As a C++ developer, one of the greatest challenges of working with Filemaker has been the discomfort of reliance on cloning. Cloning is copying the same functionality to a new context, and altering it slightly to fit. Cloning is bad. Very very bad. When functionality is copied, the bugs come along. When a bug is fixed, or functionality enhanced, it has to be done over and over and over, as many times as you’ve cloned the original code. Using cloning means that you have set your application on a collision course with extinction. Absolutely. Either it will get too expensive to maintain (think bug circus), or it will hit an extensibility wall. It’s only a matter of time. Just don’t do it. Having no alternative, watching my solutions bloat and bloat, and then bloat some more – it made me wonder if I’d made a terrible mistake.
Panic had almost set in when I came across SeedCode and modularfilemaker.org. Not stuff you see in the FM user guide, but good stuff. Like almost everything in the FileMaker world, getting things done right (or even at all) means taking advantage of loopholes and tricks in the FileMaker platform. It isn’t a language. It isn’t designed to give you latitude to do amazing and unexpected things. It wants to be a platform that doesn’t need software developers. Not even close, and far too expensive anyway. So here are the engineers stuck in this confused pseudo-developer world, and here they are finding ways to do amazing things.
Check out this methodology to make a frequently cloned piece of programming reusable. Wow. If it works, and I assume it does, maybe I can stick with this after all. Even if the veterans characterize programming Filemaker as “fighting” with it.