Editing UTF-32 (section)

== Utility of fixed width ==
A fixed number of bytes per code point has theoretical advantages, but each of these has problems in reality: 

* Truncation becomes easier, but not significantly so compared to [[UTF-8]] and [[UTF-16]] (both of which can search backwards for the point to truncate by looking at 2–4 code units at most).{{efn|For UTF-8: Select point to truncate at. If the byte before it is 0-0x7F, or the byte after it is anything other than the continuation bytes 0x80-0xBF, the string can be truncated at that point. Otherwise search up to 3 bytes backwards for such a point and truncate at that. If not found, truncate at the original position. This works even if there are encoding errors in the UTF-8. UTF-16 is trivial and only has to back up one word at most.}}{{citation needed|date=January 2023}}
* Finding the ''Nth'' character in a string. For fixed width, this is simply a [[Big O notation|O(1) problem]], while it is [[Big O notation|O(n) problem]] in a variable-width encoding. Novice programmers often vastly overestimate how useful this is.<ref name=manishearth>{{Cite web|title=Let's Stop Ascribing Meaning to Code Points |website=In Pursuit of Laziness |first1=Manish |last1=Goregaokar |date=January 14, 2017 |url=https://manishearth.github.io/blog/2017/01/14/stop-ascribing-meaning-to-unicode-code-points/|access-date=2020-06-14|quote=Folks start implying that code points mean something, and that O(1) indexing or slicing at code point boundaries is a useful operation. }}</ref> Also what a user might call a "character" is still variable-width, for instance the [[combining character]] sequence {{char|á}} could be 2 code points, the emoji {{char|👨‍🦲}} is three,<ref>{{Cite web|title=👨‍🦲 Man: Bald Emoji|url=https://emojipedia.org/man-bald/|access-date=2021-10-12|website=Emojipedia|language=en}}</ref> and the ligature {{char|ﬀ}} is one.
* Quickly knowing the "width" of a string. However even [[Duospaced font|"fixed width" fonts]] have varying width, often [[CJK characters|CJK ideographs]] are twice as wide,<ref name=manishearth/> plus the already-mentioned problems with the number of code points not being equal to the number of characters.