| Next revision | Previous revision |
| provenance:practices:public_relations [2026/08/18 15:41] – New page: working log behind practices:public_relations — every query with its denominator, both report scripts with unedited output, fold rules with residue, spot-checked quotes, external sources verified and rejected, four reviewers' findings and two re karel.kubicek.claude | provenance:practices:public_relations [2026/09/17 08:13] (current) – Qualify root start and Contributing links; Authored by Claude karel.kubicek.claude |
|---|
| Not prose. Read it if you are checking a number. | Not prose. Read it if you are checking a number. |
| |
| <wrap todo>No ''<nowiki>~~DISCUSSION~~</nowiki>'' block, matching every existing ''provenance:'' page. Comments belong on the content page, [[Practices:Public relations]].</wrap> | <WRAP todo>No ''<nowiki>~~DISCUSSION~~</nowiki>'' block, matching every existing ''provenance:'' page. Comments belong on the content page, [[Practices:Public relations]].</WRAP> |
| |
| ===== The run ===== | ===== The run ===== |
| ^ Field ^ Value ^ | ^ Field ^ Value ^ |
| | Date | 2026-08-18 | | | Date | 2026-08-18 | |
| | Page status | **new page**. ''practices:public_relations'' has been a red link since the site outline was written: promised from [[start]] ("regarding how to present your results to public and journalists") and from [[Practices:Notifying websites]] ("the other post-publication channel"). ''sitemap.mjs'' on 2026-08-18 reported 57 existing pages and 26 promised-but-missing ids, this among them. **Not an extension and not an overlap** — no existing page covers press engagement, and the nearest neighbour, [[Practices:Notifying websites]], explicitly hands off to it. | | | Page status | **new page**. ''practices:public_relations'' has been a red link since the site outline was written: promised from [[:start]] ("regarding how to present your results to public and journalists") and from [[Practices:Notifying websites]] ("the other post-publication channel"). ''sitemap.mjs'' on 2026-08-18 reported 57 existing pages and 26 promised-but-missing ids, this among them. **Not an extension and not an overlap** — no existing page covers press engagement, and the nearest neighbour, [[Practices:Notifying websites]], explicitly hands off to it. | |
| | Corpus at the time | ''data/extract/run1/extractions.jsonl'', 5,859 papers, 7 venues (CCS, IMC, NDSS, PETS, USENIX Security, TheWebConf, IEEE S&P), 2010–2026. 5,869 ''paper.cols.txt'' files under ''data/fulltext''. | | | Corpus at the time | ''data/extract/run1/extractions.jsonl'', 5,859 papers, 7 venues (CCS, IMC, NDSS, PETS, USENIX Security, TheWebConf, IEEE S&P), 2010–2026. 5,869 ''paper.cols.txt'' files under ''data/fulltext''. | |
| | Scripts written | ''scripts/report_public_relations.mjs'', ''scripts/pr_fulltext_grep.sh'' | | | Scripts written | ''scripts/report_public_relations.mjs'', ''scripts/pr_fulltext_grep.sh'' | |
| ==== Full-text sweep ==== | ==== Full-text sweep ==== |
| |
| ''bash scripts/pr_fulltext_grep.sh'' — two single passes over 20 GB of ''paper.cols.txt''. ''--binary-files=text'' is mandatory: some ''.cols'' files are detected as binary by grep and are silently skipped otherwise, which under-counts the corpus (see [[Literature:Corpus]]). | ''bash scripts/pr_fulltext_grep.sh'' — two single passes over 20 GB of ''paper.cols.txt''. ''%%--binary-files%%=text'' is mandatory: some ''.cols'' files are detected as binary by grep and are silently skipped otherwise, which under-counts the corpus (see [[Literature:Corpus]]). |
| |
| Pass 1, 20 terms, papers matching any: **674 of 5,869 (11.5%)**. Per-term paper counts are in the script's committed output. The three largest are ''journalist'' (260), ''news outlet'' (81) and ''press release'' (76) — and reading them shows why the raw count is not the finding: ''press release'' is almost always a **cited source** (an FTC, ICANN or vendor press release in the references), and ''journalist'' is almost always a **study population** (journalists as at-risk users, e.g. ''2015/USENIX/investigating-the-computer-security-practices-and-needs-of-journalists''). | Pass 1, 20 terms, papers matching any: **674 of 5,869 (11.5%)**. Per-term paper counts are in the script's committed output. The three largest are ''journalist'' (260), ''news outlet'' (81) and ''press release'' (76) — and reading them shows why the raw count is not the finding: ''press release'' is almost always a **cited source** (an FTC, ICANN or vendor press release in the references), and ''journalist'' is almost always a **study population** (journalists as at-risk users, e.g. ''2015/USENIX/investigating-the-computer-security-practices-and-needs-of-journalists''). |
| | 2025–2026* | 2,217 | 44.2% | 21.9% | 1,732 | 51.3% | | | 2025–2026* | 2,217 | 44.2% | 21.9% | 1,732 | 51.3% | |
| |
| **Residue in both directions.** ''--examples'' samples every 500th match so the sample spans the corpus rather than its first page. Called **undenominated** (correctly, in every sampled case — these are rates and scores with no population attached): | **Residue in both directions.** ''%%--examples%%'' samples every 500th match so the sample spans the corpus rather than its first page. Called **undenominated** (correctly, in every sampled case — these are rates and scores with no population attached): |
| |
| CCS/2010/dialog-based-payload-aggregation-for-intrusion-detection | CCS/2010/dialog-based-payload-aggregation-for-intrusion-detection |
| ==== node scripts/report_public_relations.mjs --examples (residue only) ==== | ==== node scripts/report_public_relations.mjs --examples (residue only) ==== |
| |
| The run above is reproduced verbatim by ''--examples'' before this block; only the sampled residue is repeated here. | The run above is reproduced verbatim by ''%%--examples%%'' before this block; only the sampled residue is repeated here. |
| |
| <code> | <code> |
| | ACM / IEEE / USENIX / IAPP / FIRST author-facing media guidance | **Checked, none found.** FIRST's multi-party coordination guidelines v1.1 cover vendor embargoes, not press. Reported as a negative on the content page. | | | ACM / IEEE / USENIX / IAPP / FIRST author-facing media guidance | **Checked, none found.** FIRST's multi-party coordination guidelines v1.1 cover vendor embargoes, not press. Reported as a negative on the content page. | |
| |
| <wrap todo>Method note worth carrying to the next run: ''WebFetch'' returned **wrong article content** for several PMC and BMJ URLs during this run (unrelated crystallography and cardiology papers), and hit Cloudflare or 403 on bmj.com, acm.org, pubmed.ncbi.nlm.nih.gov and usenix.org. Everything load-bearing was therefore cross-checked through a second channel — NCBI E-utilities, the Crossref API, or Europe PMC full-text XML, via ''curl''. **Do not trust a single ''WebFetch'' response for a numbered claim.**</wrap> | <WRAP todo>Method note worth carrying to the next run: ''WebFetch'' returned **wrong article content** for several PMC and BMJ URLs during this run (unrelated crystallography and cardiology papers), and hit Cloudflare or 403 on bmj.com, acm.org, pubmed.ncbi.nlm.nih.gov and usenix.org. Everything load-bearing was therefore cross-checked through a second channel — NCBI E-utilities, the Crossref API, or Europe PMC full-text XML, via ''curl''. **Do not trust a single ''WebFetch'' response for a numbered claim.**</WRAP> |
| |
| ===== Reviewer findings ===== | ===== Reviewer findings ===== |
| | The lead box states the page's most contestable figure with none of its caveats, 50 lines before they arrive, and "about half of this field's own writing" overstates what a model-selected sentence can support. | **Accepted.** The box now carries the proxy caveat in its own second sentence, and the "field's own writing" phrasing is gone from both the box and the trend paragraph. | | | The lead box states the page's most contestable figure with none of its caveats, 50 lines before they arrive, and "about half of this field's own writing" overstates what a model-selected sentence can support. | **Accepted.** The box now carries the proxy caveat in its own second sentence, and the "field's own writing" phrasing is gone from both the box and the trend paragraph. | |
| | "The asymmetry is not because press engagement is rare; it is because it is not reported" asserts exactly what its own footnote concedes cannot be shown. | **Accepted.** Rewritten to say the corpus cannot separate the two readings, which is true and implies the same thing for the reader. | | | "The asymmetry is not because press engagement is rare; it is because it is not reported" asserts exactly what its own footnote concedes cannot be shown. | **Accepted.** Rewritten to say the corpus cannot separate the two readings, which is true and implies the same thing for the reader. | |
| | The three subjective-statement footnotes say "add your name here if you agree" but carry **no name**, where [[Contributing]] asks for one. | **Accepted, with a caveat rather than a signature.** The footnotes now say plainly that the page was drafted by Claude and that no named researcher has vouched for the judgement calls. A model cannot satisfy a convention whose point is that a person stands behind the claim; saying so is the honest substitute. | | | The three subjective-statement footnotes say "add your name here if you agree" but carry **no name**, where [[:Contributing]] asks for one. | **Accepted, with a caveat rather than a signature.** The footnotes now say plainly that the page was drafted by Claude and that no named researcher has vouched for the judgement calls. A model cannot satisfy a convention whose point is that a person stands behind the claim; saying so is the honest substitute. | |
| | [[Practices:Legal enforcement]] promises "[[Practices:Public relations]] — publication as the other enforcement route, per {[saemann2022_fines]}", and this page never mentioned it — while resting //Publicity is an instrument// on two anecdotes from {[kotzias2018_coming]}. | **Accepted.** {[saemann2022_fines]} is now the first item in that section, with the 111-of-295 figure and the tension against {[zimmeck2023_gpc]}'s withholding. | | | [[Practices:Legal enforcement]] promises "[[Practices:Public relations]] — publication as the other enforcement route, per {[saemann2022_fines]}", and this page never mentioned it — while resting //Publicity is an instrument// on two anecdotes from {[kotzias2018_coming]}. | **Accepted.** {[saemann2022_fines]} is now the first item in that section, with the 111-of-295 figure and the tension against {[zimmeck2023_gpc]}'s withholding. | |
| | The embargo footnote contradicted itself: a row "fetched from www2026.thewebconf.org on 2026-08-18" and, in the same footnote, that page returning 404 on the same day. | **Accepted;** the footnote now states once that the row rests on the current homepage because the call-for-papers pages are gone. | | | The embargo footnote contradicted itself: a row "fetched from www2026.thewebconf.org on 2026-08-18" and, in the same footnote, that page returning 404 on the same day. | **Accepted;** the footnote now states once that the row rests on the current homepage because the call-for-papers pages are gone. | |
| * **The true number of papers that document their own press handling.** Three is a floor from a keyword sweep that is now known to have under-recalled once. The only way to raise confidence is a different method — e.g. classifying every ''ethics'' and //Disclosure// section rather than grepping — which was not run. | * **The true number of papers that document their own press handling.** Three is a floor from a keyword sweep that is now known to have under-recalled once. The only way to raise confidence is a different method — e.g. classifying every ''ethics'' and //Disclosure// section rather than grepping — which was not run. |
| * **Why the denomination rate is falling.** The script now decomposes the trend against the obvious confounder — the rising share of model-performance figures — and the confounder does not account for it: performance figures rise from 12.9% to 21.9% of sentences, while denomination among **non**-performance sentences falls from 67.8% (n=866) to 51.3% (n=1,732). So the fall is not an artefact of subject matter alone. What it //is// remains unestablished; plausible candidates the corpus cannot separate are longer and more compressed result sentences, more tables carrying the denominator instead of the prose, and drift in what the extraction model selects as an evidence quote. The content page says only that the fall survives the decomposition. | * **Why the denomination rate is falling.** The script now decomposes the trend against the obvious confounder — the rising share of model-performance figures — and the confounder does not account for it: performance figures rise from 12.9% to 21.9% of sentences, while denomination among **non**-performance sentences falls from 67.8% (n=866) to 51.3% (n=1,732). So the fall is not an artefact of subject matter alone. What it //is// remains unestablished; plausible candidates the corpus cannot separate are longer and more compressed result sentences, more tables carrying the denominator instead of the prose, and drift in what the extraction model selects as an evidence quote. The content page says only that the fall survives the decomposition. |
| | |
| | ===== Publication ===== |
| | |
| | ^ Step ^ Detail ^ |
| | | Bibliography | ''literature:bibliography'' rev 1787051260 → 1787067695, +11 entries, 351 → 362. Saved with ''%%--if-rev%%'' so a concurrent edit would have refused rather than overwritten. Diff verified as a pure append before saving. | |
| | | Content page | ''practices:public_relations'', new, 55,966 bytes | |
| | | This page | ''provenance:practices:public_relations'', new | |
| | | Reachability | [[:start]] already linked ''[[Practices:Public relations]]''; no new link was needed. | |
| | |
| | <WRAP important> |
| | **The bibtex4dw plugin serves a cached bibliography, and a stale cache fails silently.** Immediately after publishing, the content page rendered 22 inline citekey markers and a reference list numbered to [15] — but the list contained **only the 4 keys that already existed** before the bibliography edit. No purple warning, no missing-citation marker. Requesting ''?purge=true'' on ''literature:bibliography'' and then on the citing page fixed it; all 15 then rendered. |
| | |
| | So the standard check — "the rendered page shows a ''bibtex_references'' list" — **passes on a stale cache**. Count the entries in the rendered list against the number of distinct ''<nowiki>{[key]}</nowiki>'' markers on the page, and purge both pages after any bibliography edit. |
| | </WRAP> |
| |
| Back to the content page: **[[Practices:Public relations]]**. Corpus-wide caveats: [[Literature:Corpus]]. | Back to the content page: **[[Practices:Public relations]]**. Corpus-wide caveats: [[Literature:Corpus]]. |
| ====== References ====== | ====== References ====== |
| <bibtex bibliography></bibtex> | <bibtex bibliography></bibtex> |
| | |
| | ===== Markup sweep, 2026-09-17 ===== |
| | |
| | Mechanical rendering repair only: a fresh live raw/XHTML export of 188 pages was checked with ''check_wrap.mjs'' and ''check_typography.mjs''. Affected plugin tags, CLI flags and heading markup were repaired; no figures or substantive prose were changed. The resulting source and rendered DOM were re-checked after saving. |
| | |
| | ===== Link-hygiene sweep, 2026-09-17 ===== |
| | |
| | The fresh rendered-DOM sweep found bare ''start'' and ''Contributing'' links resolving inside the ''provenance:practices:'' namespace. They are now root-anchored; the post-save DOM was re-checked for red links. No figures or citations changed. |
| |
