Question mark

[Doug Ewell]

Юрий Бэкап написал:

> I've been wondering about the following question for a long time: how
> difficult and how feasible is it to add a question mark to the Unicode
> table that would be identical to the regular question mark but usable
> in Windows operating systems? Almost all characters prohibited in
> Windows OS have their equivalents in the Unicode table, allowing the
> use of characters like "/", "", ":", etc.

As far as difficulty and feasibility are concerned: One would need to write and submit a proposal that explains why encoding an identical-looking character, with identical Unicode properties, and intended for use with the same set of writing systems, as an existing character which is syntactically significant in many environments — XML, URIs, regular expressions, many programming languages — would not create a tremendous security problem.

Unicode does not typically encode lookalike, or near-lookalike, characters with identical properties for the sole purpose of looking alike. There is usually some functional difference between the two, and that difference must be defined by Unicode, not by (for example) the syntax rules of a particular vendor’s file system.

File names, like domain names, are identifiers. They are not required, and usually are not expected, to have all the flexibility of ordinary text in human languages.

--
Doug Ewell, CC, ALB | Lakewood, CO, US | ewellic.org