23 ms·
No, no, no, no, no! This tripe has been circulating the 'net for a while and it really bothers me to see this gain some semblance of acceptance/traction. Yes,
by endtwist 16y ago
No, no, no, no, no! This tripe has been circulating the 'net for a while and it really bothers me to see this gain some semblance of acceptance/traction.
Yes, studies show that poorly set text causes you to spend more time reading it because it's hard to read. However, this is not the proper solution to improve recall! This is a through-and-through design problem that can be solved without destroying readability. There is most certainly a visual solution here that would improve recall, but this is not it.
Can you imagine reading an entire textbook or novel that looked like it had been printed on a greasy old photocopier? I certainly can't.
- dkarl 16y agoThis article reports it as a research result, not as a recommendation. It links to an article that's pretty dumb, but it doesn't endorse it.
- joshrule 16y agoSo, are you saying, for example, we should get subjects with good copies in these experiments to spend the same amount of time reading as those with bad copies? Or, do you think it's some other factor?