[jats-list] Rendering expectations for footnote markers in the text flow

["Imsieke, Gerrit, le-tex gerrit.imsieke@xxxxxxxxx" <jats-list-service@xxxxxxxxxxxxxxxxxxxxxx>]

Hi,

In an STS document I stumbled across this use of a footnote in the text flow with an immediately preceding xref to it:

<xref ref-type="fn" rid="fn_1"><sup>1</sup></xref><fn id="fn_1">
  <label><sup>1</sup></label>
  <p>Available from website: <uri>...</uri>.</p>
</fn>

In our renderer, there was a duplicated footnote marker in the text flow. This is because we treat footnotes outside of fn-groups similarly to DocBook's footnote [1]:

> Processing expectations
> The footnote element usually generates a mark (a superscript symbol or number) at the place in the flow of the document in which it occurs. The body of the footnote is then presented elsewhere, typically at the bottom of the page.

Looking at the STS documentation [2], I'm not so sure any more. In the first tagged sample, there is an fn in a paragraph that is preceded by an explicit xref that points to the fn and that contains the footnote number, 1. (Interestingly, the fn doesn't contain a label -- does this mean that the marker superscript 1) will only be rendered in the text flow, not in the footnote itself?)

But this STS example seems to corroborate the explicit use of an xref for creating the footnote marker in the text flow.

However, in the JATS blue Tag Library documentation [3], the prose and the tagged samples "Author notes" and "<article-title>" seem to imply that fn elements in the text flow need not be referenced by an xref that contains the marker. The prose says:

> ... each footnote reference is marked by a number or symbol, and the text of the footnote is prefixed by the same number or symbol to help the reader associate the text with its reference(s)

This seems to imply that a marker be generated in the text flow (or taken from the label if it exists), which to me contradicts the STS example.

I could not find best practice recommendations regarding footnotes on JATS2R.

How do y'all handle this in your content and renderers? Posting this to jats-list since this is an overarching issue. If the best practice for STS deviates from the best practice for JATS, it should be explicitly mentioned on niso-sts.org. But what is the best practice?

Gerrit

[1] https://tdg.docbook.org/tdg/5.2/footnote
[2] https://niso-sts.org/TagLibrary/niso-sts-TL-1-2-html/element/fn.html
[3] https://jats.nlm.nih.gov/publishing/tag-library/1.3/element/fn.html