Whereas: A good tool should be both easy to learn and easy to master. 

tool users should be able to focus most of their attention on the uncertain goal, not on the use of the tool itself p. 66

This distinction is relevant for my work – training and consulting in CAQDAS – tools that support the analysis of qualitative data (such as ATLAS.ti, NVivo, MAXqda etc.)  All these tools are difficult to learn – and many people who start to use these tools are new to analyzing qualitative data – so they are learning two new big areas at once.  But even those people who are experienced qualitative analysts need to learn what the new affordances a software package can offer qualitative analysis.  These tools are both difficult to learn AND difficult to master – which could explain why they have been slow to be adopted, even though they have been around since the 1980s.