Should there be a dpub role for a document title?

[jnurthen]

I'm really not clear how a document title which is rendered in the document should be marked up. Its not really an H1 as it doesn't take a part in the table of contents - and it is not the <TITLE> element in HTML as it is rendered in the document. Should there be a role for this?

[mattgarrish]

I'm not sure how important it is to identify the title that occurs in the text. Users typically have other means of locating that information. In EPUB, it's more important that it be in the package metadata so it's available in the bookshelf (or included in an ONIX record for sale through a vendor). In HTML, it's generally available in the document title element, even if there might be some additional site information along with it.

[TzviyaSiegman]

This becomes a bigger question outside of the world of the book. In articles, scholarly, news, whatever, the title is usually tagged as h1, but it is also the title from a navigation and metadata perspective. This means that the same information is included multiple times. This is only problematic from the perspective of duplicating info tends to lead to errors.

[jnurthen]

note - this came from discussions with @mrbhardy who may have some more use cases and input.

[mrbhardy]

We see this in complex publications, where articles have titles, but the overall publication has a separate title. An example might be an academic journal or proceedings with multiple "documents", but all packaged as a single document in the context of the larger publication. An article might have title "Change Detection on JATS Academic Articles: An XML Diff Comparison Study", but the larger publication has the title "ACM Symposium on Document Engineering 2020". Representing the title for each sub-part of the publication would be valuable.

[aleventhal]

Note that there is already a doc-subtitle.

[mrbhardy]

@aleventhal it wasn't clear that this was the intent, but that is an interesting thought.

We still have the challenge that even for more complex single document publications where the document title is clearly in the content, but then has a strongly structured (and numbered) heading structure. I'll try to attach a sample document shortly.

[TzviyaSiegman]

@mrbhardy That example is much more along the lines of the case @mattgarrish explained with EPUB. When I stitch together multiple journal articles to form an issue, I still start each article with <h1> for the title. I keep the metadata for the issue separate. There is a title page for issue metadata (as well as all the JATS stuff). (Aside: I work on Wiley's InfoModel)

[mattgarrish]

Note that there is already a doc-subtitle.

But this isn't exclusively the publication subtitle; it's any subtitle for any section with a heading (i.e., to discourage the practice of consecutive numbered headings).

Don't get me wrong, though, we also had a proposal for doc-title in the 1.0 FPWD, broadened out beyond just the publication to avoid it being too narrowly interpreted. I'm not sure how helpful it is for books, but it's not necessarily harmful to have, either.

But, again, the consensus then was that if there were titling issues that needed addressing beyond what HTML offers this module wasn't the best place to begin patching that. Maybe things have changed, and I'm just giving a boring history lesson, but I'm concerned we end up opening the debate about whether the vocabulary is about semantics.

[aleventhal]

@mattgarrish IMO we should skip this for now. Let's wait until DPUB-ARIA really starts making an impact, and then base improvements on what evidence says will improve the UX.