Just an FYI: The f:view tag in Facelets 1.1 had the ability to set contentType and encoding of responses via two named attributes. The 2.0 RI inherited that capability, but we neglected to add it to the faclets tld. Unless there are any objections, the RI team would like to treat this as a specification bug, adding these attributes to the TLD, and documenting this in the MR. This behavior is already supported in the released RI. Please let me know if you have any concerns. Jim