(Note that the published code by Radford is not GPL—in fact, it has no license statement at all, which means that you cannot redistribute it, or code derived from it, in any form. Clean-room re-implementation is not about “GPL avoidance,” it’s about working around copyright law’s definition of “derived work.” In this case, the paper itself contains code excerpts, so it is especially tricky to re-implement the algorithm in a clean way from a copyright perspective.)
The author emailed me to say he was planning on posting a cleaned-up version of the code under a free/open-source license (his initial impulse was GPL, but I’ve urged him to consider a more permissive license for a small library routine like this). Not sure about the timeframe.