What are the reasons for diverging from BEM?

Since the design decisions in suit are all very deliberate, I was wondering what the reasons are for diverging from the BEM syntax (suit's syntax was originally closer to BEM with it's use of underscores for example).

Is it for efficiency reasons, or were there other concerns? I could not find this in the documentation or issues, so I hope that it isn't a problem that I'm asking this.

p.s.: it's not that I prefer one over the other, just wondering why.

[necolas]

BEM's original syntax isn't very easy to read at a glance. I changed it a bit. Then MontageJS riffed on that and came up with the PascalCase variant which was even clearer.

"Standard BEM" is also a poor choice for migrations (such as is taking place at Twitter), because menu is going to collide with the legacy menu code that has no isolated structure to it. Whereas the name Menu won't, and it maps more obviously with associated JS functions and View code. It's also easier to write tools that check for conformance when the naming convention doesn't sometimes result in class names that are indistinguishable from those used in non-modular code.

For example, if you were using React to build your UI, you have JS component names that map directly with the names of the CSS convention. Every name is of the form "MyComponent".

Hope that answers your question.

Hope that answers your question.

It does, thank you!