This isn't by any means intended to be an exhaustive guide on how to use this site. It is hoped that the site will be relatively intuitive for the majority of users, obviating the need for an entire manual. If something is very important to know then I will generally have included some instructions or tips at the appropriate place in the user interface. In this sense, some of the site is somewhat self documenting.

The site can certainly be useful to those who don't create an account, but without logging in one is limited to doing one-off searches and trying to build meals that will disappear once you leave the site. This is good and well and may be all that many people want. This site will perhaps be most useful to those who want to keep a food diary, and in order to do this you must login.

The notion is for a user to build up a collection of saved foods and meals, and then to use those as the building blocks for keeping an accurate food diary. The database has information on thousands of foods, but it takes time to search for them individually. It seems to me that most people generally tend to eat the same things over and over again, with some variety added in here and there. So rather than searching for those repeated foods and meals each day in order to add them to a diary, you can simply save a particular item and then with about two clicks it can be added to a diary at any time in the future.

Above, I mention the concept of a "meal." In the context of this database, a meal is just a collection of an arbitrary number of individual foods. Once a meal has been created, you can view the entire nutritional profile for all of the foods together. Logically, any meal, dish, or dinner is just a combination of a number of ingredients. Since we typically eat lots of foods in combinations, the ability to create and add a meal to a diary makes sense. Foods can be added to or taken away from an existing meal at any time, and existing meals can be used as templates for creating new, similar meals.

One legitimate complaint that is sure to arise is that not all of the foods you eat can be located in the database. To be sure, it impossible to have every possible food in the database, to say nothing of every processed, branded food. However, the USDA data has quite a large selection of the most common foods. What I have found, is that while a particular brand-name food might not be in the database, there will generally be a generic food of more or less equivalent value.

As an example, let's say you wanted to find the nutrition content for 1 pint of Ben & Jerry's ice cream. At the moment that brand is not in the database, though there are many other types of ice cream in the system. Because the system allows you to name your foods and meals, you could simply find a similar ice cream and then adjust the quantity such that the resulting nutrition information matches the brand-name product as closely as possible. What this essentially does is allow you to alias foods in the database with names that make more sense to you, or names of brand-name products.

I spent a lot of time thinking about how to handle cases such as this. Some sites allow you to enter foods into their database, pending a review by someone at the site before the food is actually entered. This is great, but has the major flaw that it is impossible to verify all the data, which process takes time and resources and introduces major problems of scalability. By allowing all users to be able to enter nutritional data on brand-name products into the database you can be nearly sure that a certain percentage of the data will be incorrect or incomplete. So, in the end I decided it was best to rely on data that I felt I could trust at all times.

Foods and meals that users find and save are searchable by everyone. What this means is that the more people use the site, the more foods and meals will be searchable by everyone. Though you cannot guarantee that the name someone gave a particular food will accurately represent the food itself, what can be counted upon is the accuracy of the data, and in any case, next to every aliased (i.e. saved) food will appear the original name the USDA gave to it, so it will be virtually impossible for someone to maliciously load the database will inconspicuously fraudulent data.