We're operating a number of liferay installations for various customers and have mixed success in keeping the look and feel of journal articles consistent. The main reason is user/author/editor preferences for formatting their own articles and the "out of the box" available formatting options in the WYSIWYG editor.
As we're working with users and systems that are not (staffed with) full time editors, the editors are not too experienced with managing web content and (sometimes) like to format headings as "font-size: xx-large; font-weight: bold; color: #33333" instead of "<h1>". If they user almost proper formatting and think that <h1> is too large, they choose <h2> or <h3> as their top level heading. Also, paragraphs are often adjusted in size depending on the authors/editors personal screens, display settings and preferences.
Together with the default options of system-dependent font-sizes
small, larger, xx-small this leads to desaster.
This is what we usually do:
- With regard to LPS-3046 I'm changing the available font-sizes to a percentage-based list of choices instead of the display-dependent
- For certain content I offer structures/templates that take some formatting options away from the user - like positioning of images and formatting/placement of additional notes for the images. Due to the low volume of articles this is usually done after a number of common article forms is identifiable.
- With varying success I'm trying to teach users to not use the formatting options the editor offers
I'm reluctant to take away all the formatting options from the editor, as problems like "pasted word content" or "h3 as top level heading format" would still remain.
I understand that the root problem is to educate the editors about proper semantic markup but I'm also interested in supporting technical means and configuration. I don't expect a silver bullet but would like to hear your experiences.
