🚀 go-pugleaf

Let me try to summarize the good(+)/bad(-) points for you:

Method 1: Create images without call-outs, add call-outs in
each authoring tool:

+ Good for images that may be imported at varying sizes,
  since the call-outs won't be resized with the image
+ Good for images that you may wish to use without call-outs
+ Possible to have different call-outs in different documents
+ May be easier if you need to translate textual call-outs

- Call-outs need to be re-inserted in each tool/document
- Call-out appearance may differ due to different tool capabilities


Method 2: Create images with call-outs directly in graphic software:

+ Only need to insert call-outs once
+ Call-outs will look the same in all documents

- Resizing the image will also resize the call-outs
- Images will always have call-outs, and the same call-outs in all docs
- Translators will need to translate text in two different places
  (body text in document, textual call-outs in images)


For translation purposes, it's always better to use numbered or
alphanumeric-coded call-outs and add the actual call-out text
outside the image (in the document). For method 1, this requires
all call-out codes to always be the same.


So, if you *always* will have the same call-outs for an image,
and all (or most) images always will be imported at the same 
relative size, method 2 may be the way to go.
If you think you may want to use some images without call-outs,
or have different call-outs in some documents, or size the
images differently within a document, then method 1 may be best.

-- 
/Thomas Michanek