See my other comments in the thread to find out about me. 1. This is true. Weeks to month(s) is usually enough to learn enough to roughly be aware of the kinds

See my other comments in the thread to find out about me. 1. This is true. Weeks to month(s) is usually enough to learn enough to roughly be aware of the kinds of problems you can solve with a tool. Finding the right and interesting problem is not quite as easy though. Usually after I know enough to have a feeling about the kinds of problems, I stop learning the new tool. It's not interesting to learn more just for the sake of learning. Once I find a good problem(s), I put months or years in learning the tool and solving problems with it. One of the reasons I moved away from medicine is because you can't learn while solving problems. You have to learn a lot upfront before you are allowed to the "field". This is really justified, but I lost interest. 2. This is not really always the case. There are fields (medicine for example), where true innovation is not always result of "deep knowledge". For example antibiotics, vaccines etc. In those cases true innovation happened because of trying to solve a problem (and quite accidentally), not because someone by his really deep knowledge predicted these phenomena. I wouldn't try to judge which way is "better". And I wouldn't use the word "unfortunately" for these. Probably we need very much both to progress.