However, Werner Sollors writes in agreement with Te-Hsing Shan that "this commonly cited translation may actually result from the conflation and rearrangement in inverted order of two poems that are clearly recognizable on the photograph of the Chinese original; and this is also the way in which an earlier translator rendered them."

The Chinese original is on the wall; isn't the writing exactly the same for the two poems?  What are the chances of that? Sometimes the characters were written on the wall and then later carved out in what is described as a Cantonese method used.  But he's right that the poem doesn't entirely make sense though we could interpret it to mean what we like.