<i>The adverse effects of frictionless computing are all around us. Email is the first example that comes to mind. It is simply too easy to send email messages, so we all send too many (I am famously guilty of this) and receive too many. It is also too easy to add recipients.</i><p>My first thought: 'Tis a poor craftsman who blames his tools.<p>Then at the end I discovered the author is the editor-in-chief of CACM. <Sigh>
I think this article is dumb. And I think it is dumb in a way that many articles are dumb: it bounces from anecdote to unrelated anecdote--from 17th century French letters! To Facebook! To the Flash Crash!--and expects readers to tag along, mistaking the textual proximity of data for relatedness. But just because you put two things into the same paragraph does not mean they arise from the same root cause.<p>I believe these shortcomings result from iconoclastic writing. Rather than observations leading naturally to a thesis, the iconoclastic writer starts with a popular opinion and inverts it into a thesis as the first step--"hey what if instead of <i>reducing</i> friction like everyone says, we should be <i>creating</i> friction?? What if friction is actually good?"<p>From that point the thesis transforms into a research filter. Anything that shows that frictionless is bad (like the Flash Crash) is in; anything that shows that frictionless is good (like telemedicine) is out. The result is a pile of unrelated things that need to be sewn together. Like Frankenstein, the finished product is not pretty.