English Language & Usage Stack Exchange is a question and answer site for linguists, etymologists, and serious English language enthusiasts. It only takes a minute to sign up.
Sign up to join this community
In serial numbers etc., there are many cases where you can't tell if the intended character is a number or a letter. For example, the number 0 and the letter O, the number 1 and the letters l or I, and so on.
Is there a name for this concept/problem?
Background: I'm writing a spell corrector for my search engine that handles this situation. I'd like to know what this concept is called so I can document it properly.
A homoglyph is a character identical or nearly identical in appearance to another, but which differs in the meaning it represents.
Wikipedia has a detailed article on homoglyphs as well as one on IDN homograph attacks where phishers take advantage of this property to mislead their victims.
As for your spell-corrector, you should be able to class such occurrences under something along the lines of homoglyph errors or homoglyph conflicts. However, it's a technical term which might possibly be a little too esoteric for your audience.
The error is called a homoglyph error.
Characters likely to lead to homoglyph errors are referred to as confusables in Unicode parlance.
The distinction is because glyphs depend upon the font as well as the character, what could be a pair of homoglyphs in one font, may not be in another (e.g. 0 and O look quite distinct to me right now, but may not in another font, while Τ and T look identical to me right now, but may be clearly distinct in another font).
Hence homoglyphs for the confusable shapes, and confusables for the characters likely to be rendered with them.
