Comments on: Inspector Scrutiny

By: Chris Hester

Chris Hester — Sun, 21 Feb 2010 20:50:13 +0000

To me, the “computed style” means “the style the browser needs to display the correct number of pixels”. Correct me if I’m wrong. Therefore “medium” is meaningless to the browser until it computes it into pixels. Therefore any inspector that shows us the computed style won’t show “medium” but whatever that computes to, eg: “16px”.

By: Eric Meyer

Eric Meyer — Sun, 21 Feb 2010 04:33:25 +0000

Those aren't the computed styles, sanbor.

By: sanbor

sanbor — Sat, 20 Feb 2010 16:11:28 +0000

Eric, I like the article, but in Firebug I see all right. Here is the screenshot http://www.twitpic.com/14c92i

By: Things I Found Interesting Around February 15th | Chris Coyier

Things I Found Interesting Around February 15th | Chris Coyier — Tue, 16 Feb 2010 14:36:32 +0000

[...] Inspector ScrutinyThis particular “bug” doesn’t worry me all that much, but I wonder how many other things are incorrectly reported in a tool like Firebug. My guess is not much. I think a cool Firebug feature would be to somehow indicate which DOM values are likely to be inconsistent across browsers. [...]

By: Some links for light reading (16/02/10) | Max Design

Some links for light reading (16/02/10) | Max Design — Tue, 16 Feb 2010 11:30:54 +0000

[...] Inspector Scrutiny [...]

By: ben

ben — Mon, 15 Feb 2010 21:11:38 +0000

…So I’m starting to build the product site for http://oreilly.com/catalog/9780596157616/ (out on the first of March, with a shout-out to you in the Acknowledgements) and one of the big pieces for that site is to be a progressive look at inspectors. I’m glad you brought this up, because it’s just the sort of thing that I would’ve mentioned in the book had I not decided to save inspectors for the product site. So, um, thanks again.

By: Eric Meyer

Eric Meyer — Mon, 15 Feb 2010 20:31:17 +0000

That's right, John. I haven't directly tested this but I will predict with a high degree of confidence that body {line-height: 1.5;} will never be reported as a unitless computed value on any element. I'll also predict with a slightly lesser degree of confidence that the reported value will be an integer in Safari and Opera even when simple math tells us it can't be an integer.

By: John Lascurettes

John Lascurettes — Mon, 15 Feb 2010 18:28:03 +0000

Wait. On line-height, if it’s unit-less, it’s not computing anything other than a relative multiplier, yes? So 1.2 would still be exactly a 1.2 multiplier to children and grandchildren wouldn’t it? It’s exactly “size * 1.2″ no matter what it’s applied to. Yes, there’s an issue of what is the font size of each element as its calculated vs. how it’s rendered; but however the browser is calculating that value, the multiplier for line height will always be 1.2.

In the case of a fixed height calculated on an ancestor, 1.2em for example, that would be calculated once (and rounded or not depending on browser-OS combination) and applied as a fixed value to each descendant as the same length, regardless of that element’s font size, would it not?

By: Eric Meyer

Eric Meyer — Mon, 15 Feb 2010 15:48:34 +0000

That's exactly the focus I'm trying to encourage, Jeff. And yes, you're quite right about line-height.

By: Jeff L

Jeff L — Mon, 15 Feb 2010 15:42:03 +0000

Wouldn’t line-height have the same issue, if you specify it without a unit? The inspector will report the computed value properly, but that’s not what is inherited down to its children. I think at some point, you have to simply focus on the fact that these tools are made for professionals, and we should have an understanding of what’s happening under the hood as well.