possible bug in the validator?

Hello,

I'm having a problem when using online HTML Validator
(http://validator.w3.org/). When validating my pages it detects correct
character encoding (specified in the tag "<meta http-equiv="Content-type"
content="text/html; charset=windows-1250" />") only if I upload the file. If
I validate the page by entering its URI (www.volny.cz/voitech/) the
validator detects default us-ascii encoding. Is it a bug? Thank you for what
you at w3c are doing, Vojtech Rynda