I think it’s useful to understand Glide’s mission. If one truly adheres to Glide’s mission, then in my opinion one cannot support css (and markup for that matter) being a necessity in Glide. Optional, why not. Necessary, no.
Layout structure, not esthetics
Bob mostly uses css to compensate for placement/structural limitations in the builder, and I think this makes good sense. So Glide could probably add a few placement/structural features in the builder. But I wouldn’t push it with colors. Most 1 billion creators couldn’t design a good looking app if their life depended on it: giving Glide builders too many design options would just about guarantee that the output would look terrible.
Losing business / We depend on css+html for customer success
“Because of some sort of design limitation.”
Well, don’t oversell Glide’s capabilities: css isn’t and never has been part of the package. It’s pretty simple.
“Because Glide projects look the same.”
Be unique with the specific business pain point you solve internally within a business. If that problem repeats across businesses, then sell the same solution. But being unique in how the app looks is not that important in B2B (with Glide you won’t be building the next TikTok). This is especially true for apps for work (business), apps for good (non-profit) and apps for fun (personal).
CSS in Glide Pages
As it’s already been announced publicly, CSS might be optional in Pages at some point, just like in any website builder. But let’s be honest, 99.99% of Glide builders would probably be better off staying away from it.
If I had to guess, a design-related and possibly component-related announcement is just around the corner. The Glide team usually makes minor enhancements here and there, is pretty quiet for a few weeks and then suddenly releases something huge on the product side: 30+ computed columns, Glide Tables, Glides Pages, external data sources come to mind. Well, I feel Glide has been a little quiet these days, and now suddenly there’s a lot of talk about layout (css and design). This could only be related to the new computational model for Apps, but I think something else is brewing. Of course I have no idea and could be totally wrong.