> Were I to rebut each of your hypothetical objections you could of course > rebut my rebuttals. I will only point out that often the coder does not have > control over how flags are defined: they are defined in some system or > "external" DSECT or similar.
I concede that readily. My thinking was that if other approaches (mine among them) were used for hopefully some non-trivial portion of the cases you posited, perhaps the number of problematic cases remaining would become small enough that they would be more of an anomaly than a recurring irritation. :-) - mb
