Character set specification

In order to satisfy the validator, I modified the HTTP headers
sent by IIS for .HTML documents to specify 

	Content-Type: text/html; charset=ISO-8859-1

rather than 

	Content-Type: text/html

as previously, as a result of which my campus search engine 
(Infoseek/Inktomi/Verity)  refused to index the pages, 
diagnosing them as "406 : not acceptable".  Since searching
is even more important than validation, I have with some 
regret re-configured IIS to once again send

	Content-Type: text/html

Since it would clearly be impossible to modify all 35000 documents
to use an HTTP-EQUIV meta-statement, can the validator team please
suggest any other way of allowing the validator to perceive the
documents as having a defined character set whilst allowing
Infoseek/Inktomi/Verity to continue to index the pages ?

Philip Taylor, RHBNC