pdf-word-translationWe are very often requested to localize PDF files for brochures and datasheet especially when their creation was outsourced to a creation agency. We always ask our customer to provide us with source files since there are a lot of drawbacks of using a compressed output format such as PDF for translation:

  • PDF files are screen renditions of printed pages essentially based on the PostScript specifications for publishing printed pages to the computer screen.
    This means that text is “often” vector based, and can be copied and pasted from the PDF into a word processor. UNFORTUNATELY, publishing applications such as Quark and InDesign¬† have more advanced text processing features than WinWORD. Which often means kerning and text spacing has an influence on the text that is copied attributing XY coordinates to each letter. This often results in EACH and EVERY LETTER of EVERY WORD being considered a separate translatable block!
  • The PDF files usually contain both text and graphics. Using the PDF to generate translatable targets has an influence on the graphics – the output quality and colour accuracy will depend on what conversion process we adopt and we are very unlikely to have anything near the source quality as PDF graphics are by nature compressed and down-sampled to provide smaller file sizes!

We hate localizing PDF files because the output quality will be significant below our standards.