UnicodePlain-TextEncodingofMathematics分析.docVIP

Unicode Plain-Text Encoding of Mathematics Murray Sargent III Microsoft Corporation, Redmond, WA 98052 8/10/94 With Unicode plus a few special symbols, it is possible to encode mathematical expressions in a readable plain text. The format is linear, but it can be readily displayed in built-up form. The method uses heuristics to recognize mathematical expressions without the aid of explicit “math-on” and “math-off” commands. Comparison is given with the standard TEX representation as well as with a Unicode TEX representation. Some aspects of keyboard entry are discussed, and export to TEX, C, and symbolic manipulation programs is outlined. 1. Introduction Mathematics is an international language that is used to varying degrees by a majority of members of the information society. The most obvious participants are scientists and engineers, but social scientists and business people use mathematics in important ways. Nevertheless, the entry and machine format of mathematical expressions remains awkward. Unicode’s large mathematical symbol set helps greatly by assigning well-documented codes to mathematical symbols. Coupled with some heuristic algorithms and a few new dedicated symbols, Unicode can be used to encode mathematical expressions in remarkably readable plain text. This text can be used in a number of ways. It can be used “as is” for simple documentation in programs or papers. It can be processed to display formulas in standard built-up form. It can be exported into TEX, SGML, and other document processing systems. And it can be exported into programming languages and symbolic manipulation systems. The export into programming languages is particularly intriguing for the programming community, since formulas can be programmed essentially as they appear in the plain-text Unicode; the same symbols used in documentation are used in the programs. This simplification is a dramatic advance in the endeavor for true FORmula TRANslation. In the mid 1950’s,