v 0.5 (July 8, 2002)
  * the released jar file is no longer signed, so it won't run as an applet inside
    Netscape or IE.  Too much of a pain to sign the jar file.
  * Websphinx is now available under an Apache-style license instead of LGPL
  * pages are now downloaded as byte arrays instead of UTF-8 strings, so
    that binary content like images and PDF files can be mirrored safely (Chad Phillips)
  * content-type "text/html charset=XXX" is handled properly (Noppadon Kamolvilassatian)
  * race condition between marking links as queued and removing them from queue 
    (Ludmilla Portugeis)
  * now uses redirected URL as base URL

v 0.4a (Jan 20 2000)
  * Recompiling from sources is easier, because the source distribution
    now includes all the class files found in the binary distribution.
  * Fixed some compile errors in the source.

v 0.4 (Apr 17 1998)
  * Outline visualization
  * Visualization options panel
  * Concatenator now converts GIFs and JPGs to inline images
  * Concatenator puts a page break between each Web page (only affects how page is
    printed)

v 0.3 (Mar 27 1998)
  * Simple mode added.
  * Link type restriction added (hyperlinks v. images+hyperlinks v. all-links).
  * Concatenator now converts text to HTML (wrapping the text with &lt;PRE&gt;
    tags and quoting special characters).
  * Known bugs (soon to be fixed):
     -Popup dialogs don't appear under IE 4.0.

v 0.2 (Mar 11 1998)
  * Internet Explorer 4.0 support added.
  * The Save and Concatenate operations change links to point to saved local
    pages whenever possible.
  * The Statistics panel is now user-drawn.

v 0.1 (Mar 6 1998)
  * Support for Netscape 4.0, Netscape 3.0, JDK 1.1, and JDK 1.0.2.
  * Known bugs (soon to be fixed):
     - The Save and Concatenate operations change all links into absolute references
       to the original versions out on the net. These operations ought to point
       to saved local pages whenever possible.
     - Server redirections (such as the <A HREF="http://www.cs.cmu.edu/">SCS
       homepage</A>) are not taken into account by the HTML parser, so relative
       links from redirected pages are misinterpreted.
