Joe Edelman suggests an unusual taxonomy for discussing how a system relates to users’ and creators’ goals:
{Funnels} are systems which {(for better or worse) push people towards a goal which someone else invented}.
Examples of {funnels} include: {Quantum Country, Facebook (when it’s trying to get you to care about others’ affirmation), most effective altruist spaces, multi-level marketing schemes, etc}
{Tubes} are systems which {help people achieve their own goals more quickly}.
Examples of {tubes} include: {Amazon, Google search, Anki (when used for material you really want to remember), an abacus, a GPS, etc}
{Fields} are systems which {facilitate or amplify a non-goal-directed activity}.
Examples of {fields} include: {a drumming circle, a friendly correspondence, improv class, Earnest Twitter, an empty canvas next to paint, etc}.
Funnels and tubes work best when relationships are {transactional} because {goal-orientation emphasizes efficiency and predictability}; by contrast, fields emphasize {expressivity, surprise, discovery}.
Joe memorably relates how our impulses arrange the purposes of funnels, tubes, and fields in our life: {the point of funnels and tubes is to be over, so we can get back to our fields}.
Why has system design mostly focused on funnels and tubes rather than fields? Perhaps the most important problem is that {we don’t have good ways to specify the success metrics of fields}.
It’s interesting to notice that my ideas around Spaced repetition memory system function in all three of these modes.
Quantum Country is like a funnel because {it invents a novel goal (“memorize the contents of this book”), convinces the reader that they want to achieve it, then helps them do it}.
Memory systems are like a tube when {used as a magic wand to serve intellectual excitement (see Spaced repetition memory systems make memory a choice)}.
Memory systems are like a field when {used to purposefully shape how we relate to or experience the world by directing our attention over time (see Spaced repetition systems as catechism)}.
Joe’s description of fields highlights one key deficiency of my ideas around Programmable attention: {they’re still single-player-centric (see Think harder about sociality and memory systems)}.