The term <specific classification> is shown in italics as a reminder that it is not an actual tag and never appears in an interchange file but, instead, is a placeholder for a series of individual elements.

Description

Both start-tag and end-tag are required for all of these elements. The content will vary from one to the next, and will be specified by individual element registrations. The content may not contain the character "<" and should not contain either "/" or ">".

Elements of This Type and Their Registration

The criteria for the registration of a classification system are as follows:

• The system is in use, or is soon to be in use, in more than one country or other administrative unit.
• The system has a clear and stable definition. The definition document must be readily obtainable and should either be on file with the INFOODS Secretariat (with permission to reproduce if other sources become unavailable) or be published in widely available literature (e.g., in a book or journal with broad circulation).

The requirement for a stable definition is important because automatic comparisons among categories require that the coding be precisely known: accidental comparisons between different versions of a system can lead to significant errors. Consequently, systems that are still evolving in ways that might change older definitions will be registered only with version identification.

The following systems meet the above criteria, and have been registered prior to the publication of this document:

<EUROCODE2> The "Eurocode 2" system, as described by Arab et al. [1]. The format of the element is <EUROCODE2> code sequence </EUROCODE2>.

For example, a cola drink would be
<EUROCODE2> 12.10 </EUROCODE2>

<FDA-FFV 8807> The version of Langual (formerly the Factored Food Vocabulary) released for study and comment in July 1988 and documented in reference 13. Other versions of Langual can be registered as needed, upon the submission of comprehensive definitions.

Additional elements may be registered as needed, independent of general revisions to the description system. Since these elements will, in general, contain informal text, the requirements for registration will be a demonstration of need for additional ones and a description of the content.

The term <food description> is shown in italics as a reminder that it is not an actual tag and never appears in an interchange file but, instead, is a placeholder for a series of individual elements, which are listed below.

Description

Each of these elements requires both start-tag and end-tag. The content will, in general, consist of unformatted text, unless a different format is specified below or in the description system. <Cmt/> elements can be used as needed.

In particular, many of the elements require keywords from the open-ended keyword list associated with the description system. If additional terms are used, they should be supplemented with <cmt/> elements.

Generic Identifiers and Descriptive Notes

The "section" listed below refer to the documentation of the INFOODS description system [32, Table 2, pp. 25 ff]. That discussion contains information about the use, format, and content of each of the elements listed below. The information below constitutes supplemental material to assist in translating between the description system documentation and the corresponding interchange system elements.

Section A: Source of food, names, descriptive terms.

<fdsource/> Source and sampling information specific to the food. Information related to the data base, laboratory, etc. should be supplied as elements subsidiary to <header>. In particular, the elements subsidiary to <source> may appear here if needed.

<fdagg/> Description of how aggregate foods were aggregated. The content is unformatted text.

Section B.: Name and identification of food

This section is used to specify the information that is usually thought of as "names of foods". These names are specified subsidiary to <classif> using the <bvname/> or <exname/> elements. In other words, any <bvname/> or <exname/> elements appearing immediately subsidiary to <classif> are considered to be food names. If desired, <cmt/> elements can be used with either of these elements to identify a particular type of name being used. Similarly, food groups and codes used in particular countries or regions are reported through the <specific classification> elements. Food names or groups may also be reflected in the <srcdbid> element, depending on how the original data base is organized.

<obtain-area/> Area or country where the food is purchased. The content of this element should preferably be an "Alpha-2" country identifier chosen from ISO 3166; see the discussion of the <country> element for further information.

Section C: Description of "single" foods

<origin/> The food source as a whole animal or plant.
<taxonm/> Taxonomic or scientific name. The content is unformatted text.
<variety/> Variety of source (origin).
<part/> Part of plant or animal.
<origin-area/> Country or area of origin. See <obtain-area/>, above; the two contents should be identical in format and usage.
<manfctr/> Identification of the manufacturer of a food. The content should contain only elements, one of which is <addr/> and another of which may be <batch>, with no end-tag, to list the batch or lot number.
<ingred/> Listing of ingredients as indicated for the description system. The different ingredients should be separated by the special delimiter <->. These are "minor ingredients" if the conventions of this section apply, i.e., if food source and scientific name are listed. Or, if the food is "multi-ingredient" (section D), the content of <ingred/> may consist of either unformatted text or elements such as those in this section to describe each ingredient.
<processing/> Processing or preparation information, including locations, listed using the open-ended keyword system.
<preserv/> Preservation method, using the open-ended keyword system.
<degree-cook/> Degree of cooking, using the open-ended keyword system.
<cond-prod/> Conditions of production.
<matur/> Maturity or ripeness, using the open-ended keyword system.
<stor/> Storage conditions, including length of storage, etc.
<grade/> Grade of food.
<contain/> Container or contact surface.
<phystate/> Physical state, shape, or form.
<colour/> Colour of food.
<photoref/> Location of photograph. The <ref/> element should be used subsidiary to this one if it is appropriate to refer to published material.

Section D: Description of "mixed" foods

<ingred/> Ingredients, see above.
<recipe-proced/> Recipe procedure.
<recipe-place/> Place processed or prepared.
<photoref/> Location of photograph, as discussed above.
<manfctr/> Identification of the manufacturer of a food. As above.
<contain/> Container or contact surface, as above.
<preserv/>Preservation method, as above.
<stor/> Storage conditions, as above.
<fnlprep/> Final preparation.

Section E: Customary uses of food

<portion/> Information about typical portion. See also <meas/>. Unlike the <meas/> element, the description system calls for identifying the portion, not just classifying it into a small number of categories.
<avail/> Availability, frequency, and season of consumption.
<dietplace/> Place in diet.
<fduser/> Food users.
<fdpurpose/> Purpose of food.

Section F: Sampling and laboratory handling of food

<smpldate/> Date of collection. The format should be identical to that for <date>.
<smplwght/> Weight of sample.
<edible/> Description of edible portion. See also <meas/>.
<refuse/> Description of refuse. See also <meas/> and <refuse/>. This is an example of the same generic identifier defining a different element (with different content) because it appears in a different context.
<smplcollect/> Place of collection.
<smplhand/> Handling of sample. Content consists of elements, each of which requires both a start-tag and an end-tag: <suppl-lab/> (between supplier and laboratory), <lab-arvl/> (on arrival at laboratory), <lab-strg/> (laboratory storage and handling).
<anal-stray/> Strategy for analysis.
<anal-ran/> Reason for performing analysis.

Example

<classif>
<bvname> Fried calf liver </bvname> <obtain-area/> au </obtain-area/>
<origin/> beef </origin/> <taxonm/> bos taurus </taxonm/>
<processing/> raw </processing/>
<cond-prod/> free range </cond-prod/>
<smpldate/> 1990.03.15 </smpldate/>
</classif>

This example is derived from several of the partial examples provided by Truswell et al. [32]

Description

Both start-tag and end-tag are required. The content is precisely the same as that for the <comp> item.

Format

The content of <fddflt> has precisely the same structure as a <comp> element (i.e., that of a <specific component>), except that all subsidiary elements are optional; the <fddflt> itself must be omitted if all immediately subsidiary elements are omitted. A data value occurring in a subsidiary of <fddflt> is determined for the entire food and the corresponding value must not occur in the same position in any other subsidiary element of the same <food>. In addition, it must not correspond to a value in the <dflt> element.

<Fddflt> is most often used to specify that a food is being reported in common household measures or as-purchased quantities, e.g., with <meas/> as its significant subelement.

The details of application and use of this element are discussed in Chapter 3.

Example

<food>
<classif> classification elements </classif>
<fddflt>
<meas/> fruit </meas/>
</fddflt>
specific component information

The above example would imply that all values for this particular food are "per fruit", rather than "per 100 grams edible portion". To specify both "per 100 grams edible portion" and "per fruit" as separate entries, the special delimiter <-> would be used both with this element and in the various specific component elements, e.g.,

<food>
<classif>
<ifri> .... </ifri>
<usda...> .... </usda...>
</classif>
<fddflt> * <-> <meas/> piece <cmt/> one fruit </cmt/> </meas/> </fddflt>
<comp>
<water> 35 <-> 7 </water>
<fat> 10 <-> 2.25 c/fat>
</comp>
</food>

The above is equivalent to writing:

<food>
<classif>
<ifri> .... </ifri>
<usda...> .... </usda...>
</classif>
<comp>
<water> 35 <->
7 <meas/> piece <cmt/> one fruit </cmt/> </meas/> </water> <fat> 10 <->
2.25 <meas/> piece <cmt/> one fruit </cmt/> </meas/> </fat>
</comp>
</food>

The term <specific component> is shown in italics as a reminder that it is not an actual tag and never appears in an interchange file but, instead, is a placeholder for a series of individual elements.

Description

Both start-tag and end-tag are required. The content of <specific component> generally consists of one required formatted data item (the data value, a numeral or a "missing" or "trace" indication) and any associated information (which may be component-specific numerals, keywords, and/or subsidiary elements), along with optional immediate subsidiaries selected from the following extensible list (the numbers in parentheses indicate the page numbers on which the element descriptions begin):

and the elements listed in this document as <data description>.

This single-data-value-and-associated-information content block may be repeated in form, with separate blocks separated by the special tag <->.

The actual content of a particular <specific component> will be as specified in the element's registered description, as maintained by the appropriate registration authority, but will always include the list above, possibly extended by the registration authority for this "generic element".

Format

Each <specific component> contains data about a particular component of a particular food (that food whose data is in the <food> element to which this element is subsidiary).

More than one measurement may be available for a single component of a single food record. Typically, the second or subsequent measurements would represent different statistical estimates (e.g., a median rather than a mean) or different reference quantities (e.g., household units, or food as purchased, rather than a measure per 100 grams edible portion). When this situation occurs, the entire content of the element may be repeated for each additional value, using the special tag <-> as discussed under "Repeated and Counted Elements" in Chapter 3.

If any single component's or derived component's element contains such a repeated content, then all of the components and derived components should contain the same number of repeated contents. There is one exception to this rule: if fewer contents appear for some components than others, the missing ones will be treated as if they were present but empty (e.g., "<a> 1 <-> 2 </a>" and "<a> 1 <-> 2 <-> <-> </a>" are treated as equivalent). The empty or missing repetitions have the same significance that would be implied if the element for the component were completely omitted (there was no other value set being reported for the food in question).

As mentioned above, the actual content of any particular <specific portent> will be as specified in that element's registered description, as maintained by the appropriate registration authority. However, the content should be thought of as consisting of two parts: (i) the required data item and any food-component-specific numerals, keywords, and/or subsidiary elements that are needed to qualify it, and (ii) the collection of optional elements that specify non-standard units or quantities measured, and statistical and sampling descriptions. This second part is optional for all <specific component> elements, and is not listed with their registered descriptions. The list of such subsidiary elements given above may be extended by its registration authority to provide for additional per-component information, but a program processing the interchange file may ignore any of the second-part elements which it does not recognize.

The term <specific derived component> is shown in italics as a reminder that it never appears in an interchange file but, instead, is a placeholder for a series of individual elements.

Description

Both start-tag and end-tag are required. The content of <specific derived component> generally consists of one required formatted data item (the data value, a numeral) and any associated information (which may be component-specific numerals, keywords, and/or subsidiary elements), along with optional immediate subsidiaries selected from the <srcfri/> or <srcorg/> elements, applicable <data description> elements, and any additional elements that may be registered in the future.

<Data description>, as used here, refers to the collection of generic identifiers that can be used to describe the statistical and related properties of the data value, e.g., what statistic is being reported, how the value is distributed, and any available information about accuracy or precision.

This single-data-value-and-associated-information content block may be repeated in form, with separate blocks separated by the special tag <->.

The actual content of a specific <specific derived component> will be as specified in the element's registered description, as maintained by the appropriate registration authority, but will always include the list above, possibly extended by the registration authority for this "generic element".

Format

Each <specific derived component> contains data about a particular derived component of a particular food (that food whose data is in the <food> element to which this element is subsidiary).

More than one measurement may be available for a single component of a single food record. Typically, the second or subsequent measurements would represent different statistical estimates (e.g., a regulatory minimum rather than an analysed mean). When this situation occurs, the entire content of the element may be repeated for each additional value using the special tag <-> as discussed with <specific component>.

In the special case of a repetition to permit <specific derived component> values for different units or measurement quantities, the corresponding <specific derived component> repetitions need only be given once, with the remaining (counting from the left) empty omitted repetitions assumed to have the same value. This is because derived components have "units" that do not involve the raw quantities sampled or the raw component amounts determined by experiment.

The <unit/> and <meas/> subelements are prohibited as subsidiaries here for the same reason.

As mentioned above, the actual content of any particular <specific derived component> will be as specified in that element's registered description, as maintained by the appropriate registration authority. However, the content should be thought of as consisting of two parts: (i) the required data item and any food-component-specific numerals, keywords, and/or subsidiary elements that are needed to qualify it, and (ii) the collection of optional elements that specify statistical and sampling descriptions. This second part is optional for all <specific derived component> elements, and is not listed with their registered descriptions. The list of such subsidiary elements given above may be extended by its registration authority to provide for additional per-component information, but a program processing the interchange file may ignore any of the second-part elements which it does not recognize.