Create a provenance-aware Record Set data frame

Construct a recordset_df, a provenance-aware contextual dataset representing members of a Record Set.

Record Set is a contextual aggregation concept defined by the International Council on Archives (ICA) Records in Contexts standard (RiC). In operational terms, a Record Set may represent:

• a project workspace;

• a research corpus;

• a synchronized working environment;

• a digital collection;

• a reconstruction context;

• or another contextual grouping of related digital resources.

The Records in Contexts (RiC) standard provides a flexible and provenance-aware approach for describing evolving digital records, their relationships, and their contextual environments.

Unlike rigid hierarchical archival models, RiC allows records and digital resources to participate in multiple overlapping contextual groupings while preserving provenance and contextual relationships.

More information:

• ICA Records in Contexts overview: https://www.ica.org/ica-network/expert-groups/egad/records-in-contexts-ric/

• RiC-O ontology repository: https://github.com/ica-egad/ric-o

A recordset_df extends the dataset_df class with lightweight contextual Record Set semantics suitable for:

• filesystem observations;

• synchronized cloud folders;

• web archive members;

• digital surrogate collections;

• curation batches;

• Digital Twin workspaces;

• provenance-aware research collections;

• contextual digital preservation workflows.

The class is designed to work together with:

• read_snapshot()

• snapshot_to_reconstruction_context()

• snapshot_to_recordset_df()

while preserving the distinction between:

• observed filesystem evidence;

• contextual grouping of related resources;

• later analytical interpretation;

• and archival or semantic enrichment workflows.

In operational terms:

• record_set_id identifies a contextual grouping of related digital resources (similar to a project workspace, collection, or reconstruction environment);

• member_id identifies one observed or asserted member within that grouping.

The resulting object inherits from:

• recordset_df

• dataset_df

• tbl_df

• tbl

• data.frame

Usage

recordset_df(
  ...,
  identifier = c(member = "http://example.com/recordset#member"),
  var_labels = NULL,
  units = NULL,
  concepts = NULL,
  dataset_bibentry = NULL,
  dataset_subject = NULL
)

Arguments

...

Vectors (columns) to include in the record set.

identifier

A named vector of URI prefixes used to generate row identifiers.

Defaults to:

c(member = "http://example.com/recordset#member")

var_labels

Optional named list of human-readable variable labels.

units

Optional named list of measurement units.

concepts

Optional named list of semantic concept URIs.

dataset_bibentry

Optional bibliographic metadata created with dataset::dublincore() or dataset::datacite().

dataset_subject

Optional dataset subject metadata.

Value

A recordset_df object.

Details

The constructor requires at minimum the columns:

• record_set_id

• member_id

Validation and class assignment are delegated to new_recordset_df.

The constructor is intentionally lightweight and does not:

• infer authoritative archival hierarchy;

• reconcile duplicate identities;

• infer canonical resources;

• construct ontology-complete provenance graphs;

• or replace curatorial or archival interpretation.

Instead, it provides a stable contextual preservation layer for provenance-aware reconstruction and human-in-the-loop workflows.

Examples

toy_recordset <- recordset_df(
  record_set_id = c(
    "heritage_digitisation",
    "heritage_digitisation",
    "heritage_digitisation"
  ),
  member_id = c(
    "inst_001",
    "inst_002",
    "inst_003"
  ),
  member_path = c(
    "scans/photo_001.tif",
    "ocr/photo_001.txt",
    "reports/collection_summary.qmd"
  ),
  member_type = c(
    "file",
    "file",
    "file"
  ),
  source_type = c(
    "filesystem",
    "filesystem",
    "filesystem"
  ),
  identifier = c(
    member =
      "https://example.org/recordset/heritage#member"
  ),
  var_labels = list(
    record_set_id = "Record set identifier",
    member_id = "Member identifier",
    member_path = "Member path"
  ),
  concepts = list(
    record_set_id =
      "https://www.ica.org/standards/RiC/ontology#RecordSet",
    member_id =
      "https://www.ica.org/standards/RiC/ontology#Instantiation"
  ),
  dataset_bibentry = dataset::dublincore(
    title = "Toy Heritage Digitisation Record Set",
    creator = person("Jane", "Doe"),
    publisher = "fscontext"
  )
)

toy_recordset
#> Doe (2026): Toy Heritage Digitisation Record Set [dataset]
#>   rowid   record_set_id         member_id member_path    member_type source_type 
#>   <chr>   <chr>                 <chr>     <chr>          <chr>       <chr>      
#> 1 member1 heritage_digitisation inst_001  scans/photo_0… file        filesystem 
#> 2 member2 heritage_digitisation inst_002  ocr/photo_001… file        filesystem 
#> 3 member3 heritage_digitisation inst_003  reports/colle… file        filesystem