In the first part of this paper, we propose a watermarking method for text documents that is less vulnerable to attacks than
previous methods. In the second part, we propose a method for protecting the actual content of the document. In a section
of independent interest, we discuss the interplay between error-correcting codes and steganography.
What we call a fingerprint is a “robust invisible watermark” for Mintzer el al. [15], and what we call a watermark is a “robust visible watermark”.