3 ms·
The author gets frustratingly close to insight, but ultimately retreats to "blame the user". Only now, instead of the usual accusations of stupidity and/or lazi
by GHFigs 17y ago
The author gets frustratingly close to insight, but ultimately retreats to "blame the user". Only now, instead of the usual accusations of stupidity and/or laziness, they're irresponsible for being afraid to take what they perceive as risks.
Why not ask why they're afraid? Why do they see risks that you do not? To what degree are these fears rational? What can be done to assuage these fears so that they can actually learn something positive from the experience?[1] Oh no, too many questions! Quick, blame the user!
Most of the time it seems that the fear of breaking the computer expands into something like: "If I meddle with this thing I don't understand, I might break it.[2] If I break it, I won't know how to fix it. If I don't know how to fix it, I will have to ask or pay for help. If I have to ask or pay for help, I will feel stupid, lose money, and won't understand the thing any better." (Note that none of this is specific to computers.)
Implicit here is the perception both a high probability of failure and the perception that failure will be costly. This is the kind of risk that people respond to with fear instead of curiosity. When it appears cheap to make mistakes, people are much more willing to risk making them.
In light of this, comparing this sense of fear that a computer user may experience to the "risk" of trying to make your own kettle corn is absurd. There was no perception that failure was going to be costly. Even if there was a high probability of doing so, making a mess of your kitchen costs almost nothing, is easily reversible, and involves no hidden mechanisms. You just clean it up.
[1]: Emphasis on learn because much effort has been expended on interfaces that are usable without being learnable. That is, even if you successfully navigate them, you gain no new understanding. The low point historically was probably when Windows XP shipped with a animated talking cartoon dog to help find your files.
[2]: "Broken" can be something technically trivial, but if the user doesn't know how to fix it, it doesn't matter how trivial it is--it's still broken for them. The idea that you can't break software only holds true if you believe that reinitializing/reformatting/reinstalling is just part of normal operation and not a "nuke the site from orbit" solution to real problems.