* Character and string types

  In Eclipse, ACL, LispWorks the type BASE-CHAR includes only Latin-1
  characters, whereas the CHARACTER type includes all of Unicode (16 bit).

  In CLISP, BASE-CHAR and CHARACTER are equivalent and include all of
  Unicode (16 bit). The memory representation of read-only strings
  (e.g. symbol print names and program literals) is optimized to 1
  byte/character if possible.

* Supported external formats of streams

  - CLISP 2000-03-06: Around 80 external formats, including all of the
    ones supported by browsers and Linux locales.
  - Eclipse: Only :ASCII (1 byte/character), :UCS (2 bytes/character),
    and :MULTI-BYTE (locale dependent multibyte representation, works
    only on OSes for which wchar_t is Unicode).
  - ACL 6.0: Lots of external formats, mostly table-driven.
  - LispWorks: Around 10 external formats, including Latin-1, Unicode
    (2 bytes/character), UTF-8, and the most important Japanese encodings
    (but not ISO-2022-JP).

  Different end-of-line conventions are indicated to OPEN through the
  :external-format argument in CLISP and LispWorks, and through an extra
  argument to OPEN in Eclipse.

* Additional API

  - CLISP: STRING-WIDTH returns the display width of a string, used by
    FORMAT ~T.
  - Eclipse: none.
  - ACL: unknown.
  - LispWorks: functions for guessing the encoding of a file (important
    for Japanese environments)

* FFI support

  - CLISP: FFI can pass strings only with single-byte encodings.
  - Eclipse, ACL: unknown
  - LispWorks: a few specialized macros for passing strings from/to C.