Authority of assertion in repository contributions to the PID graph

George Macgregor
University of Glasgow
2023-09-07
https://purl.org/g3om4c
https://orcid.org/0000-0002-8482-3973

(w/ Barbara S. Lancho-Barrantes & Diane Rasmussen McAdie)

Slides available: https://doi.org/10.17868/strath.00085167

Overview

This paper: to outline an emerging challenge in repository metadata and users' discovery of repository content

Exploring issues when levels of 'authority of assertion' are lacking or are uncertain

• Emergence of increasingly fluid, distributed, and fragmented scholarly objects
• Emergence of persistent identification as a (partial) solution
• The growing scholarly graph, e.g. 'PID graph'
• Implications for repositories and the question of 'authority of assertion'
• Scholars' 'PID literacy' as an impediment to satisfactory scholarly objective description (and ergo discovery) [1]

Raising a manifesto for action?

Metadata conventions...

• Generating statements of fact about (often) tangible things...
• Even in formal digital libraries, objects (often) static things with predictable descriptive metadata attributes (e.g. digitized objects, etc)
  • Linked Open Data (RDF/URIs)
• "...change is slow" [2]

[Photo, previous slide ] Dresden-Neustadt. Sächsische Landesbibliothek - Staats- und Universitätsbibliothek Dresden, Europeana (CC-BY-SA)

Open scholarly infrastructure

[3]

Distribution, fragmentation, fluidity

Disorientation grows for metadata creators with research domain, but also for users in discovery for research content...

• Increasing levels of distribution, fragmentation, and fluidity
  • Emergence of open, distributed, multi object (often 'non-standard') transient scholarly objects
• Multi-object exists elsewhere (e.g. METS package) but all components under direct custodianship where authority of assertion exists
• PIDs increasingly central to linking, description, disambiguation, discovery

What are PIDs?

PID = persistent identifier

• PIDs come in many flavours, e.g. DOI, ORCID, RAiD, ROR, etc. -- and beyond!
  https://doi.org/10.17868/strath.00085975
https://orcid.org/0000-0002-8482-3973
https://ror.org/05j0ve876
• Enables long-term identification but also 'actionability' because they are (generally) formed as URIs
• PIDs usually underpinned by some form of registry which registers new PIDs and resolves them (and contains metadata!) (e.g. DataCite, CrossRef)

Why have PIDs?

PID = persistent identification

• Provides unique and persistent reference to an entity normally accessible over the Internet
• Maintenance of the scholarly record - citability, verification, reproducibility, replicability [5], [6]
  • 'Reference rot' / 'link rot' - e.g Klein, Van de Sompel et al. [6], [7]
• Uniquely identify entities on the web; enables discovery these entities -- happy times!
• The quest to encode PIDs within metadata to link and grow the utility of data

PID graph

.... or Linked Open Data déjà vu... ? [4]

TIB – Leibniz Information Centre for Science and Technology - PID Service (CC-BY)

Exploring the graph with Neo4j...

RAiD envelope diagram. From Janke et al. [1], CC-BY.

Repositories on front line of PID data contributions

distribution + fragmentation + fluidity

Such metadata key to making scholarly works:

Interpretable...

• to contextualize; to provide provenance

Support reproducibility...

• to facilitate reproduciblity, verification, replication

Discoverable...

• to enable discovery through a variety of access points, some unconventional

What is 'authority of assertion'?

Concept of 'authority of assertion' exists but an under-research area; no definitions

• authority = "official permission or the legal right to do something"- Cambridge Dictionary

• assertion = "a statement that you strongly believe is true" - Cambridge Dictionary

Ergo, could we propose the following:

"The extent to which there exists a right to make metadata statements about objects and their relations, such that these statements are reliable, verifiable, and true"

The 'bread crumbs' and authority of assertion

Assertion and 'PID literacy'

• Components of multi-object, non-standard scholarly objects can often be inferred (by human & maybe machine)
• Lack of authority to assert can make reliable, verifiable, and true statements difficult or impossible
  • We want to assert the predicate / attribute of that PID, e.g. IsVersionOf, HasPart, etc.
  • We want relational understanding

Analogy - cataloguing a book and discovering that half the title page is missing...

...

All of which leads to a lack of authority to assert:

• Undermining the multi-part scholarly object
• Undermining end-user discovery and establishment of provenance
• Undermining open research, esp. verification, reproducibility, etc.
• Undermining ability of repositories or other scholarly services to contribute meaningful data to the wider scholarly PID graph

Er, what the heck is 'PID literacy'?!

What does a PID literate scholar look like?

"PID literate scholar might display the following competencies:

1. An understanding of persistent identification in scholarship, when it should be used, and its importance to the scholarly record and the wider PID graph.
2. An ability to accurately identify, reproduce, and cite PIDs in scholarship activities.
3. Cognizance of adjacent PID types relevant to scholars’ community of practice, such as those devised to identify scholarly 'things' other than academic papers." [1]

Where art thou, PID literacy?

• Challenges within the open research culture space generally
  • e.g. open data, data management planning (DMP), etc.
• Our research reveals [1] many scholars (~35%) demonstrate limited recognition of even common PIDs, e.g. DOIs & ORCIDs
  • Even less understanding of how they should be (re)used - widespread confusion
  • Concerning levels of failure in the PID tests we engaged our participant scholars in

...

But enthusiasm for PIDs in scholarship; they are perceived positively even though many do not know what they are (exactly) or how to use them

Centrality of scholars to making a better scholarly web

HELP NOW PLEASE!

• To go beyond the academic paper we need help from PID literate scholars
• Radical improvement of 'digital scholarship' skills necessary
• 'To do' for open research / scholarly communications teams: advocate; communicate; educate

Better metadata modelling of the new nature of scholarly research objects increasingly reflected in schema

• Rioxx v 3.0 and DataCite Schema

Conclusion and way forward...

• Distributed, multi-part scholarly objects are here -- and will only increase
• Modelling reality and servicing users requires that we describe this complexity in a meaningful way, contributing to the wider scholarly graph of data
• But improving PID literacy as part of wider 'digital scholarship' engagement essential to guarantee sufficient authority to assert, as well as open research goals
• Notwithstanding PID illiteracy...

Questions?

References

[1] G. Macgregor, B. S. Lancho-Barrantes, and D. R. Pennington, 'Measuring the concept of PID literacy: user perceptions and understanding of PIDs in support of open scholarly
infrastructure', Open Information Science, vol. 7, no. 1, 2023, available: https://doi.org/10.1515/opis-2022-0142

[2] G. Alemu. (2022). 'Resource Description and Access (RDA)'. In The Future of Enriched, Linked, Open and Filtered Metadata: Making Sense of IFLA LRM, RDA, Linked Data and BIBFRAME (pp. 197-212). Facet, London, available https://doi.org/10.29085/9781783304943.010

[3] G. Bilder, J. Lin, and C. Neylon, 'The Principles of Open Scholarly Infrastructure', 2020, available: https://doi.org/10.24343/C34W2H

[4] H. Cousijn et al., 'Connected Research: The Potential of the PID Graph', Patterns, vol. 2, no. 1, p. 100180, 2021, available: https://doi.org/10.1016/j.patter.2020.100180

[5] H. M. Sandy et al., 'Making a case for open research: implications for reproducibility and transparency', Proceedings of the Association of Information Science & Technology, vol. 54, no. 1, pp. 583–586, 2017, available: https://doi.org/10.1002/pra2.2017.14505401079

[6] S. M. Jones et al., 'Scholarly Context Adrift: Three out of Four URI References Lead to Changed Content', PLoS ONE vol. 11, no. 1, p. e0167475, 2021, available: https://doi.org/10.1371/journal.pone.0167475

[7] S. M. Jones, M. Klein, and H. Van de Sompel, 'Robustifying Links To Combat Reference Rot', Code4Lib Journal, no. 50, 2021, available: https://journal.code4lib.org/articles/15509