I’m struggling with the topic of variants vs. booleans in relation to memory usage of hidden layers with the latter solution. The best answer on the most detailed thread I found about this being almost 3 years old, I wonder if there have been any news about this.
I’m currently designing several products that heavily rely on tables. These tables are built using different sizes (sm, md, lg) of table cells. Each cell contains a “Content” nested component, with variants for a dozen content types (text, icon, button, ranking stars, etc.).
Some of these content types rely on base components with dozens of variants (e.g., a Text element with configurations such as Title, Title + Subtitle, Title + Subtitle + Icon, Body, Body + Subtitle, etc.).
The reason I’m using such a deep nesting of components is to allow for easier updates across all tables in the future. However, I’m concerned about performance, as changing some subcomponents already takes a while due to the number of instances being updated at once.
This brings me back to the Boolean dilemma. Suppose I need six text configurations for text-based table cells:
• Title
• Title + Subtitle
• Title + Subtitle + Icon
• Body
• Body + Subtitle
• Body + Subtitle + Icon
Should I keep six separate variants, or reduce them to just two variants with two Boolean properties (Show/Hide Subtitle and Show/Hide Icon)?
Considering that I actually have many more text configurations (since I use this base text component not only in tables but also in cards and other elements), and that I’m applying it to several tables containing 100–200 cells each, I’m concerned about memory usage. If I use Booleans, I might end up with hundreds of hidden layers in each table.
Have there been any recent updates or official recommendations from Figma on this matter?
I have purchased recent high-quality UI kits and seen both solutions used. There doesn’t seem to be a consensus.
