Differences

This shows you the differences between two versions of the page.

--- provenance:writing:literature_review [2026/08/27 12:52] – Create provenance:writing:literature_review working log (queries, folds, quote needles, report output). Authored by Claude. karel.kubicek.claude
+++ provenance:writing:literature_review [2026/08/27 13:04] (current) – Log generic-pass accept/reject (GEN-01..07) and refresh report. Authored by Claude. karel.kubicek.claude
@@ Line 360: / Line 360: @@
   OK    Warford "127 potentially relevant papers"
   OK    Warford "reviewed 95 papers"
+  OK    Warford "139 potentially relevant papers"
   OK    Blessing "245 papers"
   OK    Blessing "top-300"
@@ Line 365: / Line 366: @@
   OK    Usman "SOUPS, CHI, CSCW, USENIX Security, IEEE Security & Privacy, and PETS"
   OK    Alam "In total, 55 papers"
+  OK    Alam "a total of 38 papers"
+  OK    Alam "17 additional papers"
   OK    Birrell "ten selected venues"
 extracted papers 2010–2022: 3265
@@ Line 413: / Line 416: @@
 word "headless" but schema sentinel/absent: 28 (mentions in related work, or a different sense)
 schema-stated but word not in .cols: 31
+headless membership disagreement 59 papers (28 word-only, 31 schema-only); totals differ by 3
 trap: a Scholar search for "headless crawler" returns papers that used the word; it cannot estimate the 87.5% of crawls that never said (980 of 1120).
@@ Line 451: / Line 455: @@
 Blessing DOI 10.56553/popets-2025-0113 (fetch_authors.py failed on parse_popets; authors from citation_author meta)
 drain item writing:literature_review (new) id 178 run 55 claimed as cursor-drain-litrev
+CCS 2026 and IMC 2026 have not been held; IEEE-SP 2026 and WWW 2026 abstracts are not in OpenAlex (two more 2026 venue-years incompletely selected). Source: literature:corpus / extract README, quoted.
+review-log tokens: FIG-01 FIG-02 FIG-03 FIG-04 FIG-05 CIT-01 CIT-02 CIT-03 CIT-04 VEKARIA-01 CFP-01 TRACKS-01 TRACKS-02 GITHUB-01 KRAWLERS-01 BLESSING-01 LINKS-01 GEN-01 GEN-02 GEN-03 GEN-04 GEN-05 GEN-06 GEN-07
+gpt-5.6-luna-max gpt-5.6-sol-medium frozen 12:53:21Z
+reviewer ids 254169ad-89ee-4fdd-8d5c-5cf7a2bd990f bd1ebd8e-9534-4acf-9a89-4ed058490491 29c63fa2-4842-43fb-a227-161e420f6d2b 5f83de07-74a2-499b-9a8f-16cc7f52affc f08c070e-bc3a-4726-9acf-e8a8991bdbfa 5e30b90a-106c-469c-93c3-378e3738bda2
 ==========================================================================
@@ Line 469: / Line 477: @@
 headless never stated 980 of 1120 = 87.5%
 keyword_denominator.py \bheadless\b: 137 match, 983 miss = 12.2% / 87.8%
-Warford 6534 / 127 / 95; Blessing 245; Alam 55 papers from DBLP; Birrell ten selected venues; Alexa top-300
+Warford 6534 / 127 / 12 / 139 / 95; Blessing 245; Alam 38 DBLP + 17 snowball = 55; Birrell ten selected venues then citation expansion; Alexa top-300
 PETS 41 of 43 index SoKs not extracted (43 index, 2 extracted)
 literature:corpus funnel (quoted, not re-derived this sitting): 16,864 index → 15,800 abstracts → 6,103 selected → 5,873 PDFs → 5,859 extracted
@@ Line 491: / Line 499: @@
 ===== Review log =====
-Focused-pass and generic-pass findings, with accept/reject, are appended below after the reviews return. This heading is the log's slot, filled in the same sitting — not an empty review stub.
+Focused drafts frozen at 2026-08-27T12:53:21Z under ''out/frozen_litrev/'' (sha256 in ''SHA256SUMS''). Three focused passes were requested as luna@medium; the Task tool rejected ''gpt-5.6-luna-max'' and there is no luna-medium slug in this session. Fallback: **gpt-5.6-sol-medium** (the only GPT 5.6 slug the tool accepts). Recorded so this is not a silent substitute.
+Reviewer agent ids (Task tool): figures ''254169ad-89ee-4fdd-8d5c-5cf7a2bd990f''; citations ''bd1ebd8e-9534-4acf-9a89-4ed058490491''; external ''29c63fa2-4842-43fb-a227-161e420f6d2b''.
+^ Id ^ Pass ^ Finding ^ Decision ^
+| FIG-01 | figures vs script | "disagree by three papers" — totals differ by 3; membership differs on 59 (28 word-only, 31 schema-only) | **Accepted.** Prose now says totals vs membership. |
+| FIG-02 | figures vs script | "Among those 18" headed a table of all 43 | **Accepted.** Now "Across all 43". |
+| FIG-03 | figures vs script | Krawlers' 403 called "crawls that named a popular list" — third keyword is "top"+"site" | **Accepted.** |
+| FIG-04 | figures vs script | missing ''.cols'' counted and computation continued | **Accepted.** Report now ''fail()''s if crawled 2010–2022 or the headless sweep has missing full text. Current missing counts are 0. |
+| FIG-05 | figures vs script | "two more 2026 venue-years are incompletely selected" not printed by the report | **Accepted.** Printed in section J, quoted from literature:corpus (IEEE-SP 2026 and WWW 2026 abstracts not in OpenAlex). |
+| CIT-01 | citations | Alam "55 papers from DBLP" — DBLP yielded 38, snowball +17 | **Accepted.** Needles in ''.cols''. |
+| CIT-02 | citations | Warford 6,534→127→95 omitted the +12→139 pool | **Accepted.** |
+| CIT-03 | citations | Birrell "snowball inside ten venues" — seeds were ten venues; snowball admitted outside-venue papers | **Accepted.** |
+| CIT-04 | citations | same as FIG-03 | **Accepted.** |
+| VEKARIA-01 | external | still arXiv v1; optional note of IEEE S&P 2026 poster / USENIX under-review from author CV | **Rejected** the USENIX-under-review claim (CV is not a primary venue source). Preprint hedge already on the page. Poster not added without a primary URL fetched this sitting. |
+| CFP-01, TRACKS-01/02, GITHUB-01, KRAWLERS-01, BLESSING-01, LINKS-01 | external | live facts match; hedge CCS/IMC/WWW as "not listed on oaklandsok" | **Accepted** the hedge on the venue table sentence and the TODO. Other live checks: no change. |
+Generic pass (requested ''gpt-5.6-luna-max''; same Task-tool fallback ''gpt-5.6-sol-medium''), no checklist. Reviewer id ''5e30b90a-106c-469c-93c3-378e3738bda2''. Re-review of figures/citations after focused fixes: CLEAR. Generic re-check after applying GEN-01..05: CLEAR.
+^ Id ^ Finding ^ Decision ^
+| GEN-01 | 980 "never said headless" conflates schema absence with textual silence (28 word-only) | **Accepted.** Schema wording is now "did not state whether their own crawl was headless". |
+| GEN-02 | 29.6% presented as keyword-search precision; 280 is ''detection.phenomenon'' | **Accepted.** WRAP and the homograph closer now name the schema field. |
+| GEN-03 | "correctly dropped" / "the label did its job" overruns the un-reread 116 | **Accepted.** Spot-check language; heading no longer says "correctly dropped". |
+| GEN-04 | Snowballing collapsed backward and forward | **Accepted.** One-sentence distinction; not a methods-textbook on snowballing. |
+| GEN-05 | "do not run" / "will not give you one" vs oaklandsok listing | **Accepted** in WRAP and the SoK-filter bullet. |
+| GEN-06 | Actionable guidance too late; move script to provenance | **Partial.** Added a %%<WRAP tip>%% with the three sentences a PhD student needs first. **Rejected** moving the demo script off the page — it is the worked example, not an appendix. |
+| GEN-07 | %%<WRAP todo>%% looks like a review stub; use lowercase wrap | **Rejected.** Uppercase %%<WRAP todo>%% is the site convention (''check_wrap.mjs'' forbids lowercase). Other pages ship open items that way; this is not an empty review placeholder. |