American Standard Code for Information Interchange

Technology Dictionary -> American Standard Code for Information Interchange

American Standard Code for Information Interchange



The basis of character sets used in almost all present-day computers. US-ASCII uses only the lower seven bits (character points 0 to 127) to convey some control codes, space, numbers, most basic punctuation, and unaccented letters a-z and A-Z. More modern coded character sets (e.g., Latin-1, Unicode) define extensions to ASCII for values above 127 for conveying special Latin characters (like accented characters, or German ess-tsett), characters from non-Latin writing systems (e.g., Cyrillic, or Han characters), and such desirable glyphs as distinct open- and close-quotation marks. ASCII replaced earlier systems such as EBCDIC and Baudot, which used fewer bytes, but were each broken in their own way.

Computers are much pickier about spelling than humans; thus, hackers need to be very precise when talking about characters, and have developed a considerable amount of verbal shorthand for them. Every character has one or more names - some formal, some concise, some silly.

Individual characters are listed in this dictionary with alternative names from revision 2.3 of the Usenet ASCII pronunciation guide in rough order of popularity, including their official ITU-T names and the particularly silly names introduced by INTERCAL.

See V ampersand, asterisk, back quote, backslash, caret, colon, comma, commercial at, control-C, dollar, dot, double quote, equals, exclamation mark, greater than, hash, left bracket, left parenthesis, less than, minus, parentheses, oblique stroke, percent, plus, question mark, right brace, right brace, right bracket, right parenthesis, semicolon, single quote, space, tilde, underscore, vertical bar, zero.

Some other common usages cause odd overlaps. The "#", "$", ">", and "&" characters, for example, are all pronounced "hex" in different communities because various assemblers use them as a prefix tag for hexadecimal constants (in particular, "#" in many assembler-programming cultures, "$" in the 6502 world, ">" at Texas Instruments, and "&" on the BBC Micro, Acorn Archimedes, Sinclair, and some Zilog Z80 machines). See also splat.

The inability of US-ASCII to correctly represent nearly any language other than English became an obvious and intolerable misfeature as computer use outside the US and UK became the rule rather than the exception (see software rot). And so national extensions to US-ASCII were developed, such as Latin-1.

Hardware and software from the US still tends to embody the assumption that US-ASCII is the universal character set and that words of text consist entirely of byte values 65-90 and 97-122 (A-Z and a-z); this is a major irritant to people who want to use a character set suited to their own languages. Perversely, though, efforts to solve this problem by proliferating sets of national characters produced an evolutionary pressure (especially in protocol design, e.g., the URL standard) to stick to US-ASCII as a subset common to all those in use, and therefore to stick to English as the language encodable with the common subset of all the ASCII dialects. This basic problem with having a multiplicity of national character sets ended up being a prime justification for Unicode, which was designed, ostensibly, to be the *one* ASCII extension anyone will need.

A system is described as "eight-bit clean" if it doesn't mangle text with byte values above 127, as some older systems did.

See also ASCII character table, Yu-Shiang Whole Fish.

(1995-03-06)


© Art Branch Inc.

SQL Tutorial