Joe, et al,
Not frivolous at all for those of us in a portion of the world that is not perfect by any means and not well managed; specifically, many governments.
In working with these folks I've come to realize they treat and "manage" an IBM i like a PC. That means, things are all over the place in any and all libraries, personal and generic, that you can imagine. So, regardless if you're looking for a character string of some sort in a source file or a data file, you may not know where it is or how many files contain that string so you get rid of duplicates. You many have to go on a form of a scavenger hunt to find what you need. Does that mean starting at the "root" (all libs)??? Yep.
Because the IBM i, in this case, is treated like a PC, it sure would be nice if it had tools like a PC; native as a PC does. Am I in love with PCs??? Not the way they are often used, but they are there and often abused just like some IBM i's and that's the way some portion of the world works. BTW, do I have to do this "root" based search on PC files? You betcha, for the same lack of management reasons.
As to using the SYS tables...would love to, but they are often "meta" oriented and I need to search the files, or should I say, base tables on a global basis. But, yes, sometimes the SYS tables can be helpful and I use them.
And, as Shannon said, will it take some time for the search?? Probably, but A BUNCH of time shorter than trying to find it manually.
Build my own?? I guess I'll have to, but WHY; should be there here in 2008 in the i/OS.