[jjmarr]

Tools are created to reduce the amount of time one spends on a task. The greater the ratio between benefit and effort the better a tool is. If a tool requires the "barest minimum of effort" to be moderately useful, it is an amazing tool because that ratio is high. A tool that requires high effort for high benefit isn't.

That being said, people will voluntarily learn new features if it creates tangible benefits for them. But the learning curve can't be too steep—it has to be intuitive on top of what they already know with consistently increasing rewards.

[cubefox]

I think this isn't necessarily true for a company environment though. If a tool is useful and is made the "official" solution for some problem, people will be forced to learn how to use it. Example: Jira. It has lot of complex structure and it can take quite a while to understand just a fraction of its features, but a lot of companies use it. (Granted, most people don't use or need all of its features.)