the purpose of the <text:author-name> etc text fields is to display the
metadata elements from meta.xml somewhere in the document, so they do
apply to the whole document.
the reason why these may be children of a bunch of other elements is
simply that they occur in paragraph content.
where there is authorship information for elements in the document, such
as <office:annotation>, the <dc:creator> element is used.
i think the only way to add authorship to arbitrary elements currently
is RDF metadata, for example with the <text:meta> element, which is
more-or-less a <text:span> with an xml:id or RDFa attributes.
--
Michael Stahl
Senior Software-Entwickler LibreOffice
–––
allotropia software GmbH
Versmannstr. 4
20457 Hamburg
Germany
–––
michael.stahl@allotropia.dehttps://www.allotropia.de–––
Registered office: Hamburg, Germany
Registration court Hamburg, HRB 165405
Managing director: Thorsten Behrens
VAT-ID: DE 335606919
–––
Original Message:
Sent: 1/14/2025 1:09:00 PM
From: Francis Cave
Subject: RE: ODF-Authorship-Metadata uploaded
I think that the descriptions of <text:author-name> and <text:author-initials> in 7.3.7.1 and 7.3.7.2 are incorrect. They state that they represent the full name/initials of the "author of the document". However, these two elements are child elements of <text:h>, <text:p>, <text:span> and <text:a> (among others), so the name and initials of the author is actually specific to that element. I cannot believe that these two elements have been made children of such granular text elements without it being understood that they are to be used to specify the author of the parent element.
Tracked changes can include identification of the author of the change. When tracked changes are accepted, such author information needs to be preserved in some way.
I consider authorship metadata to be important from several points of view. We have talked about how one might differentiate AI and human authors. We have also talked about the need to describe the authorship of documents that are the result of collaboration. More generally, I think we need to think of documents as being creative works, and it is important from a variety of legal, commercial and moral perspectives that there should be ways of recording and determining the provenance of the content of documents. A major topic in international standardization circles currently is the whole business of trust and trustworthiness. How can one determine whether to trust the content of a document, if one doesn't know its provenance?
Francis
Original Message:
Sent: 1/13/2025 11:22:00 AM
From: Francis Cave
Subject: RE: ODF-Authorship-Metadata uploaded
Patrick
Thanks for putting together this draft!
In ODF 1.4 I note that:
- 4.3.2.6 is <meta:initial-creator>, not <text:author-name> as in your draft
- 7.3.7.2 is <text:author-initials>, which I think is what you meant to type!
I am puzzled as to why <text:span> (6.1.7) and <text:a> (6.1.8) do not have the xml:id attribute, while most other text elements do have it.
Best wishes,
Francis
Original Message:
Sent: 1/12/2025 10:07:00 PM
From: Patrick Durusau
Subject: ODF-Authorship-Metadata uploaded
Submitter's message
Greetings!
A very rough cut on the ODF AI/Human editing metadata draft!
Hope everyone is at the start of a great week!
Patrick
-- Patrick Durusau
---------------------------------
Patrick Durusau
Patrick Durusau (Personal)
Covington GA
---------------------------------</text:span></text:meta></dc:creator></office:annotation></text:author-name>