This filter is for importing PDF comments which have been exported to XFDF format (an XML variant). Since memoQ does not recognize the extension type, the import dialog will have to be set to "All Files" or a template configured to recognize the extension must be used. Line breaks within the comments are set to be converted to inline tags.

I'm still investigating various export/import scenarios with these comments, and I have lost any respect I had for Adobe's programmers or their designers who write specifications. When someone first mentioned the other format (FDF), I looked and that was the only export format I saw. I was looking at a single comment or a few selected comments to export. As you may have noted in the other discussion, FDF is only for ANSI characters - everything else gets trashed. Only when I chose to export ALL comments was I offered the XFDF format, which will support any language. That is just weird and stupid to limit the more selective exports that way. I actually discovered the existence of XFDF when looking at a different (3rd party) tool, though I don't remember if that was offered when exporting only a few comments from a larger number.

I'm interested in possible use scenarios for all of this. The colleague who brought this up didn't mention her client's intentions, but I can see how something like this could be useful in some of the legal cases I've been involved with. I've found that I can take sets of comments in any format from different copies of the PDF and "merge" them by importing to the same PDF. So you could, for example, send out a PDF for commentary to a group of people and them compile all their comments in whatever languages in a single PDF. These comments can overlap too, as you can see in the screenshot I've included. It's kind of fun to think of all the things you could do with this in a teaching exercise, including using that NLP locking feature of memoQ to isolate a single source language.

Another user's comment: Kevin, one scenario we used the "PDF comments only translation" for is those image only PDFs, getting transcription performed in a comment attached to the respective part of the pdf, then we could send the translated commented PDF back to the design team to sort out (copy paste back into Photoshop)

So... go into the Resource Console and import it under "Filter configurations":

Then to import an XFDF file, set the file type to "All Files" so you can see it. You'll get a message that the filter type is unknown, but then just select the XML configuration designed for XFDF files.

You can also set up a template that will essentially do all this automatically. See the books on memoQ Project templates by Marek Pawelec for the best advice on this..