VolgaCTF 2015 Poem

So after looking at the text it's all the same as the the source, but the formatting of the document looks a little odd, some of the line spacings are different for no obvious reason.

To take a better look at the structure of the document I grabbed the tool iText RUPS

Looking at the paragraphs, it seems that they alternate between 14.xxx and 17.xxx, so copy them all out, and then swap them out for 0 and 1 respectivly, splitting into groups of 8 gives a few chars and then breaks down.

Looking at where it breaks down it's on the page boundry, so add an extra space (I've added the value of the last paragraph to see them)

     01111011
     01110011
     01101001
     0111101  -37.904
     01100101
     01011111
     01100100
     01101111
     0100101   -40.738
     01110011
     01011111
     01101101
     0100001  -37.904
     01110100
     01110100
     01100101
     0111010  -34.452
     01011111
     01100001
     01100110
     0111000  -46.408
     01100101
     01110010
     01011111
     0110000   -34.452
     01101100
     01101100
     0111110   -190.854

A quick python script to output it allong with options for each of the missings (should be easy to work out from context and hopefully will split allong some line which is aparent in hindsight)

#!/usr/bin/python

print chr(0b01111011)
print chr(0b01110011)
print chr(0b01101001)
print "%s %s" % (chr(0b01111010),chr(0b01111011))
print chr(0b01100101)
print chr(0b01011111)
print chr(0b01100100)
print chr(0b01101111)
print "%s %s" % (chr(0b01001010),chr(0b01001011))
print chr(0b01110011)
print chr(0b01011111)
print chr(0b01101101)
print "%s %s" % (chr(0b01000010),chr(0b01000011))
print chr(0b01110100)
print chr(0b01110100)
print chr(0b01100101)
print "%s %s" % (chr(0b01110100),chr(0b01110101))
print chr(0b01011111)
print chr(0b01100001)
print chr(0b01100110)
print "%s %s" % (chr(0b01110000),chr(0b01110001))
print chr(0b01100101)
print chr(0b01110010)
print chr(0b01011111)
print "%s %s" % (chr(0b01100000),chr(0b01100001))
print chr(0b01101100)
print chr(0b01101100)
print "%s %s" % (chr(0b01111100),chr(0b01111101))

which gives us the following:

      {
      s
      i
     z {
      e
      _
      d
      o
     J K
      s
      _
      m
     B C
      t
      t
      e
     t u
      _
      a
      f
     p q
      e
      r
      _
     ` a
      l
      l
     | }

Clearly I've misplaced those bits in a few places, but it's clear from what's there that the answer is supposed to be {size_does_matter_after_all}