They read every symbol or character that is displayed. Normal errors are verbose but at least they have meaning. Pictures that use ascii or other characters end

They read every symbol or character that is displayed. Normal errors are verbose but at least they have meaning. Pictures that use ascii or other characters end up having all the symbols read too because there are no semantics associated with the picture. I like to ask this question on posts about new libraries or developer tools, because most developers don't consider the accessibility of the features they're building.