q19861951)”: item must contain property “instance of (P31), subclass of (P279)” with classes “food (Q2095), recipe (Q219239), type of food or dish (Q19861951)” or their subclasses (defined using subclass of (P279)). (Help)

Exceptions are possible as rare values may exist. Exceptions can be specified using exception to constraint (P2303).
List of violations of this constraint: Database reports/Constraint violations/P7971#Type Q2095, Q219239, Q19861951, SPARQL

Units: “kilojoule (Q4989854), kilocalorie (Q26708069), kilojoule per kilogram (Q21077849), kilojoule per gram (Q106808096), kilojoule per 100 gram (Q107213614), kilocalorie per 100 gram (Q107883587)”: value unit must be one of listed. (Help)

Exceptions are possible as rare values may exist. Exceptions can be specified using exception to constraint (P2303).
List of violations of this constraint: Database reports/Constraint violations/P7971#Units

Citation needed: the property must have at least one reference (Help)

Exceptions are possible as rare values may exist. Exceptions can be specified using exception to constraint (P2303).
List of violations of this constraint: Database reports/Constraint violations/P7971#citation needed

Absolute vs. relative (molar) energy

From the property discussion I was not able to find a good motivation for the unit kilojoule per mole (Q752197). Moreover, a property - to be really useful - should not mix absolute and relative values. Given that all the examples list absolute values I have to assume that this property is meant to model "how much energy does the whole item provide". This is now reflected in the unit constraint (I changed kilojoule per mole (Q752197) -> kilojoule (Q4989854)). To prevent misunderstandings as to how this item should be used the description should clearly state that this is for the whole item. (A related property - which in my opinion would be even more useful - could be proposed to measure the relative food energy, that is, values measured in kJ / 100 g or kcal / 100 g.) Toni 001 (talk) 08:13, 16 March 2020 (UTC)

Differences between countries, kilocalories vs kilojoules, variations in what defines a "item" of food, generic food items, variability over time, seasonal variability, etc.

It's a shame I didn't see the property proposal for this, because I would have raised the same question that I did at Wikidata:Property_proposal/food_composition -- how do you effectively deal with the dozens/hundreds/thousands of possible values this could have for any given "food"?

• Country variability. The reported energy value of a Big Mac (Q506796) differs from country to country. That's potentially almost 200 values just from this alone. For some foods it may even differ from region to region. Different databases may give different values within the same country.
• Some countries report energy in both kilocalories (a.k.a. food Calories) and kilojoules for some or all foods, meaning potentially 2 values for each country from this.
• Since this property is for a "per-item" value, you have to define what an "item" is. Nutritional databases often have multiple definitions (by size and/or weight, skin-on vs skin-off, cooked vs raw, etc.) and therefore values for each.
  • Or there may not be a per-item value reported at all, just a per-weight and/or per-volume one -- how do you express this? Would you just take the 100g/100mL value (qualified to say so obviously)? See the "Absolute vs. relative (molar) energy" topic above.
• Some food items are just too generic. apple (Q89) is listed as a Wikidata property example (P1855), yet according to Wikipedia apples have more than 7500 known cultivars, the majority of which will have their own distinct nutritional characteristics.
• Values can and will change over time due to a variety of factors (e.g. recipes changing). Since Wikidata makes it possible, how many historical values do you retain?
• The value for some foods may be seasonal/periodic.
• Probably some other stuff I've forgotten.

Clearly all this stuff can be modeled with qualifiers. With that in mind, how does Wikidata deal with the combinatorial explosion that these variables represent? My argument is that it shouldn't be attempting to, and we should instead just be ensuring that we are linking to external nutritional databases via IDs. This is what the linked data paradigm is for. --NoInkling (talk) 22:54, 24 March 2020 (UTC)