Resolves discussion in github PR:

Character sequence -> byte sequence
Remove parens

Author: Etienne Vouga

spec/2023-07-draft.md

@@ -987,7 +987,7 @@ Arguments                    | Description
 The default output validator parses the submission output and answer files as a string of single-byte characters and tokenizes the files by splitting on sequences of one or more consecutive whitespaces.
 Whitespace characters are: space (`0x20`), form feed (`0x0c`), line feed (`0x0a`), carriage return (`0x0d`), horizontal tab (`0x09`), and vertical tab (`0x0b`).
 In its default mode, the default output validator then ignores the whitespace and compares the submission output and answer files token by token. 
-(If there is a disagreement in the number of tokens in the two files, the validator rejects the submission instead.)
+If there is a disagreement in the number of tokens in the two files, the validator rejects the submission instead.
 The validator may also reject any submission output that is unreasonably large (including due to containing unreasonable amounts of whitespace).
 
 #### Comparing Floating-Point Tokens
@@ -1009,7 +1009,7 @@ For problems containing a mix of integer and floating-point output, a custom out
 
 #### Comparing String Tokens
 
-If `case_sensitive` is specified, the two tokens must match exactly, i.e. consist of exactly the same character sequence.
+If `case_sensitive` is specified, the two tokens must match exactly, i.e. consist of exactly the same byte sequence.
 
 Otherwise, the submitted output token is accepted if it matches the answer file token up to case. The default output validator treats uppercase ASCII letters `A`--`Z` as equivalent to their lowercase counterparts.