Comment by splatzone
3 years ago
Interesting that the memo claims that sans serif fonts are preferred to serif fonts for screen reader users and people with disabilities. I thought the exact opposite was true. Is there any definitive research on which type of font is more accessible?
"Results showed a small, but significant advantage in response times for words written in a sans serif font."
https://www.tandfonline.com/doi/full/10.1080/20445911.2011.5...
"Sans serif, monospaced and roman font styles significantly improved the reading performance over serif, proportional and italic fonts."
https://dl.acm.org/doi/abs/10.1145/2513383.2513447
and here a whole book about that question
https://library.oapen.org/handle/20.500.12657/53344
Dyslexia associations have recommended Comic Sans as a font that allows for easier reading comprehension. Experts say that opposition to Comic Sans is ableist: https://www.thecut.com/2020/08/the-reason-comic-sans-is-a-pu...
I, for one, think all State Dept cables should henceforth be transmitted in Comic Sans, so as to provide maximum accessibility.
While I really hope the claim that hating on Comic Sans being ableist is facetious, I do kinda like the font, and use Fantasque Sans[^1] in my terminal.
[^1]: https://github.com/belluzj/fantasque-sans
1 reply →
That last link (the book) is a fantastic literature review, but every chapter basically concludes that despite some results going each way, there is no statistically significant readability difference between sans serif and serif typefaces, whether on paper or on screen, or whether the reader has various disabilities.
I had a laugh about that.
Isn't 'small but significant' quite ironic.
Fascinating, thank you
Sans serif fonts are preferable only on low-resolution displays (this includes 1080p) or at small point sizes, because in this cases the good serif fonts cannot be rendered correctly.
The classic serif fonts cannot be rendered correctly at low DPI not only due to their serifs (which must be enlarged at low resolutions, to avoid their disappearance; this transforms all classic serif typefaces into uglier slab-serif typefaces), but also due to their "contrast", i.e. because they have both thin and thick lines (which are equalized in width at low resolutions, distorting the glyphs). Some of the better modern sans serif typefaces, e.g. Optima, also cannot be rendered correctly at low DPI, due to their variable-width lines.
On high DPI displays, the preference for sans serif or for serif is driven mainly by the familiarity with the tested typefaces, so it is impossible to predict from the results of a half hour test, where the people may see for the first time some typefaces, which of them they would prefer after using them continuously for six months.
“Disabilities” is an extremely wide spectrum, and it’s pointless to try to group people on that scale.
Let me introduce a term to your venacular that will serve you well: 508 compliance.
Not sure I get your point?
3 replies →
Am I being thick, or does a screen reader not care what the font is?
One of our employees is blind, and screen readers (JAWS is what he uses) can be fairly finicky. For a case where you are looking at or editing the document in MS Word it doesn't matter, since if the text is readily accessible it will ignore the font. However, in something like a pdf or image it needs to OCR the text, and certain fonts are easier to read.
I happen to know this because we ended up needing to rework our training documentation, which we supply as pdfs that were exported from Word. The screen reader needs to figure out the order of the text in the pdf, and by default word can end up exporting the text laid out in a way that causes the screen reader to parse the text out of order. There are accessibility options that can resolve this, but we ended up changing the font since that resolved the problem without needing to ensure the correct options were set everytime someone needs to update the training manuals.
There are PDF's to embed OCR info on it.
The issue is with OCR. Imagine something that was printed(for being physically signed or filed into a cabinet, for example) and rescanned into an image or PDF. Both printing and scanning is lossy with lots of noise and artifacts. The article is saying that serif fonts are harder to OCR.
For example, something like this https://h30434.www3.hp.com/t5/image/serverpage/image-id/6945...
I wouldn't expect it would make a difference either, but the memo made this claim