3 ms·

Surprised no one has mentioned Flesch-Kincaid yet. https://en.wikipedia.org/wiki/Flesch%E2%80%93Kincaid_readability_tests https://en.wikipedia.org/wiki/Flesch%

Surprised no one has mentioned Flesch-Kincaid yet. https://en.wikipedia.org/wiki/Flesch%E2%80%93Kincaid_readability_tests https://en.wikipedia.org/wiki/Flesch%E2%80%93Kincaid_readabi... There's also social register. More complex writing/speaking assumes knowledge of cultural markers which are intended to exclude those who aren't familiar with them. So writers/readers can use them as caste markers. Simplification has to be a mix of as-simple-as-possible with as-neutral-as-possible. I don't think the examples hit that, because they read as if they've been simplified, while successful simplification doesn't. I like the info pages on the UK NHS site. They're as uncomplicated as they can be, they get the key facts across, but they're also very neutral. https://www.nhs.uk/conditions/flu/ https://www.nhs.uk/conditions/flu/

nonrandomstring 5y ago
> I like the info pages on the UK NHS site. They're as uncomplicated as they can be, they get the key facts across, but they're also very neutral. Not just NHS, all of UK .gov online, leaflets and telephone services have adopted a super high standard of communication in recent years. Compared to the deliberately twisted legalese for tax, benefits, and services in the 1980s it's a huge transformation. Something has changed in the core culture.
IshKebab 5y ago
It is mentioned in the article. They give it as an example of a pretty useless reading level algorithm. I would have thought GPT-3 or similar could probably solve the reading level algorithm problem very easily. I thought that's where the article was going to be honest. I agree the NHS pages are very clearly written.