You have a point. I really don't think that the code is all that difficult, though. What you might gain is replacing brute-force, or lengthy, DDL statements for this fictitious 100 column table. If you're faced with a challenge to capture ten columns - not recommended.

Another example where this was used in a most clever way, where I first saw it as such, was in the original Match.com search engine, created back around 1996-7. The mask stored many, many values which allowed the "first cut" of a search stored procedure to eliminate many unwanted candidates for match making. From that result set, the search process then looked to the actual tables where the data was stored for further refinement, processing using WHERE and EXISTS clauses.

Again, just an approach to think about. It isn't applicable for everything that is stored in the database!