This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
A lot if windows-world editors tend to prepend a BOM to unicode files, and the CSS parser of the validator does not appear to strip the BOM. In some cases (especially when stylesheet starts with @media ) it causes a parse error and false invalidation. In other cases it only adds spurrious characters to the output.
examples: http://jigsaw.w3.org/css-validator/autotest/testsuite/bugs/4828-bom_atmedia.css has a BOM at the beginning, starts with @media => parse error Parse Error - ���@media screen http://jigsaw.w3.org/css-validator/autotest/testsuite/bugs/4828-bom_notatmedia.css has a BOM, starts with body selector => source display has garbage characters
In the soon-to-be-released development version, which uses a different template engine for the display and the tagsoup parser, this looks fixed. http://qa-dev.w3.org:8001/css-validator/