The entire article doesn't contain compelling support for the argument you say he is making. Its a bunch of hand waving about how google gets bored after the engineering. That may be so, but he provides no tactile examples of how this is the case.

I can see your point about how he's attempting to argue the implementation is lame, but its also easy to construe his argument as 'glass-type' hardware is lame. This is an issue with the authors presentation, not with the OPs argument.