DISTRIBUTION STATEMENT A: Approved for public release; distribution unlimited

3.14 Internationalization 3.14-

3.14-1 Coded character sets standards 3.14-

3.14 Internationalization. Internationalization is the adaptation of a computer system's interface to present data according to local conventions and to use character sets that support the local language.

NOTE: Throughout Part 14, all tables shall have abbreviations listed under the column (Standard Type) as follows:

3.14.1 Character set and data representation. A character set is a subset of all letters in different alphabets, numbers, punctuation marks, mathematical symbols, and other characters used by computers. These services include the capability to input, store, manipulate, retrieve, communicate, and present data independent of the coding scheme used.

3.14.1.1 Coded character sets. (This BSA appears in both part 5, Data Interchange, and part 14, Internationalization.) A character set is a subset of all letters in different alphabets, numbers, punctuation marks, mathematical symbols, and other characters used by computers. These services include the capability to input, store, manipulate, retrieve, communicate, and present data independent of the coding scheme used.

3.14.1.1.1 Standards. Table 3.14-1 presents standards for coded character sets.

3.14.1.1.2 Alternative specifications. Alternative character coding schemes include Encoded Binary Decimal (EBCDC) and the Macintosh character set.

3.14.1.1.3 Standards deficiencies. The greatest deficiency any of these standards have is narrow applicability to a single application or language or no standard means of translation from set to set.

3.14.1.1.4 Portability caveats. Character sets are generally portable, but there are sometimes questions about conversion between sets.

3.14.1.1.5 Related standards. The following standards are related to coded character set standards:

3.14.1.1.6 Recommendations. ISO 6937 is recommended for ordinary English-only alphabetic applications.

3.14.1.2 Seven-bit coded character sets. (This BSA appears in part 5, Data Interchange, and part 14, Internationalization.) Character sets which contain only as many characters as can be uniquely identified using a seven-bit number (i.e., 128 characters numbered 0 to 127).

3.14.1.2.1 Standards. Table 3.14-2 presents standards for seven-bit coded character sets.

ISO 646 describes a set of 128 control, alphabetic, digit, and symbol characters. It includes the use of the control characters and describes the option of national replacement characters. It is the standard that formed the basis for creating additional standards that extend the character set to include many languages. A variant, ISO 646:1991 IRV, left open an additional 128 codes to be used to represent symbols for other languages.

3.14.1.2.2 Alternative specifications. Alternative character coding schemes include Encoded Binary Decimal (EBCDC) and the Macintosh character set.

3.14.1.2.3 Standards deficiencies. The greatest deficiency any of these standards have is narrow applicability to a single application or language or no standard means of translation from set to set.

3.14.1.2.4 Portability caveats. Character sets are generally portable, but there are sometimes questions about conversion between sets. FIPS 19-2, a catalog of widely used code sets that lists and briefly describes code stes in wide use in the United States and might be used in Federal data systems, may be helpful to consult.

3.14.1.2.5 Related standards. The following standards are related to seven-bit coded character sets:

3.14.1.2.6 Recommendations. FIPS 1-2, which adopts the ASCII character set, is recommended for common applications. ISO 646 is also recommended.

3.14.1.3 Eight-bit coded character sets. (This BSA appears in part 5, Data Interchange, and part 14, Internationalization.) Character sets which contain only as many characters as can be uniquely identified using an eight-bit number (typically, 256 characters numbered 0 to 255).

3.14.1.3.1 Standards. Table 3.14-3 presents standards for eight-bit coded character sets.

3.14.1.3.2 Alternative specifications. Alternative character coding schemes include EBCDC and the Macintosh character set.

3.14.1.3.3 Standards deficiencies. The greatest deficiency any of these standards have is narrow applicability to a single application or language or no standard means of translation from set to set.

3.14.1.3.4 Portability caveats. Character sets are generally portable, but there are sometimes questions about conversion between sets.

3.14.1.3.5 Related standards. The following standards are related to eight-bit coded character sets:

3.14.1.3.6 Recommendations. ISO 4873 is recommended.

3.14.1.4 Eight-bit single byte character sets. (This BSA appears in part 5, Data Interchange, and part 14, Internationalization.) Character sets which contain only as many characters as can be uniquely identified using an eight-bit number in a single byte (typically, but not always, 256 characters numbered 0 to 255).

3.14.1.4.1 Standards. Table 3.14-4 presents standards for eight-bit single byte character sets.

ISO 8859 defines a set of 191 graphic characters with a single 8-bit byte. It uses the characters 0x20 through 0x7F to represent those used in the US-ASCII (ISO 646) set.

3.14.1.4.2 Alternative specifications. Alternative character coding schemes include EBCDC and the Macintosh character set.

3.14.1.4.3 Standards deficiencies. The greatest deficiency any of these standards have is narrow applicability to a single application or language or no standard means of translation from set to set.

3.14.1.4.4 Portability caveats. Character sets are generally portable, but there are sometimes questions about conversion between sets.

3.14.1.4.5 Related standards. The following standards are related to eight-bit single byte character sets:

a. NIST FIPS 19-2: Catalog of Widely Used Code Sets
b. Optical Character Recognition Character Code Sets

3.14.1.4.6 Recommendations. ISO 8859, parts 1-9, is recommended.

3.14.1.5 Control functions. (This BSA appears in part 5, Data Interchange and part 14, Internationalization.) This service area is for definition and coding of control functions for inclusion in character sets.

3.14.1.5.1 Standards. Table 3.14-5 presents standards for control functions.

ISO 6429 defines 7-bit, 7-bit extended, 8-bit, and 8-bit extended character set control functions.

3.14.1.5.2 Alternative specifications. There are no alternative specifications.

3.14.1.5.3 Standards deficiencies. Deficiencies in the existing standards are unknown.

3.14.1.5.4 Portability caveats. Portability problems with the existing standards are unknown.

3.14.1.5.5 Related standards. There are no related standards.

3.14.1.5.6 Recommendations. ISO 6429 is recommended.

3.14.1.6 Character set conversion. (This BSA appears in part 5, Data Interchange, and part 14, Internationalization.) Character set conversion deals with the problem of translating from one character set to another.

3.14.1.6.1 Standards. Table 3.14-6 presents standards for character set conversion.

ISO 6936 specifies conversion between the 58 character ITA2 set and the 128 character ISO 646 set.

3.14.1.6.2 Alternative specifications. There are alternative specifications that are sometimes necessary:

a. Mac to ASCII
b. EBCDC to ASCII

3.14.1.6.3 Standards deficiencies. The greatest deficiency any of these standards have is narrow applicability to a single application or language or no standard means of translation from set to set.

3.14.1.6.4 Portability caveats. Character sets are generally portable, but there are sometimes questions about conversion between sets.

3.14.1.6.5 Related standards. The following standards are related to character sets conversion:

a. Transliteration standards.

3.14.1.6.6 Recommendations. There are no recommendations. Character set conversion standards depend on which sets are involved.

3.14.1.7 Code extension techniques. (This BSA appears in part 5, Data Interchange, and part 14, Internationalization.) There is also a need to define standard techniques for expanding the number of characters represented by a character set. Switching between character sets in mid-string is done by escape sequences.

3.14.1.7.1 Standards. Table 3.14-7 presents standards for code extension techniques.

3.14.1.7.2 Alternative specifications. Alternative specifications would include other, larger, forms of character sets (8-bit instead of 7-bit, or multiple-octet sets instead of 8-bit).

3.14.1.7.3 Standards deficiencies. Deficiencies in the existing standards are unknown.

3.14.1.7.4 Portability caveats. Few systems support the ISO 2022 encoding architecture because escape sequences present difficulties to processing.

3.14.1.7.5 Related standards. There are no related standards.

3.14.1.7.6 Recommendations. ISO 2022 is recommended.

3.14.1.8 Universal character sets. (This BSA appears in part 5, Data Interchange, and part 14, Internationalization.) Universal character sets are an approach to defining the broadest possible character set. This involves using more than an 8-bit code. Use of a 16-bit code allows for a character set of 32,768 characters, which is sufficient to cover several complete alphabets, including accented letters. The object of UCS is to represent the written form of world languages unambiguously to facilitate information interchange

3.14.1.8.1 Standards. Table 3.14-8 presents standards for universal character sets.

ISO 10646 is an extension of ISO 8859. A separate part of 8859 is defined for a variety of character sets. The 10646 is multiple-octet character set that can be encoded using 8-, 16-, or 32-bit character sizes. All existing character sets in 8859 are included as pages in the 10646 encoding, along with virtually all known characters on the planet. The 10646 is effectively the dictionary of coded character sets.

Unicode is an implementation of ISO 10646 that defines a set of 16-bit characters and is not exactly a superset of 8859.

3.14.1.8.2 Alternative specifications. There are no alternatives for a universal character set.

3.14.1.8.3 Standards deficiencies. Only a small number of modern languages are unrepresentable by these satndards, but are expected to be supported soon.

3.14.1.8.4 Portability caveats. The portability problems with universal character sets involve their multi-byte nature. Translation to and from single-byte sets is full of chances for errors.

3.14.1.8.5 Related standards. There are no related standards.

3.14.1.8.6 Recommendations. If multiple-octet representations (16- or 32-bit) of characters are required, ISO 10646 is recommended.

3.14.1.9 Currency and funds representation. (This BSA appears in part 5, Data Interchange, and part 14, Internationalization.) Covers characters for and the representation of currency and monetary values.

3.14.1.9.1 Standards. Table 3.14-9 presents standards for currency and funds representation.

3.14.1.9.2 Alternative specifications. There are no alternative specifications.

3.14.1.9.3 Standards deficiencies. Deficiencies in the standard are unknown.

3.14.1.9.4 Portability caveats. Portability problems in the standard are unknown.

3.14.1.9.5 Related standards. Numerical value representation standards and internationalization locale specifications are related.

3.14.1.9.6 Recommendations. ISO 4217 is recommended.

3.14.1.10 Country name representation. (This BSA appears in part 5, Data Interchange, and part 14, Internationalization.) These standards provide for a short character combination that can be used to represent the names of countries.

3.14.1.10.1 Standards. Table 3.14-11 presents standards for country name representation.

ISO 3166 defines a 2-letter, a 3-letter, and a numeric code for each country. The 2-letter names are well-known and accepted as internet domain names. The 3-letter codes are often used in international sports.

3.14.1.10.2 Alternative specifications. Alternative specifications would include the international codes to designate the country of registration of automobiles.

3.14.1.10.3 Standards deficiencies. Deficiencies in the existing standards are unknown.

3.14.1.10.4 Portability caveats. Portability problems with the existing standards are unknown.

3.14.1.10.5 Related standards. There are no related standards.

3.14.1.10.6 Recommendations. There is no recommendation.

3.14.1.11 Representation of human sexes. This BSA concerns the uniform representation of human sexes for the interchange of information.

3.14.1.11.1 Standards. Table 3.14-12 presents standards for representation of human sexes.

3.14.1.11.2 Alternative specifications. There are no alternative specifications.

3.14.1.11.3 Standards deficiencies. ISO 5218 does not meet the requirements of specific medical or scientific applications.

3.14.1.11.4 Portability caveats. ISO 5218 does not prescribe file sequences, storage, media, programming languages, or other features of information processing to be used in its implementation.

3.14.1.11.5 Related standards. No related standards have been identified.

3.14.1.11.6 Recommendations. ISO 5218 is recommended for use.

3.14.1.12 Representation of names of languages. (This BSA appears in part 5, Data Interchange, and part 14, Internationalization.) This BSA presents standards for code to represent the names of languages.

3.14.1.12.1 Standards. Table 3.14-13 presents standards for representation of names of languages.

3.14.1.12.2 Alternative specifications. Alternative specifications may include abbreviations in common use in entomology.

3.14.1.12.3 Standards deficiencies. Deficiencies in the existing standards are unknown.

3.14.1.12.4 Portability caveats. Portability problems in the existing standards are unknown.

3.14.1.12.5 Related standards. The following standards are related to representation of names of languages:

3.14.1.12.6 Recommendations. ISO 639 is recommended.

3.14.1.13 Date and time representation. (This BSA appears in part 5, Data Interchange, and part 14, Internationalization.) Date and time representation and storage require consideration and standardization. Problems include representation of twelve or twenty-four hour time, the order in which the day and month are presented, and dropping of the century digits from the year.

3.14.1.13.1 Standard. Table 3.14-13 presents standards for date and time representation.

3.14.1.13.2 Alternative specification. There are no other available specifications.

3.14.1.13.3 Standard deficiencies. In the early days of computer technology, information storage space was at a premium. Engineers saved space by using only the last two digits of the year rather than using full four-digit year representation since they did not anticipate that existing systems would still be in operation in the year 2000. This is a problem to be kept in mind during data design for information systems and their databases. The internal representation of the year and dates is expected to cause enormous difficulties as the year 2000 arrives.

3.14.1.13.4 Portability caveats. The difference between a little-endian (i.e., 11 May 1995), a big-endian (i.e., 1995 May 11), and mixed mode (i.e., May 11, 1995) date representation can be a portability problem for systems. The stated DoD data element for date format is "YYYYMMDD" where YYYY is the year, MM is the month, and DD is the day. NIST highly recommends that four-digit year elements be used and that two-digit year elements NOT be used for data interchange. On March 25, 1996 NIST published a change notice to FIPS 4-1 that highly recommends four-digit year elements, and states that two-year elements specified in ANSI X3.30:1985 (R1991) should not be used for the purpose of any data interchange among U.S. Government agencies.

3.14.1.13.5 Related standards. The following standard is related to date and time representation:

3.14.1.13.6 Recommendations. For purposes of data interchange, DoD requires that year, month, and day be represented as 'YYYYMMDD'.

3.14.2 Cultural convention services. These services provide the capability to store and access rules and conventions for cultural entities maintained in a cultural convention repository.

3.14.2.1 Numerical value representation. (This BSA appears in part 5, Data Interchange, and part 14, Internationalization.) Numerical value representation deals with the presentation of numerical values as character strings in machine- and human- readable form.

3.14.2.1.1 Standards. Table 3.14-14 presents standards for numerical value representation.

ISO 6093 specifies three presentations of numerical values as character strings in machine-readable form for data interchange.

3.14.2.1.2 Alternative specifications. There are no alternative specifications.

3.14.2.1.3 Standards deficiencies. Deficiencies in the existing standards are unknown.

3.14.2.1.4 Portability caveats. Portability problems in the existing standards are unknown.

3.14.2.1.5 Related standards. The following standards are related to numerical value representation:

3.14.2.1.6 Recommendations. ISO 6093 is recommended.

3.14.2.2 Customization to local norms. (This BSA appears in part 3, User Interface, part 13, Human Factors, and part 14, Internationalization.) Customization to local norms involves modification of the key mapping to accommodate the local language and display of data in the commonly-used format (e.g., numbers, dates, time).

3.14.2.2.1 Standards. Table 3.14-15 presents standards for customization to local norms.

DODI 8120 mandates use of the DOD HCI Style Guide.

Motif 1.2 is the current version of the OSF specification for GUI behavior and appearance and programming and data interfaces. X11R5 is the current release of Version 11 of the X Windows standard.

3.14.2.2.2 Alternative specifications. Several applicable consortia or de facto style guides are available for internationalization. However, conformance with one or more the style guides listed below does not guarantee conformance with ergonomic standards:

3.14.2.2.3 Standards deficiencies. Currently, conformance to parts 12-17 of the ISO 9241 standard is on a part-by-part basis. There is concern that the overal standard may thus fail to address potential ergonomic problems arising from interactions between the user interface elements covered by the individual parts.

3.14.2.2.4 Portability caveats. Although Morif supports the X/Open Native Language System, it also aupports a number of its own internationalization extensions which makes it compatible with some legacy applications (e.g., OpenLook).

NIST FIPS 158-1 (User Interface Componenet of the Applications Portability Profile) mandates the use of the X Window protocol, X library, and X toolkit intrinsics. IEEE P1201.2, when completed, is intended to increase the level of user interface consistency (and thus user interface portability) across X Windows-based environments. There are potential conflicts here.

The DOD HCI Style Guide is based on (and intended to supersede) the Army, Navy, Air Force, and DODIIS Style Guides cited in the table above. The goal of this effort is to minimize unnecessary user interface diversity across DOD systems. There are potential problems with systems designed to accommodate different style guides.

3.14.2.2.5 Related standards. The following standards are related to cultural convention services:

3.14.2.2.6 Recommendations. Procurements that require software user interfaces to be addressed by ergonomic standards can require conformance with standards for menu structures, command languages, direct manipulation dialogs, forms-based dialogs, windowing, icons, screen formatting, information coding, and user guidance.

Parts 1 and 2 of the ISO 9241 standard are informative; parts 10 and 11 are expected to be informative on completion. Part 3 of the ISO 9241 standard is normative; parts 2-9 and 12-17 are expected to be normative on completion. Conformance with the overall ISO 9241 standard is based on conformance with all normative parts that apply to a particular product.

Procurements must recognize the difference between informative and normative parts of the standard in question. Where possible, both the informative and normative parts should be required for the best implementation of modern human factors/ergonomic thinking. In general, conformance tests for informative parts will not be available.

The DOD HCI Style Guide is recommended for customization to local norms.

3.14.3 Natural language support services. These services provide the capability to support several languages simultaneously.

3.14.3.1 Keyboard device layout. (This BSA appears in both part 3, User Interface, and part 14, Internationalization.) Keyboard device layout standards specify the arrangement of keys on a keyboard.

3.14.3.1.1 Standards. Table 3.14-16 presents standards for keyboard device layout.

3.14.3.1.2 Alternative specifications. The only other available specifications are proprietary.

3.14.3.1.3 Standards deficiencies. Deficiencies in the existing standards are unknown.

3.14.3.1.4 Portability caveats. Portability problems related to the existing specifications are unknown.

3.14.3.1.5 Related standards. No standards are related to keyboard device layout standards.

3.14.3.1.6 Recommendations. Conformance to all ISO and ISO/IEC keyboard specifications conforming to DIS or IS levels is recommended. This is especially important for equipment that will interoperate with that of U.S. allies (e.g., NATO).

3.14.4 Related standards and programs. This MLSA includes services supporting internationalization indirectly.

3.14.4.1 Character set registration. (This BSA appears in part 5, Data Interchange, and part 14, Internationalization.) Character set registration provides a mechanism for identifying and defining graphic character sets

3.14.4.1.1 Standards. Table 3.14-17 presents standards for character set registration.

ISO 7350 spceifies procedures for preparing, registering, publishing, and maintaining the register of graphic character sets and procedures for assigning identifiers to the sets.

3.14.4.1.2 Alternative specifications. There are no alternative specifications.

3.14.4.1.3 Standards deficiencies. Deficiencies in the existing standards are unknown.

3.14.4.1.4 Portability caveats. Portability problems in the existing standards are unknown.

3.14.4.1.5 Related standards. The following standards are related to character set registration:

3.14.4.1.6 Recommendations. There are no recommendations.