I’m not confident in coming up with good design. However, I know a good design when I see one (at least I’d like to think I do). One approach I like to take when coming up with a design is throwing things on to the screen and then fiddling with them as I go. Many small changes later and hopefully I have a decent design set up.
All those small changes can become tedious though, so streamlining that process will save a lot of time. Having a UI Framework potentially does just that. It changes the html and css code to something more in tune with design ideas. “I want a menu that goes across the top here with a couple of dropdowns.” Having a UI framework can make implementing that relatively painless (along with the subsequent deletion). Of course, you can still sprinkle in a bit of html and css for fine adjustments, but a UI framework will help you to get that concept on screen right away. I’ve been learning how to use Semantic UI for a bit, and while I can’t say I’m very proficient at it, I can definitely see how it can help me to design a page easily.
But why not just use raw html and css? To me, it is like the difference between low and high-level programming languages. If you want complete control over your design, then maybe you want to use the low-level language. However, the high-level language offers to streamline the process without losing much at all. From my experience so far, Semantic UI is similar. It restructures the logic in a way where you can think less about code and more about design. Just like how when you write a for loop, you think more about the function of the for loop more than how it is actually run by the processor.