So I’ve separately mentioned that:
- it’s very hard to write a good WYIWYG style editor that can adequately
deal with real world styles
- we need to improve style previewing to catch problems with styles
To be concrete as an illustration of both issues, in general problems can
happen when you assume data will be present that in many cases will not be.
This is particularly so with different kinds of contributors.
I can think of the following common cases:
- author substitution: what to do when there’s no author, as is common
in reports and news articles
- translators and other non-primary contributors
So as we work on this, we need to include test data for previewing that
includes these sorts of data conditions.