Skip to content

Instantly share code, notes, and snippets.

@brito
Last active March 23, 2017 16:40
Show Gist options
  • Save brito/9ee56d020bb9bae5ba10e7b3d0d5baf1 to your computer and use it in GitHub Desktop.
Save brito/9ee56d020bb9bae5ba10e7b3d0d5baf1 to your computer and use it in GitHub Desktop.

tl;dr;

Suggestion: Look for abstract taxonomies instead of categories, to find patterns where we can converge and agree to a subset of symbols that can be reused in multiple contexts. Iconography is a language and it will shape the apps built with it.

What

Categorize categories to identify high-value icons and provide as much flexibility as possible for the multiple cases where there won't be a concrete icon and instead we can use a concept abstractly. The better the abstractions, the more reusability we get from the set.

Why

Asking for individual icons won't scale (too many asks, not enough icons) and instead issue tracking has shifted to "suggest categories". This has the same problem at a minor scale where it may or not work because there are SO many categories too.

For everyone here and everyone else who is trying to build some UI using these icons, we can assert the following about all of them:

if UI is useful then it fulfills needs

By focusing on needs we can find out which icons and categories represent usefulness to a user. The key here is that human bio-psychological needs also overlap our devices by extension, which is why it's funny-but-true to think of a plug as a fundamental need. Without energy, devices cannot function as intended and your phone dies.

There is a few ways to go about this, I'm sure the collective internet can come up with far better:

We could use something like Maslow's hierarchy of needs to categorize the kinds of needs our UIs fulfill:

  1. Physiological, basic: Bathroom, food, rest, plug, wifi, time, space, light, energy, health.
  2. Safety: Encryption, login, backup, emergency services, help, home, lock/unlock, crypto.
  3. Belongingness: In SDT, we also find that one of three things to aim for is relatedness/purpose (1/3). This would then relate to the relative value that a person/device/service has, so we can introduce concepts like community, chat, market, currency, social...
  4. Esteem: Love. I suppose not all of these are straightforward to give examples for. I mean, if we build interfaces that honestly fulfill the need for esteem I think we are winning the internet.
  5. Self-actualization: identity, fashion, branding, shiny things, art, expression, entertainment
  6. Karma, good/bad, politics, philosophy, morals, ethics, logic, recursivity, ... dunno this one is hard.

Another way to look at needs is by audience age using Erik Erikson's Stages of Psychosocial Development. Here you get trust, autonomy, initiative, industry, etc. with similar examples for how our services and their UIs build trust with the user, let the user do things, add value, express individuality, learn, have fun and ultimately be ok with all of this at the end of the session.

Keeping in theme with SDT, there is also autonomy (2/3) which could relate to mobility, access, transportation, privacy; and mastery (3/3), a category to which we could apply Bloom's taxonomy.

Aside, but along these lines of increasing abstraction and reusability is the idea to map these symbols to their unicode representations in a pictorial language, so the icon for house could map directly to another language logogram (chinese, japanese, etc words for the same thing). Your icons would then be somewhat better mapped as a language bundle and your apps would benefit from automatic translation in more cases.

Yet another example to catalog knowledge could be using as a basis the Diderot and d'Alembert figurative system of human knowledge.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment