5 ms·

Perhaps a naive question, as I don't understand this domain very well. How are these fonts different than, say just changing the color of the text?

by hdesh 4y ago

WorldMaker 4y ago
Colors (and gradients) per-stroke rather than per-glyph. (Coloring all of an "A" versus the three or more strokes that make an "A", for instance.) The most obvious examples are emoji which may need hundreds of strokes in many colors to be properly colored as we see them in most places (that aren't HN). Many emoji were rendered as bitmap images on many platforms to be able to get the colors correct prior to font support for color fonts.
hnarn 4y ago
Did you click the link and note the headline saying "What’s so special about the COLRv1 font format?" before asking this question?
nerdponx 4y ago
You can do something like building a brand logo and embedding it in webpages and documents as actual text, in rather than an image.
- jfoster 4y ago
  If you did that, what's the advantage? Instead of loading an image file, it loads a font file, which seems a lot like an image file when used that way.
  - lalaithion 4y ago
    Text is easier for screen readers to read, easier for search engines to parse, easier for control-F to find, easier to copy and paste. Text is a better format for text, and this allows for companies to embed their logotype design as text with a specific font as opposed to incentivizing them to use an image.
    - jfoster 4y ago
      Screenreader: image "alt" tag should be equivalent, shouldn't it? Arguably better, because the intent of it is more widely understood. Search engine parsing: image "alt" tag should be equivalent again. Copy-Paste: Arguably not better with a font. A raster image will be the same regardless of where you copy-paste it. A logo implemented with a font will only look the same when copy-pasting between applications that are aware of that font. If every company creates their own font-logo, it will just break copy-paste everywhere except their domain.
      - lalaithion 4y ago
        WCAG specifically asks designers to avoid text in images because they're not equivalent. If I embed an image in some text, like `foo <img alt="bar"> baz` then I doubt any search engine can find this for the search query `"foo bar baz"`. I would like to copy and past text-like things as text, without having to copy and paste images. When I want to copy and paste text, I rarely care about matching exact styles. In cases where I want the exact style, I can just screenshot.
rsoto 4y ago
They're multi-color. For example, you could have a font with a 3D-effect, its shadow and lighting could be different from the actual block. Look at the first image of the post to see what I mean, the big "Color Fonts" text.

Perhaps a naive question, as I don't understand this domain very well. How are these fonts different than, say just changing the color of the text?

by hdesh 4y ago

Perhaps a naive question, as I don't understand this domain very well. How are these fonts different than, say just changing the color of the text?

WorldMaker 4y ago

Colors (and gradients) per-stroke rather than per-glyph. (Coloring all of an "A" versus the three or more strokes that make an "A", for instance.) The most obvious examples are emoji which may need hundreds of strokes in many colors to be properly colored as we see them in most places (that aren't HN). Many emoji were rendered as bitmap images on many platforms to be able to get the colors correct prior to font support for color fonts.

hnarn 4y ago

Did you click the link and note the headline saying "What’s so special about the COLRv1 font format?" before asking this question?

nerdponx 4y ago

You can do something like building a brand logo and embedding it in webpages and documents as actual text, in rather than an image.

jfoster 4y ago
If you did that, what's the advantage? Instead of loading an image file, it loads a font file, which seems a lot like an image file when used that way.
- lalaithion 4y ago
  Text is easier for screen readers to read, easier for search engines to parse, easier for control-F to find, easier to copy and paste. Text is a better format for text, and this allows for companies to embed their logotype design as text with a specific font as opposed to incentivizing them to use an image.
  - jfoster 4y ago
    Screenreader: image "alt" tag should be equivalent, shouldn't it? Arguably better, because the intent of it is more widely understood. Search engine parsing: image "alt" tag should be equivalent again. Copy-Paste: Arguably not better with a font. A raster image will be the same regardless of where you copy-paste it. A logo implemented with a font will only look the same when copy-pasting between applications that are aware of that font. If every company creates their own font-logo, it will just break copy-paste everywhere except their domain.
    - lalaithion 4y ago
      WCAG specifically asks designers to avoid text in images because they're not equivalent. If I embed an image in some text, like `foo <img alt="bar"> baz` then I doubt any search engine can find this for the search query `"foo bar baz"`. I would like to copy and past text-like things as text, without having to copy and paste images. When I want to copy and paste text, I rarely care about matching exact styles. In cases where I want the exact style, I can just screenshot.

rsoto 4y ago

They're multi-color. For example, you could have a font with a 3D-effect, its shadow and lighting could be different from the actual block. Look at the first image of the post to see what I mean, the big "Color Fonts" text.