If you haven’t heard of calibration cases, you should read Commoncog’s introduction to them. A brief definition:
Calibration cases are business narratives that teach you to see. A calibration case is simple: it is merely a narrative of a specific business situation. What makes calibration cases special is how they are meant to be consumed: they are organised in sequence according to concept, so you are able to do rapid case comparisons across the same concept. This exercise helps you calibrate your expectations of the business concept in question.
If you haven’t heard me rant about how much of interface design is done by people saying insighful/stupid-sounding shibboleths, you can follow my Twitter.
Maybe they have their reasons to be so non-specific, but it means that most advice is scarcely better than personal opinion. “Form follows function. Make apps fun again! Make apps look old again!” I can’t recommend that to learn from.
Instead, I used to say something like “just look at lots of good designs and use many apps” to people who asked me for a how-to guide or direction of study. Because reading Refactoring UI can only give you a few simple rules, that can nearly always be safely applied. How do you become good enough to learn/break those rules?
I now think the I wasn’t far off with my original answer, you have to gather a corpus of cases in your head. Luckily there is more than one way to do this for design:
- Make something and watch someone use it
- Find good design writeups (like this one), but those are very rare
- Spend time using apps that are considered to be good
- Read someone else’s experience with specific software
I belive that the “organised in sequence according to concept” part of the process can be done without, because design doesn’t really have that many distinct concepts. Typography, spacing, heirarchy, and content is the majority of what I do.