As the things we build grow more complex, labeling them becomes harder. Do we try to label them in a way that tries to simplify the things they represent and obscure underlying complexity? Or do we want labels that more accurately reflect the underlying systems?
In the Hinton reading I enjoyed his analysis of the labeling of Apple’s iCloud service. I’ve often thought that cloud computing systems are difficult to understand, but have never considered how the labeling plays an important role at shaping expectation and helping users construct a mental model that allows sufficient understanding of the system.
Hinton argues that “cloud” obscures complexity rather than making cloud services understandable. Its often difficult to know what files are on what devices, what are the consequences of deleting a file on a local computer, etc. He argue that ontology is the reason for the problems, but offers no solution. From my computer science background I know how difficult and confusing managing sync across devices is. How to know what is the correct file? If there are duplications or differences how are they resolved? How to delete a file from one device and not another. What if you want to delete a file from all of these devices?
I agree with his point that in the attempt to simplify to such a degree, we have actually made using the system harder. There is no model for users to understand what is going on behind the scenes. The difficulty is presenting enough information and and creating an appropriate model so users can understand what is going on and have control over what happens, but not be overwhelmed with complexity.