Reset v2.0 | Hacker News Reader

Reset v2.0

meyerweb.com

44 pointsduck11 comments

These kinds of resets often seem like overkill. Are there any reset libraries out there that do the bare minimum by normalizing against user agent defaults?

I.e., those described here: http://css-class.com/test/css/defaults/UA-style-sheet-defaul...

I'd also like to hear some pragmatic CSS professionals weigh in on this topic. From what I've seen so far, I agree that some of the resets are over the top, but I'm not sure whether the opposition to them is a result of purism over pragmatism. I really just want to speed up CSS development...

Never use them, never will. I just don't see any benefits and I hate how reset stylesheets pollutes firebug or webinspector CSS panels with inheritance nightmare.

I'm finding that they are less relevant than they used to be. We are beginning to have more consistency across browsers, fortunately. I've been working on projects lately that don't employ them with very minimal issues.

> I really just want to speed up CSS development...

Meyer's reset is not supposed to speed up development. On top of ensuring basic consistency, Mayer's goal is (was?) to avoid taking default styles for granted, think more about document and re-create all those styles for it:

http://meyerweb.com/eric/thoughts/2007/04/18/reset-reasoning...

If you just want consistency, but don't want to spend extra time re-creating basic defaults, then unbolding of strong, unitalicizing of em and few other such resets don't make sense.

http://sixrevisions.com/css/a-comprehensive-guide-to-css-res...

Good roundup of popular resets.

I don’t think this one is overkill at all. It fits in my browser window!

> Are there any reset libraries out there that do the bare minimum by normalizing against user agent defaults?

Looking at that chart, and looking at Meyer's reset, I'd imagine that your suggestion would lead to a much larger CSS file. As it is, reset2 is fairly short.

I don't mean in bytes, I mean in how many elements it applies rules to (see rimantas' comment). I'm also not sure it _would_ lead to a much larger CSS file (only 16 tags differ among browsers, and I doubt 16 CSS rules would be required for normalization).

Gotcha. But still, the chart is old, and lacks browsers people use. The benefit of the CSS reset is that it resets everything in browsers so everything is equal. Normalization wouldn't exactly do that. And if things changed, the normalization would have to change as well.

Good idea, but practically speaking, I think it defeats the purpose.

Perhaps, but both the Mozilla and WebKit stylesheets are available for reference in their respective repositories (and likely unchanged, for the most part), and IE9's purported stylesheet is here: http://www.iecss.com/

It would be doable to keep up to date if someone wanted to, though whether it's worth the effort is another thing entirely.

Reply on news.ycombinator.com