Preserving the Visual Presentation Form of Digitized Traditional Mongolian Script While Waiting for Unicode Specifications to Stabilize

Abstract

Purpose: Digital encoding for traditional Mongolian text has evolved over time. The two main strategies for encoding Mongolian text are graphetic and phonetic. Graphetic solutions encode the glyph shapes while phonetic solutions encode the letters and rely on contextual shaping rules to determine the letter shapes. Unicode takes the phonetic approach by encoding the letters of the traditional Mongolian alphabet, but it does not specify any shaping rules. Variation between fonts and variation as the specifications evolve over time cause Unicode text to be malformed when displayed in an environment different than it was created in. Attempts have been made to standardize the Unicode shaping rules. However, not all entities have agreed on a single standard yet.

Methods: In order to preserve the correct visual appearance of digitized traditional Mongolian text until an unambiguous specification is standardized and agreed upon by all parties, this paper proposes using a graphetic encoding model to store the glyph shapes. This text can then be converted back to Unicode on demand.

Results: An open source converter mongol_code was developed and published, which converts between Menksoft code and Unicode following the GB/T 25914-2023 specification. The converter achieved a 100% pass rate across 1,401 unit tests with 88.7% code coverage.

Conclusions: Because Menksoft code encodes both the glyph shapes and phonetic letters, it successfully preserves these elements during conversion to and from Unicode, which allows for long term storage while waiting for Unicode specifications to stabilize.