Most clients of mine that have pushed the issue of locking data have not had data worth protecting. The ONLY content of a binary file I can see that would give me an advantage would be if a very clever algorithm was used for a very unique (e.g. medical analytical) purpose. Even in such a case the time to find and decipher the process using that algorithm will take an inordinate amount of time.

I, myself, having reverse engineered a clients product after a fire at a that client (how is that as an argument for regular offsite backups) consider the lock bits the least important protection. The most important protection is the fact that ANY code is darn difficult to convert from object to source.

If you feel you need "super" protection beyond lockbits of your code, you are most likely holding yourself in way too high esteem.

Erik