Wandering about the universe trying to track down the reason for a change in CGI.pm and came across this node.

Lincoln Stein changed the code in escapeHTML from using ‹/&155; to using ‹/›.   Change comment was:

• Version 2.84
  HTML escaping code now replaced 0x8b and 0x9b with unicode references ‹ and ›

So the equivalent code would become:

    s{\x8b}{‹}gso;
    s{\x9b}{›}gso;
[download]

Still haven't found _why_ the change ... I guess the Unicode values were more correct than assuming a particular charset?