5 ms·

I’m confused by your reply. Did your eye skip over the word “not” in my comment?

by leephillips 6y ago

chrismorgan 6y ago
Sorry, you are correct. Drop the final parenthetical, then, but I stand by the rest.
- leephillips 6y ago
  Yes, your other point was something I hadn’t thought about. I guess his advice would be good for some people, but I’m not sure about the 99%. It’s not uncommon to use small images as text-like elements, and when you do that, if you have this reset, you will need to explicitly make them inline. Not a huge deal, but people tend to copy and paste their own CSS libraries between projects, and redefining basic default semantics of elements may create confusion and extra work down the line when things don’t behave the way that any documentation you may consult assumes.
  - chrismorgan 6y ago
    I’m going to stick with the >99% figure. I think it is uncommon, rare even. Other than chat systems with custom emoji rendered as an img, I honestly can’t think when I last saw a site with inline <img> tags. (I’d guess it to be within the last month, maybe even the last week, but it wouldn’t surprise me if it wasn’t.) Twenty years ago, sure, but they’ve gone out of fashion stylistically as well as technically. Technically, I suspect background-image and icon fonts are both used more commonly for small iconography.
    - leephillips 6y ago
      It depends on what kind of sites you tend to visit, I guess. I see this several times per day. But I often visit pages with math on them. Before mathjax became standard, inline images was the only way to do it. And visit any Wikipedia page with math: they use mathML and have fallbacks, but the end result is, you are looking at inline images. Also, I often see such things as sparklines and similar things, also inline images. The >99% may be true of what you consume, but without data I am skeptical about the numbers. Also, some people use CMSs that substitute inline images for missing Unicode glyphs, and other things. If it’s done well, you may not even notice. You don’t know if the correct figure is 99% unless you have examined the source of a sample of sites.

I’m confused by your reply. Did your eye skip over the word “not” in my comment?

chrismorgan 6y ago

Sorry, you are correct. Drop the final parenthetical, then, but I stand by the rest.

leephillips 6y ago
Yes, your other point was something I hadn’t thought about. I guess his advice would be good for some people, but I’m not sure about the 99%. It’s not uncommon to use small images as text-like elements, and when you do that, if you have this reset, you will need to explicitly make them inline. Not a huge deal, but people tend to copy and paste their own CSS libraries between projects, and redefining basic default semantics of elements may create confusion and extra work down the line when things don’t behave the way that any documentation you may consult assumes.
- chrismorgan 6y ago
  I’m going to stick with the >99% figure. I think it is uncommon, rare even. Other than chat systems with custom emoji rendered as an img, I honestly can’t think when I last saw a site with inline <img> tags. (I’d guess it to be within the last month, maybe even the last week, but it wouldn’t surprise me if it wasn’t.) Twenty years ago, sure, but they’ve gone out of fashion stylistically as well as technically. Technically, I suspect background-image and icon fonts are both used more commonly for small iconography.
  - leephillips 6y ago
    It depends on what kind of sites you tend to visit, I guess. I see this several times per day. But I often visit pages with math on them. Before mathjax became standard, inline images was the only way to do it. And visit any Wikipedia page with math: they use mathML and have fallbacks, but the end result is, you are looking at inline images. Also, I often see such things as sparklines and similar things, also inline images. The >99% may be true of what you consume, but without data I am skeptical about the numbers. Also, some people use CMSs that substitute inline images for missing Unicode glyphs, and other things. If it’s done well, you may not even notice. You don’t know if the correct figure is 99% unless you have examined the source of a sample of sites.