Understanding the philosophy of Materialish can help you to use it successfully.
Materialish's goal is to provide visual components with a Material Design-like appearance.
What this means is that it is not always the case that the components will provide all of the interactions that you need.
For example, the modals in Materialish do not include features such as scroll locking, animations, focus trapping, or fixed positioning. This is an intentional omission, and there are several reasons for Materialish's focus on visual components.
One reason is longevity. New patterns and APIs frequently emerge in the React ecosystem that provide improved solutions to old problems. By separating out the visual components from trending implementation patterns, we hope to increase the library's lifespan.
A second reason is flexibility. The unique requirements of applications often necessitate specific implementation details that can be difficult to abstract. Consider dropdown menus. It would be impossible for Materialish to provide a single functional dropdown, or even a collection of functional dropdowns, that solve the needs of every application out there.
What Materialish can (and does) do is provide a common visual languages to support the wide variety of interactions you see in components like dropdowns. With these visual components, you can add interactions to a dropdown as needed. You may recognize the value in separating these distinct concerns if you have ever struggled to change the interaction of a component that provides them for you.
We recognize that you cannot build an application without interactions, and we plan to provide guides and examples that demonstrate how to add common interactions to Materialish components, but they do not exist at this time.
Generally, our preferred method for introducing interactions is through render prop components that do not render to the DOM themelves. One example is the Downshift library, which is a great solution to implement dropdowns and autocompletes.
Materialish's visual design is influenced by several sources:
Materialish visual language is dynamic, and will continue to change and adapt as time goes on.
Theming and extensive customization of the visual language of Materialish components is not a design consideration of the library. Although the library supports "basic" adjustments, such as the primary color, base font size, sizing, and font, you will fight the library if you attempt to make more drastic changes to the visual appearance of the components.
If the out-of-the-box visual appearance of Materialish does not align with your requirements, then it may not be the right solution for your project.
Materialish is meant to be lightweight and unopinionated. You should be able to use Materialish as long as your application can render React components and load CSS files.
When it comes to CSS, Materialish uses "plain" CSS with a BEM class naming convention to reduce the likelihood of collisions. Whenever possible, we strive to keep the specificity of selectors low so that you can more easily override styles when needed.
The authors recognize the growing popularity of CSS-in-JS, so we feel the need to add a note about our decision to not use CSS-in-JS.
Most importantly, we do not believe that using CSS-in-JS would have provided any value over using "plain" CSS, given the project's design goals.
Secondly, different CSS-in-JS libraries have different APIs. Choosing any one library increases the overall learning curve of Materialish if we were to to expose that API publicly.
Thirdly, the CSS-in-JS space is evolving rapidly. There is no way to know which library will win the overall market share over the next few years, so we believe it is a risk to adopt any one library at this time.
By implementing Materialish using plain CSS, we believe that the library is more likely to remain useful for a longer period of time, with fewer large refactors.
Materialish is built for internal applications at Netflix, which affords us the ability to target newer browsers only.
If your project requires support for older, or legacy browsers, then Materialish may not be the right solution.
Materialish was built for an internal suite of applications at Netflix. At this time, small screens support is not a priority. Typically, we won't do anything that completely breaks a small screen device (if we do, let us know). Instead, what this means is that some parts of Materialish won't be optimized for small screens.
Perhaps the best example of this is the Table component. It isn't responsive.
Our small screen support is in some ways embodied by this very website. Although it generally works on small screens, there are some improvements that we would like to make to it.