HTML verbosity

[dgarijo]

It's difficult to edit the generated HTML due to its verbosity.
I would like to maybe separate the content of the sections in the html as markdown files or similar, so it's easy to edit.

[davidjsherman]

Do you intend to provide the markdown files in the final RO, or only produce them as intermediate files that the user can edit before re-rendering the HTML? I thin there are three scenarios:

1. Only the markdown files are provided in the RO. Since markdown is designed to be readable text files, this might be good enough.
2. Both the markdown and rendered HTML files are provided in the RO. Then there is some redundancy because the same information is provided twice, and a risk of inconsistency, if one version is changed and the other is not updated.
3. Only the rendered HTML is provided in the RO. The markdown files are generated locally for the user but not put in the RO. This makes it hard to update an RO, if that ever happens.

In all cases we need a separate function to update the rendered HTML from the markdown.

---

For (2.), I suppose that theoretically the rendering of the HTML from the markdown intermediate could be added to the RO as a workflow, with the markdown as input and the HTML as an output. I can't decide whether this is an elegant solution, or an unnecessary source of complexity. We would not want all RO to be classified as workflows, just because formatting the metadata used a workflow.

[dgarijo]

@davidjsherman thanks for sharing your thoughts.
I always think of the HTML as a serialization of the RO for humans, and therefore I had not thought on including it IN the RO itself. My idea would be to have sections as markdown, which would then be loaded from the main HTML and rendered using one of the existing JS libraries that are available. Another question is whether this method could be used to modify some of the metadata of the RO itself (my intention was not to, initially)