>>>Keep in mind that the text you see in a PDF may not be text in the PDF - it could be an image. It depends on how the PDF was created.
>>
>>True - but unlikely that it's in an image. Most likely it's in ASCII, hex or an encoded stream - all of which, I would assume, would be understood by Word....
>
>That's not quite true:
>
>- A few years back, in the early days of free PDF printer drivers I ran into a least one, might have been more, of those drivers that created PDFs by creating image snapshots of each page and then stitching them together, one image per page, into a PDF

Any scanner will offer that way to create a PDF from paper - as images, or else there would inevitably be OCR errors and support calls. So images they are.

>- The original source could be images of text. I for one do this quite often, sending screen shots or snipping tool images for support purposes. I also train my primary contact at each client how to use PrtSc, Alt-PrtSc and the Snipping Tool so they can send me error messages "verbatim"; they invariably think that's cool and start using it with other people

And it takes only a few years to teach them to paste straight into their email clients (who were all capable of inserting an image from clipboard for at least ten years now) and not into a word document which they'd attach to email. Before docx format (i.e. zipped xml instead of old legacy format) these would include the image in who knows which format plus a .bmp preview and whatnot. The emails were huge.