I am also researching a Chinese character telegraph encoding scheme, and I will provide a brief overview here.
Referencing the dual-character plus character-based input method scheme, reduce the length of individual code to 4 English characters and minimize redundancy.
Scope: National Development and Reform Commission Notice [2013] No. 23, "General Specification for Chinese Characters"
Encoding example
Han (han 部件 氵小字 又 => (hjdy)
Character (zi), component 宀, small character 子 => (zibz)
Hang (hang xiaozi mu xiaozi qi => (hhmj)
State (zh ou) - first stroke, last stroke, | => vzdl)
The current problems with this scheme are:
- For any combination of 4 letters, there are theoretically only 7963 valid codes, which is less than the standard telegraph code.
- Given the current rules and existing encoding schemes, it's impossible to achieve a situation where common characters have no duplicate codes.
- There are 529 instances where one character corresponds to two words.
- There are 50 instances where one character corresponds to three characters.
- There are 6 combinations where one character corresponds to four characters.
- There are 2 instances where one code corresponds to five characters.
Disadvantages
- You need to memorize the dual-pinyin rule (i.e., using two letters to represent a complete pinyin), Chinese character shapes, rules for breaking down characters, and the mapping relationship between components (i.e., which component is represented by which letter).
Advantages
- Fixed-length encoding, with the ability to distinguish at higher speeds.
- Less re-encoding, more precise expression.
- Once familiar with the rules, you can encode and decode directly on-site without referring to a table. When precise expression is not required, placeholders can be used to omit glyphs, allowing the decoding system to complete decoding based on context and semantics.
It's currently just a draft.
闽ICP备2021006864号-7
闽公网安备35020602002794号