I think we should add this information to the design rationale, near the end.  Comments?

===========================================
<p>
This SRFI is intentionally silent on the interpretation of unprefixed
square brackets, because
different Schemes (as well as other Lisps) interpret
square brackets differently.
One
<a href="http://trac.sacrideo.us/wg/wiki/BracketsBraces">survey of
Scheme implementations with brackets and braces</a>
shows these differences;
several Scheme implementations follow the R6RS specification that
accepts <samp>[...]</samp> as a synonym for <samp>(...)</samp>,
GNU Kawa interprets <samp>[...]</samp> as the redefinable constructor
<samp>($bracket-list$&nbsp;...)</samp>,
and two implementations (Rep and FemtoLisp) use them as vector constructors.
By intentionally not defining the interpretation of unprefixed
square brackets, implementations are free to continue to use whatever
interpretation their users are used to, and users can easily access
that interpretation.
</p>

--- David A. Wheeler