I really like where this is going. Preprocessors have had a pretty easy ride so far with very little criticism. For all their awesome, they can trick us into writing poor CSS.
Nesting often leads to overly specific, unmaintainable code. Using mixins for vendor prefixing can get pretty clumsy. It's very easy to get lost in preprocessor land and lose sight of your CSS output.
Myth looks like it will retain some of the power of preprocessors, while letting us write CSS that's much closer to the output.
Looking forward to seeing more.
Shit has no one had anything positive to say about this? Aside from the beautiful landing page built for a CSS library, it serves its purpose well and doesn't aim to replace sass or less.
The Hacker News thread is rough on two fronts:
- Nitpicky HN bullshit about implementation. It's not perfect, but that's the the Pull Request is for dudes!
- The marketing site is too slick, or it makes promises the plugin can't quite deliver. Again, HN should sack up and put in a PR.
- Complaints about the marketing site design.
- "The text is too thin and it renders badly on Windows." Ditch XP already and turn on Cleartype!
- "I didn't realize there was more content 'below the fold.'" I guess HN commenters don't ever use the internet.
I'm not sure about the implementation, but the premise Myth is un-assailable.
I'm seeing text rendering issues on Windows 7 :/ (Cleartype is turned on by default)
Source Sans Pro is a well-used font (my own application uses it). Does the font render poorly for you on the Adobe sample page?
That page looks fine, no problems at all.
Would really like to hear the developers views on this. I'm super excited about Myth, but I was jumping wholeheartedly into it without testing it (can't at the moment). At the same time, I'll also take this Gist with a grain of salt.
That dude did co-create Sass and created Compass.
I've been using this for the last hour or so, and I can't really see a reason to switch from SASS.
Hey, creator here. Glad to hear you've been trying it out :)
One of the motivating factors was that since browsers are going to be adding support for a lot of the features CSS preprocessors have traditionally had claim over, this is a way to help ease the transition. Basically instead of needing to convert things from the Sass or LESS syntax to the real syntax later you just write in the real deal from the start, and still get most of the feature benefits. (And as more things are speced, or barely speced, they'll get added to Myth too.)
The other reason you might want to switch is if you want the benefits and don't want to have to use Sass or LESS. We've actually used all 3 of the "big" preprocessors (LESS, Stylus, Sass) at one point over the past two years while working on Segment.io and now we're back to just using plain CSS. We've found that the preprocessors generally lead to less maintainable code and not the other way around. So we wanted a way we could keep writing plain CSS, but still kickstart some of the new features benefits that are coming down the pipe.
The other reason is that it's waaay faster to use than Sass, and we were getting sick of Ruby slowing down our build step since we're an all-Node company. Rework (what Myth uses internally) is awesomely fast.
I'd love to hear a little more about your guys thoughts on CSS organization/maintainability with preprocessing vs without. Blog post sometime?
Could you tell me a bit about how you plan to support changes in the specs? I'm having a rough time trying to picture the compiling and source vs. changing specs and output (that doesn't really make sense, does it? Sorry.) What is a user to do with their old sources when you update Myth with the latest and greatest from the ever-changing CSS spec? Is this a non-issue, as it's something that all the other x-processors have to deal with too?
Anyway, I'm absolutely in fucking love with this. I just found AbsurdJS the other day, and that was looking way better than SASS or LESS, but then I found this gem and just about shat my pants.
Can't see yet the benefit over Sass or Less.
It's a polyfill, a temporary measure that implements future CSS technology in today's technology. It provides some functionality of Sass or Less, in addition to functionality provided by custom polyfill mixins or packaged libraries like Bourbon.
Preprocessors provide a lot of power...but you can quickly get into big problems, especially around specificity. I see this as a kind of minimal way to get some big preprocessor bang, without the pitfalls.
Got it Andrew, it does make sense indeed. I'll give it a try in my next project. The page looks beatufiful by the way. Well done!
For sure, wish it were mine!
@Jamie same thing for LESS, less with less-hat covers more than myth. For me the syntax is longer and more complex than less.
The font rendering is really bad in my Win 7 Chrome.