- Contrast is designers currency. It's like a fuel in a car. Use as little amount of contrast as possible to distinguish two elements. Create distinctions only in its range. One can always subtract contact, but at some point there is no way to increase the contrast. The most contrasting element will set the ceiling of all the contrast in the composition.
- Use grids. Everything looks way better with them
- Avoid designing as a committee. Even if the group of people are carefully curated averaging decision will leave you with sub-optimal outcomes.
"If you want to make the wrong decision, ask everyone" Design as committee on wikipediaNaval Ravikant
- when FB invented newsfeed people were appaled. Number one FB group was bring back the profile centric interface. Decade later all social media platforms are newsfeed centric. Zuckerberg on this 19:35
- Vibe over content of the company. The function of a brand communicating what the company does is overvalued. The function of a brand communicating vibe of the company is undervalued
Systems & Interfaces mental models
Designin systems general
- Start with a simple system that works. A complex system that works is invariably found to have evolved from a simple system that worked. A complex system designed from scratch never works and cannot be made to work. You have to start over, beginning with a simpler system. – John Gall, The systems bible
- Sacrificing precision for consistency of rules. It's necessary to sacrifice little inefficacies in details (component level) for uniformity, simplicity, and aesthetic uniformity and simplicity of a whole (system)
- Shallow folder structure. It's better to have more elements with a shallower folder structure than deeper structure and more folders with fewer element each.
Systems are about constraints
Creating a system (interface) is building constraints on the unimaginably complex world. There is infinite amount of potential actions. Interface (and every system) is about creating a constraint, a choice of which actions to display.
Tragedy of the System 🎨
Every system (interface) breaks with its growth.
It is impossible to design a system for every case. The more entries in the system, the more different cases, the more things that doesn't work with how the system was designed.
Every entry demands a little bit different treatment. So never redesign a system because one entry needs it. Redesign the system based what most entries need or what highest priority entries need.
It's computationally almost impossible to oversee what users will need from an interface two steps ahead In your product development.
When adding entries to a system eventually it will bloat and it will be harder to navigate because of a lot out of date entries. That's why one should regularly restart the system from scratch. A good tool for it is loss aversion deleting method.
In order to take a value from the system one need to agree that it won't be ideal and some things won't be adjusted correctly or will not work.
Designing systems with Loss aversion in mind
When subscribing to podcasts or articles to a collection it's important to create ability for opt-in mechanisms. You need to opt-in to consume it and by default nothing is saved. Because it is harder to delete stuff opting-out or deleting is more mentally straining operation.
The most effective folder structure for human to process is relatively shallow. Don't do very deep folder structures with little content in each step. Instead create shallow folder structure with more content in each step (Design Details are talking about it around beginning of Jan 2020)
Subtle, vague things that break design
Once X happens nobody reads: once its a folder (like encyclopedia), once not everything is designed to be consumed in one time, once its a draft