Other Xerox devices reportedly affected

Reading last night's mails, I get the information that readers have been able to reproduce the error on the following devices (thanks, Christian and others):

  • WorkCentre 7530, WorkCentre 7328, WorkCentre 7346, WorkCentre 7545

As usual, the following holds: I did not reproduce the error on these devices myself, so consider this as hearsay, even though plausible. My own Observations have been made on the following devices:

  • WorkCentre 7535, WorkCentre 7556

Other mails reported the reproductability of the issue on

  • Xerox ColorQube 9203, Xerox ColorQube 9201

There are also some readers able to reproduce errors, but understandably not willing to be named here for credits at the same time. I respect this attitude and then give credits to anonymous readers, so feel free to mail me even though you do not want to be named. After writing this article, I will add a “list of reportedly affected machines” to the original blog post.

Whats more, there is a new download. Lots of readers asked for a original scanned PDF of one of the affected devices, so there is one now in the lower part of the original blog post.

Xerox scanning errors: Cat images reportedly unaffected

Today's deadpan humor award goes to Metafilter with respect to their post about the Xerox Machines: Cat images reportedly unaffected - Xerox scanners/photocopiers randomly alter numbers in scanned documents. Nicely put!

As a computer scientist, I have to express my objections though. Please be most careful when scanning HTTP status cats.

First assumptions on the xerox scan error cause

The following information has also been appended to the original article in order to give new readers all information on one single page.

It seems that the thoughts about excessive image compression have not been that wrong at all. Several mails I got suggest that the xerox machines use JBIG2 for compression. This algorithm creates a dictionary of image patches it finds “similar”. Those patches then get reused instead of the original image data, as long as the error generated by them is not “too high”. Makes sense.

This also would explain, why the error occurs when scanning letters or numbers in low resolution (still readable, though). In this case, the letter size is close to the patch size of JBIG2, and whole “similar” letters or even letter blocks get replaced by each other.

Of course, if Xerox would have chosen the patch size in a way enabling whole, readable letters to fit into the patches, this would be grossly negligent. Also, it would shed light on how these machines are tested, as when using some patch-based compression algorithm, it kind of suggests itself to test it with low-resolution, albeit still readable letters.

I am curious how Xerox is going to react and what will come out. Until then, thanks for spreading the word, please go on doing so – and of course, I am looking forward to getting further helpful emails!

Edit: Just got this by eMail – Thanks, Boris! :-)