3 ms·

Reader mode is nice when it works, but since it’s hueristics-based I find it often either misses important content like images and code blocks that should be in

Reader mode is nice when it works, but since it’s hueristics-based I find it often either misses important content like images and code blocks that should be included with the article, or includes text that shouldn’t be there. Even if it selects all the right elements, syntax highlighting doesn’t work which is an issue for me since a lot of the articles I read involve code. These are issues with every reader mode I’ve tried – iOS Safari, iOS/desktop Firefox, desktop Chrome. I wish there were a standard CSS media type that sites could implement, like `@media reader`, to help specify which elements should be included and to support things like syntax highlighting.

busymom0 5y ago
I recently worked on implementing readability in my Hacker News client app and here's the resources I found most useful. Basically, if once designs their site following the correct structure, then it works correctly for both reader mode as well as accessibility. These are "standards" which sites are supposed to follow but they don't unfortunately: A very long read but this covers everything: > How to Section Your HTML https://css-tricks.com/how-to-section-your-html/ https://css-tricks.com/how-to-section-your-html/ These are specifically important for accessibility users. Another short read: > HTML5 sectioning elements, headings, and document outlines https://www.456bereastreet.com/archive/201103/html5_sectioning_elements_headings_and_document_outlines/ https://www.456bereastreet.com/archive/201103/html5_sectioni... And you can use Markup Validation Service to validate and get tips: https://validator.w3.org/#validate_by_input+with_options https://validator.w3.org/#validate_by_input+with_options Check out the excellent answer by Christian Kohlschütter on how readability works: https://stackoverflow.com/questions/3652657/what-algorithm-does-readability-use-for-extracting-text-from-urls https://stackoverflow.com/questions/3652657/what-algorithm-d...

busymom0 5y ago

I recently worked on implementing readability in my Hacker News client app and here's the resources I found most useful. Basically, if once designs their site following the correct structure, then it works correctly for both reader mode as well as accessibility. These are "standards" which sites are supposed to follow but they don't unfortunately: A very long read but this covers everything: > How to Section Your HTML https://css-tricks.com/how-to-section-your-html/ https://css-tricks.com/how-to-section-your-html/ These are specifically important for accessibility users. Another short read: > HTML5 sectioning elements, headings, and document outlines https://www.456bereastreet.com/archive/201103/html5_sectioning_elements_headings_and_document_outlines/ https://www.456bereastreet.com/archive/201103/html5_sectioni... And you can use Markup Validation Service to validate and get tips: https://validator.w3.org/#validate_by_input+with_options https://validator.w3.org/#validate_by_input+with_options Check out the excellent answer by Christian Kohlschütter on how readability works: https://stackoverflow.com/questions/3652657/what-algorithm-does-readability-use-for-extracting-text-from-urls https://stackoverflow.com/questions/3652657/what-algorithm-d...