Depends.
Factors:
-- Are you starting from scratch, or stealing from an existing object?
-- what does the object need to do? It is one level of difficulty to make a thing, another again, to make it parametric. It is usual to commission an object that is complex and has an important commercial application. I made a stacked wall of log joints once for a company using odd-shaped joints. Make a thing that cut a wall of varied log diameters and other settings, give them cut choices and whatever - you are looking at several days work. Then get out the bugs.
-- Does it have downmarket potential? I did an eagle head for the end of a log home ridge beam. Not much call for that on the Objex-on-lion. If you make an object, is it useful and widely applicable? To approach a developer with an idea for a commercial element might be inexpensive since he can sell copies later.
-- Does it need a fancy interface and idiot-proofing? GDL is developed to the stage where preparing the interface and Properties aspects can be more time than actually writing the element. If not used exactly as designed, the object can mess up - so you spend a lot of time making sure that implausible settings get neutralized during script execution. Time time time time time.
-- Are you giving professional service? or is it just a "quick thing?" it takes time to adhere to GDL maker standards for syntax and variables to make the script decipherable by others. Wouldn't want the script police to show up. There's a fine and then GDL jail time for repeat offenders. Writing code for toilets all day as punishment.
-- how accurate does it need to be? After suffering through the chainsaw-carved-by-a-blind-man "Woman in Red Dress 65" we see the importance of accurate curvature. GDL is easy to make singly-curved surfaces but a nightmare with undulations or bulbous things that aren't lathed.
Dwight Atkinson
