* naming of classes
* creating components whose sole purpose is styling
* eliminating unused stylesheets
It’s not like we became any less strict about specificity or used inline styles any more or less than we did with BEM. I find the differences to be purely logistical. Styled Components simply automates a lot of the grunt work and allows us to focus on more important things.
I've found it much easier to not have left over or unused cruft with component styling JSS/StyledComponents etc.
We import shared styles for typography, color, and some layout basics, but each component gets its own CSS file.
Used to be a hardcore Sass lover, and CSS modules took a small amount of getting used to, but worth it in the long run.
The only part I dislike about CSS Modules is having to use camelCased class names, or string references (styles['class-name']), which always feel awkward. I prefer kebab-case for CSS because it's a lot more flexible (more word separators: "-", "--", "_", "__", etc.), but I'm sure I'll adapt.
I try to simplify as much as possible to a single word and repeat them a lot since there’s no collision, so you can usually expect a .container, .items, .item in most components, all top level but SASS helping with other annoyances like pseudo selectors, parents and nesting where it makes sense.
1. naming rules almost always lead to verbose inelegant names that humans would not really use at some point. (I said almost to ward off a war)
2. The part of CSS that I am generally better at than most developers is specifically in understanding the cascade. BEM, in order to help the people who are not good at the part that I am good at, pretty much makes it impossible to use what I am good at. I feel like I'm being Harrison Bergeroned every time I use BEM. But also the names I have to give things rankle.
Then I tried in a practical, everyday development setting and found out I was just wrong :)
In the hundreds of projects I've used it on it has only make maintenance easier and intentions much more clear. Having a flatter css structure is much easier to work with too. The classes do look ugly though, and was a point of contention when I first looked at it. It was the same when I first saw JSX in React too. I hope a native scoped css solution becomes commonplace so we can avoid having to use a naming convention to achieve the same thing.
You're welcome to disagree of course, I held a somewhat similar view to you until I tried BEM in a production setting.
And of course, different products call for differing methodologies. I haven't yet switched my personal blog over to BEM, for example.
My experience and all the data I've gathered about performance, even coding the equivalent in UCSS was shocking: 400% render speed improvement. 600% loading resources improvements and radical painting speed. To set you in context: one corporate website from a bank, had an average of 230kb resource loading, we lowered it to 33ms. 650ms rendering, we lowered to 90ms. Painting speed avged 90ms, we lowered to 12ms. If you plan things properly, you can radically improve the speed.
It's true that UCSS doesn't make that much sense for very small websites. Although you will render close to a plain HTML doc for sure. It makes sense when you need to repeat and componetize many things. There's when you gain true power. For example, in our tests, we made another equivalent 590kb uncompressed CSS file to 32kb total code need in UCSS. Compressed was about 12kb against 190kb compressed file. That's how you make great improvements.
If a human isn't going to read the CSS and the computer handles it without performance problem, I say no.
Those are of course two quite important ifs!
First, not all users need your huge CSS file, in terms of downloading. Specially when they're on mobile phones and cache is crap.
A ~9000 rules CSS file will load and become COSSM slower than one that has 1000 or 10. It is linear increment. Make a test to see it for yourself.
A bloated CSS is more network time. Therefore, slower loading.
A bloated CSS will contain slower selectors, therefore slower rendering time and painting speed.
When you count all these things, you will see radical speed performance.
Rule 1 of performance: Your intuitions are not reliable.
Definitely worse than my usual approach of writing some LessCSS and trying to be quite specific.
CSS is decent enough for overall theming, but quite bad for actually laying things out.
I think CSS's cascading and inheritance are bad design choices for most of the things that people actually use CSS for.
The alternatives need to do better. Or, I need to find out which alternatives are already better.
To start with, the only good way to do CSS in JS is to use a library that compiles your styles at runtime and inserts them into the document under a style tag. If you're not doing that, you lose the ability to use pseudo-elements and pseudo-selectors like hover. Probably 95% of the time, using CSS will be easier than setting up an event system to monitor all of this stuff yourself.
React's default tutorials will almost always tell you to pass styles manually into components, and ironically this reintroduces a lot of the problems with cascade that we're trying to avoid, because it's very easy to apply multiple styles in the wrong order or forget that some style is getting overridden. It can also make it very difficult to figure out where styles are coming from in code. So definitely don't follow that advice -- at the very least use a runtime compiler that doesn't spit out any inline styles.
There are plenty of libraries that will handle runtime compilation, but many of them come with performance costs. Enterprise software has a reputation of being slow, for good reason -- it usually is slow, because it's framework heavy and handles weird edge cases, and it's developed by multiple people over multiple years. I worry about performance more at work than I do on personal projects.
Aside from performance, I was finding that it was irritating to track class prefixes for dynamically compiled styles. One advantage of BEM is that it's really, really easy to debug. You always know exactly where a style is coming from, you always know where in code to find it, and you always know what class to use in the (hopefully rare) case that you need to hack together some selectors someplace in your code.
On the ecosystem side of things, there are a number of tiny annoyances if you're using React -- Webpack can handle live style updates if you're compiling a stylesheet. If you're compiling your CSS in Javascript though, Webpack can only reload the entire page. I found this made iteration and design a lot slower. Maybe there's some setting someplace to fix that behavior, but I couldn't find it and I didn't want to waste the time trying to build my own solution.
Finally, on a personal level, I was finding that putting my CSS in the same file as my component logic was making it very hard to organize everything and quickly grep sections of the code. This is probably personal preference for a lot of people, but I've found that disorganization is one of the biggest risks for large applications. Everyone tells me that you can still keep your JS-styles in a single file, and people will be good, and they won't split it across multiple sections of code or files -- I don't think that reflects the reality of most software development. I want the people I'm working with (and myself) to fall into a "pit of success" where organization is concerned.
Putting all of that together, the biggest thing was that I was messing around with all of this technology, finding that it was not as tightly integrated or as seamless as everyone said it was, was spending a lot of time debugging stuff, and I asked myself -- why?
IMO the biggest problems to solve with external CSS is style cascade, it's conflicting class names, it's selectors that apply across multiple components. BEM solves all of these problems for most codebases I've encountered, and doesn't require any extra libraries or frameworks. I hate debugging third-party code, and I can count on one hand the number of third-party libraries I've used on the web that I've never needed to debug. So a solution that solves all of the problems we face at work without introducing edge-case behaviors or extra dependencies is nearly always the right choice to make.
There is a rare case where you want to make your own custom CSS behavior that can only be handled through Javascript, but this is usually a bad idea because of performance concerns, and it's usually better to handle that as a separate, global polyfill anyway.
For me the argument for was: maintaining a component tree and a style tree is a non-start. Too much for me to keep sense of.
But admittedly I didnt think too much further than that.
Experimented a bit with CSS-in-JS, which is designed to get rid of this separation, and I found it to ultimately be harder to maintain than SCSS. This process was also what got me hooked on BEM after being initially pretty skeptical about whether it had any real value.
I now believe that BEM would meet the needs of most teams, with the exception of a few massive software houses like Facebook. And most companies don't fall into the Facebook category, even if they think they do.
I've also taken some time to reflect on older companies I've worked for, particularly some work I did at Oracle, which was for a reasonably large team spread across several continents. I think BEM would have been good enough for us there as well.
I would propose the following test: Is your front end in a monorepo? If so, BEM is probably good enough to solve any problems with scope and style conflicts. If your code can fit in a monorepo, I think its unlikely you will ever have a legitimate reason to duplicate a component name in multiple places.
Take it or leave it, just my experience, I'm sure other people would advocate different things. But it's by no means a settled debate, different people/orgs have come to different conclusions. It's just that one side writes a lot more blog posts.