
Bug:
-- the name of the web site that issued a cookie is not tracked.
   thus, if webclient is used to bounce from one website to another,
   it will be sending all cookies to all sites, which is not correct.
   This shouldn't hurt in most cases, but may cause confusion, and 
   it certainly leaks "confidential" cookies. 


	-- Shirley has seen situations where the
	   average number of gif files per run
	   is not an integer...  guess is that
	   somehow the gif files are not getting
	   fetched correctly and repeatedly

	
If you have comments in a webclient script, and you retrain the file
(using the -v option) the comments are all moved to the front of the
file.....

-v option in webclient loses the <<END>> line.

Can we go to a real checksum of the entire
html file rather than this kludgy thing we use now?

Looks like there still is a problem with gif_cache_emulation.
Fractional gif files fetched per trial in statistics reports
for workload.out.short.

The trial, request, etc numbers are incorrect when a check sum error
occurs.... (all zero)
