4 ms·

> JPG+RAR files was used to smuggle information past American checkpoints Have any more information about this? I couldn’t find anything on google.

jeroenhd 4y ago
Google is notoriously bad at finding older news articles and its competitors don't do nuch better. If memory services, the technique used relies on one file format putting its offsets and metadata in the header (like JPEG does) and another putting everything in the footer (looking into this, ZIP makes more sense). Concatenate the two files and you've got a file that's both a valid JPEG with some garbage at the end and a valid ZIP with some garbage before the first header. The JPEG parser would start at the front and seek until the length specified in the header has been processed, the ZIP parser would start at the central directory at the end and work its way backwards through offsets to find the first header. In Windows this could be as simple as `copy /b picture.jpg /b data.zip secret.jpg`. Rename secret.jpg to secret.zip and you've got a working ZIP file you can add files to as you please, rename back to secret.jpg and you've got an inconspicuous JPEG of about 3GB sitting on your desktop. Perhaps this simple methodology was just an urban legend, it certainly made seemed like an impressive trick when I was in high school. Articles online regarding this topic mention steganography but describe more advanced methods (i.e. encoding data in the least significant bit in pixels of a frame, encoding data in the cosine table of a JPEG, etc.). From what I can tell, this all seems to be based on a single security researcher explaining the concept of steganographry to a single news source followed by simplification and loads of copy/pasting.
- Anthony-G 4y ago
  OT: > Google is notoriously bad at finding older news articles and its competitors don't do nuch better. It’s not just Google. I’ve found the same to be true for Duck Duck Go and Bing. :(

jeroenhd 4y ago

Google is notoriously bad at finding older news articles and its competitors don't do nuch better. If memory services, the technique used relies on one file format putting its offsets and metadata in the header (like JPEG does) and another putting everything in the footer (looking into this, ZIP makes more sense). Concatenate the two files and you've got a file that's both a valid JPEG with some garbage at the end and a valid ZIP with some garbage before the first header. The JPEG parser would start at the front and seek until the length specified in the header has been processed, the ZIP parser would start at the central directory at the end and work its way backwards through offsets to find the first header. In Windows this could be as simple as `copy /b picture.jpg /b data.zip secret.jpg`. Rename secret.jpg to secret.zip and you've got a working ZIP file you can add files to as you please, rename back to secret.jpg and you've got an inconspicuous JPEG of about 3GB sitting on your desktop. Perhaps this simple methodology was just an urban legend, it certainly made seemed like an impressive trick when I was in high school. Articles online regarding this topic mention steganography but describe more advanced methods (i.e. encoding data in the least significant bit in pixels of a frame, encoding data in the cosine table of a JPEG, etc.). From what I can tell, this all seems to be based on a single security researcher explaining the concept of steganographry to a single news source followed by simplification and loads of copy/pasting.

Anthony-G 4y ago
OT: > Google is notoriously bad at finding older news articles and its competitors don't do nuch better. It’s not just Google. I’ve found the same to be true for Duck Duck Go and Bing. :(