Add a memo to a ResearcherReflectionLog

R is pass-by-value: callers must capture the return: log <- add_memo(log, "..."). The function appends the memo to log$memos (preserving order) and updates log$last_updated. When an audit_log is supplied, a memo_added decision is recorded so the methodology paper has a timestamp trail of when each memo was authored relative to AI calls.

Usage

add_memo(log, body = NULL, ..., memo = NULL, audit_log = NULL)

Arguments

log

A ResearcherReflectionLog.

body

Character: the memo's Markdown body (or pass a pre-built Memo object via memo instead).

...

Forwarded to make_memo (type, author, linked_codes, linked_themes, linked_entries, linked_prior_memo, timestamp, id). Ignored when memo is supplied.

memo

Optional pre-built Memo object; supplying this bypasses make_memo construction. Mutually exclusive with body.

audit_log

Optional AuditLog; when supplied, a memo_added decision is recorded.

Value

The updated ResearcherReflectionLog.

Details

Memos are immutable once added: there is no update_memo or delete_memo – if a researcher needs to revise a thought, they add a NEW memo with linked_prior_memo pointing at the old one. The chain is the audit trail. This is intentional per the Birks/ Chapman/Francis 2025 "Memoing in qualitative research: two decades on" guidance that memo evolution itself is data.

See also

make_memo (constructor); persist_memos (write all memos to disk as Markdown with YAML frontmatter); load_memos (read them back).

Examples

log <- create_reflection_log()

# Add a theoretical memo linked to a theme
log <- add_memo(
  log,
  body = paste0(
    "Adoption themes are over-weighted by contributor X's posts.\n\n",
    "Need to interrogate this concentration before publishing."
  ),
  type = "theoretical",
  linked_themes = "Adoption"
)

# Add an operational memo as a revision of the prior
log <- add_memo(
  log,
  body = "Merged codes async_routine + daily_batching into async_adoption.",
  type = "operational",
  linked_codes = c("async_routine", "daily_batching"),
  linked_prior_memo = log$memos[[1]]$id
)

list_memos(log)
#> # A tibble: 2 × 9
#>   id      timestamp author type  n_linked_codes n_linked_themes n_linked_entries
#>   <chr>   <chr>     <chr>  <chr>          <int>           <int>            <int>
#> 1 memo_2… 2026-08-… resea… theo…              0               1                0
#> 2 memo_2… 2026-08-… resea… oper…              2               0                0
#> # ℹ 2 more variables: body_chars <int>, body_preview <chr>