F.1:11 - SoTA-Echoing
| Practice question | Best-known line | Serious alternative or default | Defect overcome and F.1 mutation | Source roles and limits | Reopen condition |
|---|---|---|---|---|---|
| How should a small source cut remain question-led while exposing decisive heterogeneous rivals and limits? | Critical interpretive synthesis is the best-known-line candidate for this bounded comparison because it begins from a compass question, uses purposive iterative selection, critiques assumptions, and can synthesize heterogeneous evidence. Perlman, Ben-Sheleg, and Ellen’s 2026 scoping review makes the method’s recurring phases, variation, and gaps inspectable. | Protocol-first systematic-review reporting and one-canon selection are the serious defaults. PRISMA and PRISMA-S are retained only as an official/popular comparator when a protocol-defined review is actually required. | Fixed eligibility and reporting can preserve search history without deciding which conceptual claim changes the answer; one-canon selection hides rivals and transfer failure. Adapt: F.1 uses a stated question, inspected roles, iterative reopen, explicit gaps, and a separate F.0.2 synthesis boundary. Reject: a saturation claim, health-domain appraisal rules, and an exhaustive protocol for every bounded question. | Perlman, Ben-Sheleg, and Ellen, Making sense of conducting a critical interpretive synthesis: A scoping review (2026), is the current critical synthesis; PRISMA/PRISMA-S are the explicit protocol comparator; purposive-sampling studies support the variation problem but set no FPF quota. | Reopen if a stronger current source-selection synthesis offers a cheaper method that exposes the same action-changing rivals, counterexamples, limits, and gaps. |
| How may search automation reduce effort without deciding the source cut? | Transparent active-learning screening is the best-known-line candidate for prioritizing inspection in a large candidate set when its corpus, model, and stopping interpretation remain recoverable. | Ranking, similarity, citation count, or an LLM answer used directly as the membership gate is the serious default. | The default converts an attention proxy into relevance, truth, or completeness. Adapt: F.1:4.4, invariants, self-checks, and regression checks let search readings order inspection but require the underlying claim before membership changes. | van de Schoot et al., An open source machine learning framework for efficient and transparent systematic reviews (2021), supplies an efficient transparent-screening branch, not evidence that a retained source is relevant or sufficient. Search-tool documentation and latest-model status are identity/currentness only. | Reopen if a current method demonstrates lower inspection effort while preserving claim-level membership decisions and explicit omission risk. |
| How should finding adapt to a revised question and stop with enough material for its next use? | Adopt exploratory, source-led search and adapt it with an explicit comparison of the material still needed by the receiving use. In :5.10 this directs the next search to the missing preparation condition and then permits a sufficient stop. | Developed berrypicking already changes the information need, follows references and changes techniques. A sufficient direct lookup remains preferable to further exploration. | Changing terms or documents can leave a necessary contribution unrepresented. Adopt: evolving inquiry and varied routes. Construct: the connection from their results to :4.4.1’s revised question, :4.4.4’s remaining contribution and :4.4.5’s bounded return to selection or subject work. The deliberate cost is this local comparison; no extra search is required when the material is sufficient. | Bates, The Design of Browsing and Berrypicking Techniques (1989), §§II–IV, is the developed historical comparator. BRIGHT-PRO (2026), §6.3, supplies contemporary failure evidence: repeated evidence, neglect of a required aspect and continued search after sufficient retrieval in its evaluated agent setting. Neither source establishes an FPF retrieval advantage or a universal stopping rule. | Reopen when the contribution comparison misses a needed aspect, imposes effort without changing the return, or a stronger method supplies the same bounded result more affordably. |
Choose the cheapest attainable route that supplies the needed material and its consequential limits. Source selection still leaves synthesis and assurance to their own methods.