Take milk into account for cereal nutritions

[teolemon]

Add a mode to add milk+cereals vs cereals

https://world.openfoodfacts.org/product/7613035144699/cereales-au-chocolat-nesquik-nestle

[hangy]

I think those values are pretty bogus and probably don't need to be added to OFF at all. The benefit of having this additional data seems to be marginal at best. They don't even know which milk you will be using, so I never use that value when tracking my nutrition. 😄

[teolemon]

This is related to the Nutriscore. Some stuff needs either milk, or water, or added eggs (like cake mixes) to be edible. Thus they get a B while they should be C or D.

[hangy]

OK, but for that case, cereals with milk are probably a bad example. Not only is the score of milk as good as or better than the cereal, but also the portion size on most cereals is just ridiculous (30 g 😆) and does not scale at all (you don't need 250 ml of milk to eat 60 g of cereals).

Anyways, there are several problems that we might need to consider for this to be implemented.

• Similar products might use different ways of suggested cooking/serving instructions (more sugar, less wheat, …), which may lead to different Nutriscores. This means that there would need to be multiple Nutriscores for a product - at least one for the "raw" product and one for the additional serving suggestion. That could be really important, so that the "raw" products can still be compared by their score.
• I've seen products that have several suggested serving sizes in their nutrition table. For example, there was product+skim milk or product+whole milk nutrition information. Similarly, I'm pretty sure I've seen a product that had one column for adding the producer's soy drink and another column for any skim milk. Which would mean that we would need 1..n nutrition tables, and 1..n nutriscores for each product.
• If we have multiple nutriscores, which score would be the main one to be displayed/graphed/colour coded?

I'm probably missing something, but to me this feels like a lot of effort for little gain. ❔

[stephanegigandet]

For the Nutriscore, only products that cannot possibly be eaten as sold need to have the Nutriscore computation done on the prepared product. E.g. powdered milk, soup etc.

The guidelines mention that the score for breakfast cereals should be computed without milk.

A few weird things: Nutriscore for dry pasta is computed on uncooked pasta. Same for raw pie dough.