Well, with a quantum of class, they could 1) acknowledge that their judgement was erroneous, 2) publish that as a correction, listing the project, ideally

Well, with a quantum of class, they could 1) acknowledge that their judgement was erroneous, 2) publish that as a correction, listing the project, ideally with a small apology and at least some details on how the entry would have been judged if not for the GitHub link and 3) create a separate prize for "projects our judges learned valuable lessons from". As a good example of a similar precedent, look at this entry from 1994: https://web.archive.org/web/20200214094614/http://www0.us.ioccc.org/1994/smr.hint https://web.archive.org/web/20200214094614/http://www0.us.io...