Acu-Rite scale datamatrix barcode encoding scheme

Does anyone happen to know how the 2D datamatrix barcode on the acu-rite scales is encoded? I was hoping that the individual accuracy spec would be encoded into the barcode, as the models I have seen only list the resolution in plain text (e.g. 0.5 um, 1 um, etc...) yet the manuals list two different accuracy specs, ±3 um and ±5um/1000mm.

I made an excel sheet to do the decoding but converting to ASCII yields 100% gibberish. The spec documents (for the "standard" datamatrix barcodes) suggests some "randomisation" using a pseudo-random code scheme (actually two different schemes) but reversing the algorithm produces yet more gibberish. I would not have thought that they would be encrypted, so my conclusion thus far is that it is either a very large unique identifier (18x18, or 40 8-bit codewords) used during production or they have some byzantine encoding/decoding scheme that is not to standard. The numeric serial number printed on the same readhead has 8 places, or 100,000,000 unique numbers, which is likely more scales than they will ever make. There is way more entropy in the 2D code than the serial number, so I'm therefore leaning towards it being some bizarro encoding system, hence my question.

Also, I did try searching the forum using google + "site:practicalmachinist.com" but came up with nil. Hopefully I'm asking the right crowd as well, I kind of figured the metrology wonks would be the most likely to know the answer...