Protocol: how the pool of sources was built

This file records every search, count and rule of the progressive-disclosure review. Shared rules: evidence README. Searches ran on 2026-10-29. Its terms are defined in frame.

1. Why grey literature

Garousi 2019, Table 4, answered for progressive-disclosure: six of seven questions get a yes.

# Question Answer for this review
1 Complex, not solvable from formal literature alone? yes: accordions, details, outliners and folding come from practice
2 Lack of volume or quality of evidence? yes: one review (Cockburn 2009) finds no clear guidelines; few studies use documents
3 Context important? yes: mobile against desktop, reference against learning, human against model reader
4 Corroborate research with practice? yes
5 Challenge practice with research, or the reverse? yes: practice nests accordions and trees of summaries; studies question both (position.md)
6 Would a synthesis serve both communities? yes
7 Large practitioner volume? no for documents as such; yes for outliners (state-of-the-art.md counts)

2. Questions

The sub-questions Q1 to Q7 of frame.

3. Existing reviews (Garousi Guideline 2)

  • Searched first, by strings R1 to R4 (section 4) and by the lead named in the brief (Cockburn 2009).
  • Found: 7. Read in full or in the relevant sections: Cockburn 2009, Brusilovsky 2001. By abstract: Chen 1996, Dillon 1998. By title only: Knutov 2009, Lam 2011. Held by a linked review: DeStefano 2007.
  • No systematic review or MLR of folded, expandable or nested text in documents was found; that is a statement about this search.

6. Stopping rule

  • Stated before searching: effort bounded, N = 15 per formal string and 10 per web string, with saturation, k = 5 consecutive results that add no new concept.
  • Formal: evidence exhaustion for F1, F18, F19, F21, F22, F24 and A1 to A4 (small sets); effort bounded for the rest.
  • Grey web: G1 and G2 stopped by saturation after the first page.
  • Q/A sites: as in section 5.

7. Selection

  • Include: a source on showing nested text to a reader (folded, expandable, zoomable, fisheye, layered, outlined).
  • Include also: hierarchical against networked structure for reading, or trees of summaries read by a model; English; readable on the search date.
  • Exclude: graphics and maps with no text or hierarchy; distortion techniques for images; tree visualisation for phylogeny or data; adaptive systems that adapt only links; vendor pages selling a product; duplicates.
  • A source whose text and abstract could not be read is kept only as "located", never as support.

8. Quality of grey sources (Garousi Guideline 11, Table 7)

  • The 20 criteria and the threshold: evidence README.
  • Outlet tier: standards, government guidance = tier 1 (1); company pages, project documentation, wiki and Q/A pages = tier 2 (0.5); blogs = tier 3 (0).
  • Threshold: 10 of 20. Below it a source is an example of practice only.
  • A count read from a public listing (stars, downloads) is the reviewer's measurement, not a source, and is not scored.
  • Scores, with subtotals, are in sources.

9. Extraction and synthesis

  • Extracted per source: kind, year, the claim, the evidence, any number with where it was read, marked "(secondary)" when read in an abstract or another work's summary.
  • Synthesis is narrative, by question Q1 to Q7, and every claim traces to a row of sources.

10. Limits

  • One first rater for progressive-disclosure; the second rater's blind sample is reported in the evidence README.
  • The ACM Digital Library and ScienceDirect refused automated reads on the day. ACM papers were read in author or course copies where found, otherwise through their abstract or through Cockburn 2009, and marked so.
  • Shared limits: evidence README. Here: one arXiv call timed out (F2).
  • Forward snowballing was not paged in a citation index.