CLIMBING THE SMALLTALK MOUNTAI N MARY BETH ROSSO N JOHN M .CARROLL Part of the promise of object-oriented softwar e technology -- and much of the fascination wit h objected-orientedness â stems from its psycho logical implications . Thinking in terms of object s sending and responding to messages is suppose d to be more "natural" than thinking of variou s data structures being operated on by generi c procedures . Programming by using or specializing existent objects and their methods is sup posed to make code reuse accessible even t o beginners . But does natural mean usable ; does it mean eas y to learn? The small amount of empirical evidence in hand suggests that the psychologica l consequences of object-oriented programmin g may indeed be profound, but that they are no t without cost (O'Shea, 1986) . For example , interactively searching and managing a larg e code library, like Smalltalk's Class Hierarchy , can be a daunting task. I-Iaving hundreds o f classes and thousands of methods available ca n empower the experienced Smalltalk programmer . But it can be a formidable obstacle to gettin g started with Smalltalk programming . It ca n
/lp/association-for-computing-machinery/climbing-the-smalltalk-mountain-7CwbysEWlc