Damien Cassou wrote: 
On Sun, May 11, 2014 at 3:20 PM, Ben Coman <btc@openinworld.com> wrote:
  
I noticed that many of the figures are pdf, so I am really curious why this
was done and not left as a JPG or PNG?  It may not matter for a book but it
seems awkward if a html version is needed at some point. Also I added a
.gitignore since `git status` showed way too much irrelevant info to be
useful - and the PDF-figures prevent *.pdf being a filter (but *.pier.pdf is
an effective alternative for our purpose).
    

- PDF is a vector graphics format.
- JPG and PNG are not.

http://en.wikipedia.org/wiki/Vector_graphics
But these screen snapshots (which I didn't clarify earlier, see attached) are bitmaps not vector graphics. From line 22 of attached file: "<< /Length 8 0 R /Type /XObject /Subtype /Image /Width 1428 /Height 1070 /ColorSpace 9 0 R /BitsPerComponent 8 /Filter /FlateDecode >> stream"

Okay I see a benefit in annotations on top of the bitmap being re-editable later, but it makes the underlying bitmap harder to edit.  'll have see how this balance pans out in practice.  Maybe the bitmaps wont need much editing.

cheers -ben