A while ago, I wrote about both my love of the free and open source translation environment tool OmegaT and my frustration with tags in OmegaT. Since then, I’ve found that the easiest solution, as long as the document doesn’t contain complex formatting, is to save the document as plain text, then translate in OmegaT and then use your word processing program to put the formatting back in.
In order to make sure that the tags are really gone, I save the document as .txt in OpenOffice.org, then open it in a text editor (I use Kate, but any text editor will do) and save it again, then place that file into the source folder in the OmegaT project. This method is especially helpful when a client sends a “Word document” that is actually an OCR of a PDF (talk about tags!!).