Do you scan documents? Has your company scanned and retained documents from as far back as 2005 using a Xerox Scanner? Are these documents required for legal, contractual or even commercial reasons?  Then it’s possible that the data you are storing is no longer reliable.  

The bug identified by David Kriesel has some serious implications, if you haven’t retained the original document, the electronic copy no longer offers non-repudiation, and therefore reasonable doubt of it accuracy.

See http://www.dkriesel.com/en/blog/2013/0802_xerox-workcentres_are_switching_written_numbers_when_scanning

The bug has been traced back to an implementation of the JBIG2 Compression Algorithm, which at a high level chops up an image into smaller chunks to create a dictionary of blocks, the algorithm then looks for similar blocks and replaces them with a dictionary version provided that the error rate meets a set criteria.  So in this case a 6 can look like an 8.

The issue has been reported with the “Normal” setting that the scanners can be set to which uses the JBIG2 compression scheme.  Xerox advise that the default method is High or Higher which doesn’t use the JBIG2 compression algorithms.

It will be interesting to see how this pans out, especially with legal evidence that may be required for cases past or present as the corruption of your documents aren’t obvious at first glance.

Smith And Brooks Please Help Support Us As Independent Developers 🙂