Hi
I believe there might be a slight bug with the method of 
parsing the links to multiple java classes, as I found 
while retrieving pages on this site:
----
URL: <http://www.ocw.mit.edu\18\18.013a\f01\required>-
readings\chapter05\section02.html
Original source:

<applet code="PlanesInSpace" codebase="../applets/" 
archive="planesInSpace.jar, go.jar, goText.jar, mk_lib.jar, 
parser_math.jar, jcbwt363.jar" width=760 
height=450></applet>
----
Now in the above code, there are multiple jar files 
mentioned under the archive= field. WinHTTrack isn't able 
to recognize them as multiple file names, and instead 
attempts to download a file with the same name as that 
string, which obviously doesn't exist (it even tried to 
look for .class files, even though all the code is in the 
jars).
The htmls that are locally saved are like this:

<applet  archive="planesInSpace.jar,%20go.jar,%
20goText.jar,%20mk_lib.jar,%20parser_math.jar,%
20jcbwt363.html" width=760 height=450 
codebase="../applets/" code="PlanesInSpace.class"></applet>

So I manually downloaded all the jars in the /applets dir 
with another download prog, replaced the %20s back to 
spaces, and renamed the jcbwt363.html in this case to the 
actual jar extension and it worked.
Guess you might wanna look into this for the next version.
-MT

p.s. can httrack download all files within a specific 
folder (like the /applets/*.jar in this case), even if 
those files are not linked in html pages (which I had to do 
here because these jar paths weren't properly resolved by 
httrack)?