← Voltar para o blog

Design do Caos e outras coisas

Eu tava lendo o artigo 'Chaos Design' e esse trecho me pegou:

"So what’s it like to be good at interface design in 2019? As far as I understand it’s pretty easy. You just need to make sure you get this simple checklist done.
Loads instantly
60fps
Usable with a keyboard
Screenreader friendly
Accessible contrast
Award winning content
Semantic HTML
Works on the following screen sizes: All of them, also watches
Works well in low light environment
Works well in a bright light environment
Supports right to left text
No scroll jank
Handles variable length content
Works if CSS doesn't load
Works if JS doesn't load
Well balanced if user has different zoom level
Looks amazing
Has proper error states
Only use design system
Hover state
Active state
Focus state
Loading state
Empty state
Easy to use navigation
No superfluous animations and handles reduce motion
Consistent with other parts of interface
Follows all of the fundamental UI patterns
100% accessible, no bugs
Everything is fully documented
Make it pop"

E ele diz como tudo é muito frágil, como o design de hoje parece estagnado (isso pq o post é de 2019), e sugere o Chaos Design

What if we had… Chaos Design

The discipline of experimenting on a component in order to build confidence in the components capability to withstand turbulent conditions in production.

Chaos design in practice. If we were to reword the above a bit. We might come up with something like:

To specifically address the uncertainty of distributed components at scale, Chaos Design can be thought of as the facilitation of experiments to uncover weaknesses in a components implementation or design. Define a ‘steady state’ as some measurable output of a component that indicates normal behavior. Hypothesize that this steady state will continue in both the control group and the experimental group.

The harder it is to disrupt the steady state, the more confidence we have in the behavior of the system. If a weakness is uncovered, we now have a target for improvement before that behavior manifests in the system at large.