There's a balance between the value that comes from having experienced it yourself and the value from being able to step back and generalise the experience of o

There's a balance between the value that comes from having experienced it yourself and the value from being able to step back and generalise the experience of other people. First-hand accounts always run the risk of generalising the particular but academic accounts run the risk of being too general. In both cases the problem comes from falling in love with your own model and not paying enough attention to inconvenient data.