{"@context":"https://schema.org","@type":"DigitalDocument","altText":"Comment attachment by janedoe1","alternateName":"WIJDR_Published_Comments_Deep_Dive_Oct_2026.pdf","caption":"","contentUrl":"https://lucycraig.blog/media/689/raw","datePublished":"2026-10-30T13:28:20+00:00","description":"","encodingFormat":"application/pdf","identifier":"689","isAccessibleForFree":true,"machineReadable":{"json":"https://lucycraig.blog/media/689/index.json","markdown":"https://lucycraig.blog/media/689/index.md"},"name":"Comment attachment by janedoe1","ocrStatus":"ready","ocrText":"--- Page 1 ---\nWIJDR | Published comments research | Oct 2026\n1\nWIJDR | PUBLISHED COMMENT ANALYSIS\nDeep-dive research report | October 6\u201330, 2026 (UTC)\nPrepared from a published-only XML export generated October 30, 2026 at 13:03 UTC. This report analyzes comment\ncontent, timestamps, public reply structure, and reaction totals. No commenter identifying attributes, network addresses,\naccount identifiers, or unpublished entries are used.\nExecutive findings\n3,286 published comments across 174 page keys. 2,274 (69.2%) are replies and 1,012 (30.8%) are top-level comments.\nThe archive is unusually concentrated: 2,717 (82.7%) appeared October 8\u201317. The top day, October 10, produced 656\n(20.0%) comments. October 9\u201314 together account for 2,058 (62.6%).\nBoth pro-Jane Doe and anti-Jane Doe arguments are present, and the loudest conflicts are interactive. Supportive\narguments emphasize intoxication, consent, alleged group coordination, institutional failures, and victim-blaming. Critical\narguments question credibility, subsequent conduct, motivations, evidence, and procedural fairness. Opposing commenters\nrepeatedly criticize the site itself. These are positions expressed by commenters, not adjudicated facts.\nImportant limitation: the automatic stance scan identified 169 strong anti-position signals, 51 strong pro-position signals, 33\nmixed signals, and 3,033 not confidently classified. Those numbers are not a valid pro/anti percentage of the conversation.\nMuch of the debate uses sarcasm, replies, quotations, indirect disagreement, or short language. Reporting 169:51 as\naudience preference would materially misrepresent the record.\nVolume over time\n10-06\n10-10\n10-13\n10-17\n10-22\n10-29\nDaily published comments (UTC); peak = 656 on Oct 10\nDate (UTC)\nComments\nReplies\nInterpretation\nOct 9\n327\n198\nRapid buildup and active exchanges\nOct 10\n656\n421\nPrimary spike; broad sitewide activity\nOct 11\n304\n223\nContinuation of major arguments\nOct 12\n283\n221\nDebate sustained mainly through replies\nOct 13\n287\n202\nInstitutional actions and contested evidence\nOct 17\n235\n202\nSecond, strongly reply-driven episode\n\n--- Page 2 ---\nWIJDR | Published comments research | Oct 2026\n2\nDate (UTC)\nComments\nReplies\nInterpretation\nOct 29\n44\n27\nDocumentary-linked discussion, not mass-volume resurgence\nHow sentiment and stance were handled\nStance means a substantive position toward Jane Doe and/or her allegations; it is distinct from positive or negative emotional\ntone. A hostile comment aimed at the website may support Jane Doe. A legal criticism of the court may be neutral about the\nsexual-assault allegations. Comments that condemn harassment but question evidence can be mixed. The automated scan\nused explicit phrase patterns only, deliberately abstaining from most classifications rather than guessing. No claim of hand\nvalidation or exhaustive machine-learning classification is made.\nConservative high-signal counts: 169 anti (5.1% of all comments), 51 pro (1.6%), 33 ambiguous/mixed (1.0%), 3,033\nunclassified (92.3%). These are lower-bound counts of selected explicit expressions, not prevalence estimates. The anti\nphrases cover more common short insults than the pro phrases, producing an asymmetric detector; comparing the two counts\nas an unbiased ratio is inappropriate.\n\n--- Page 3 ---\nWIJDR | Published comments research | Oct 2026\n3\nArgument map: supportive and critical positions\nPro-Jane Doe / critical of anti-Jane-Doe narrative\nAnti-Jane Doe / skeptical of allegations\nIntoxication and capacity: accepting some activity does not\nnecessarily establish consent to other activity or continued\nparticipation.\nConsent and subsequent conduct: some commenters argue earlier\nexpressions of agreement or later communications contradict the\nallegations.\nAlleged coordination: several commenters emphasize claims of\ngroup messaging, others joining, and unequal knowledge.\nCredibility and motive: allegations characterized by some as regret,\nreputational concern, money-seeking, or changing accounts.\nInstitutional response: criticism of Cornell, fraternities, campus\nenforcement and the handling of sexual-assault reports.\nDue process: objections to reputational consequences, the\npresumption of guilt, sealing practices, and media narratives.\nPrivacy and abuse: complaints that identification efforts,\nsexualized ridicule, and the website itself amplify harassment.\nSpeech and publication: arguments that public records, court\ntransparency, and criticism remain legitimate topics.\nInterpretive caution: consent is conduct-specific and context-dependent. A person can consent to one act and not another;\nsubsequent communications do not, by themselves, settle what happened. Likewise, a lawsuit is not proof of the pleaded\nallegations, and the absence of charges is not proof the allegation was false. The comments repeatedly collapse these\ndistinctions.\nTopic fingerprint: conservative keyword matches\nTheme\nComments with match\nShare of corpus\nconsent/intoxication\n578\n17.6%\ncredibility/evidence\n437\n13.3%\ncourts/due process\n346\n10.5%\nmedia/censorship\n400\n12.2%\ncornell/fraternities\n266\n8.1%\nclass/wealth\n169\n5.1%\nprivacy/identity/doxxing\n143\n4.4%\nprotest/institutional\n90\n2.7%\ninsult/sexual shaming\n441\n13.4%\ndocumentary\n21\n0.6%\nTheme categories overlap; a single comment can match several. These are reproducible word/phrase matches, not\ncomplete semantic counts. In particular, some references to \u201crape\u201d are rebuttals or quotations rather than endorsements of\nan assertion. The 441 insult/sexual-shaming matches indicate a material amount of aggressive rhetoric; they do not mean\neach matched comment attacks the same side.\nConversation mechanics\nReplies dominate overall (69.2%) and particularly the Oct 17 wave (202 of 235, or 86.0%). That indicates a few contested\nconversations can generate many posts without an equally large increase in new opening statements. Oct 10 was both the\n\n--- Page 4 ---\nWIJDR | Published comments research | Oct 2026\n4\nbiggest absolute reply day (421) and a significant top-level-comment day (235).\nLargest reply trees by total comments including the initial comment: 59 (Oct 11; defense of Jane Doe and criticism of\nmisogynistic attacks), 45 (Oct 13; off-topic university-access argument), 33 (Oct 8; sexually explicit harassment), 29 (Oct 9;\ndispute over ability to overpower the accused), 25 (Oct 10; criticism of the blog and reference to the claimant\u2019s legal team),\nand 22 (Oct 17; dispute over identity claims). These thread summaries are anonymized and do not validate any participant\u2019s\nclaimed firsthand knowledge.\nThe second-largest tree is largely off-topic, demonstrating why \u201cmost-commented\u201d does not equal \u201cmost relevant.\u201d The top\nsupportive tree and the prominent blog-critical tree demonstrate that pro-Jane-Doe voices were not absent from the published\narchive.\n\n--- Page 5 ---\nWIJDR | Published comments research | Oct 2026\n5\nEvents and temporal alignment\nDate\nVerified external or site event\nCommentary relationship / confidence\nOct 5\u20136\nCornell campus protests and Day Hall property\ndamage reported by AP/Washington Post.\nThe export starts Oct 6 with just 3 comments. An\nopening-stage news climate is established, but no\nwithin-export before/after baseline exists.\nOct 8\u201310\nOct 8 reporting on a party safety monitor; Oct 9\njudicial sealing order following misidentification and\nthreats.\nComments jump 164 \u2192 327 \u2192 656. Legal secrecy, free\nspeech, privacy, identity, consent and credibility appear.\nStrong temporal coincidence; causal attribution unproven.\nOct 12\u201314\nOct 12 WIJDR article challenging the Oct 9 order; Oct\n13 Cornell fraternity-party suspension; Oct 14\ninstitutional leadership debate.\n283 \u2192 287 \u2192 201. Discussion includes court process and\nCornell institutional issues. No clean single-event attribution.\nOct 16\u201317\nTrustees met; protesters demonstrated; New York\nTimes report Oct 17.\n115 \u2192 235. Of Oct 17 comments, 202 are replies; this points\nto renewed disputes as well as news attention.\nOct 22\u201323\nWIJDR redaction article on Oct 22; judge grants Jane\nDoe pseudonymity Oct 23.\n45 \u2192 20. No large fresh volume spike; important events are\nnot necessarily reflected in sitewide posting volume.\nOct 28\u201330\nNetflix announced the Instadocs installment Oct 28,\nreleased Oct 29; WIJDR published a same-day\narticle.\n31 \u2192 44 \u2192 22 (Oct 30 partial). Clear topical linkage within\nseveral Oct 29 posts, modest total-volume bump.\nTiming is expressed in UTC, following the XML timestamps. A comment near midnight UTC may fall on the preceding local\ncalendar date in New York or Bogot\u00e1. The October 30 data end at the export timestamp, so that day is incomplete. External\narticles were independently checked; their publication dates do not establish when each commenter first encountered them.\nDistinguishing correlation from causation\nFor Oct 10, the document shows an unusually broad increase on the home page (205 comments), discussion landing page\n(110), an article page (40), and multiple individual discussion topics. The evidence fits wider awareness of the website, highly\nvisible controversy and self-reinforcing argument chains. It does not isolate whether a specific article, repost, search-ranking\nchange, or judicial action drove the traffic. No page-view/referral history, first-publication timestamps for every post, or\nplatform reach series were supplied.\nFor Oct 17, the high reply share and large existing-topic contributions suggest conversational reactivation, possibly alongside\nexternal reporting. For Oct 29, topical mentions of Netflix show a more direct content link even though aggregate comments\nremained modest.\n\n--- Page 6 ---\nWIJDR | Published comments research | Oct 2026\n6\nReader reaction system: what it can and cannot prove\nReaction\nTotal recorded\nInterpretation\nThumbs up\n6,886\nPotential approval, not verified stance\nTrain / clown\n5,026\nSite-specific symbol; meaning may vary by context\nLaugh\n3,184\nCan be agreement or mockery\nMiddle finger\n2,588\nHostility to comment, speaker, or subject unclear\nGold star\n1,327\nPotential approval\nTrophy\n1,204\nPotential approval\nTotal recorded reactions: 20,215. This is a snapshot of reaction counters at export time, not reactions created on each\ncomment\u2019s posting date. They cannot be dated to news events or reliably reclassified as pro/anti sentiment. The\nhighest-reacted items include a site-feature announcement, a terse criticism of the site, support/hostility concerning\nsexual-assault claims, and profanity. Reaction popularity does not imply factual accuracy.\nNarrative evolution\nOpening phase (Oct 6\u20138): direct accusations about dishonesty and sexual conduct coexist with a lengthy pro-Jane-Doe\nargument about intoxication, alleged group coordination and delayed trauma recognition. Off-topic link drops are also evident\nearly in the dataset.\nMass-engagement phase (Oct 9\u201313): the debate widens from the specific incident to rape law, evidence, credibility,\nfeminism/misogyny, press freedom, anonymous speech, judicial sealing, political commentary and Cornell leadership.\nComments critiquing the site itself become an important counter-narrative. This phase produces most of the archive.\nRebuttal phase (Oct 14\u201317): argument depth exceeds opening-message growth. Fights over sources, identity assertions\nand the meaning of consent persist. Do not infer that a commenter claiming personal acquaintance has verified any identity.\nLower-volume phase (Oct 18\u201328): overall output contracts to tens of comments per day; legal/privacy and class concerns\nrecur, but there is no second October 10-scale rise.\nDocumentary phase (Oct 29\u201330 partial): discussion touches documentary production, Netflix coverage, alleged uneven\naccess or inclusion, and the blog\u2019s editorial response. The dated film release makes this link relatively direct.\nQuality, risks and methodological boundaries\nThis is a self-selected, contentious, site-specific audience\u2014not a representative poll of Cornell students, the public, or either\nparty\u2019s supporters. Raw entries are not unique persons; the analysis deliberately does not attempt identity matching,\ndeduplication by technical metadata, or coordinated-behavior attribution. The report excludes all nonpublished statuses\nentirely. Violent fantasies, sexually derogatory remarks and unsupported allegations appear in published comments; they are\nanalyzed as rhetoric, not repeated as factual claims.\nA reliable final population-wide pro/anti split would require blind, human-reviewed annotation of all 3,286 comments or a large\nstratified sample, with separate pro, anti, mixed, neutral, off-topic, unclear labels, paired agreement tests and context-aware\nhandling of sarcastic replies. The strong-signal scan included here is a transparent starting point, not that completed\nvalidation.\n\n--- Page 7 ---\nWIJDR | Published comments research | Oct 2026\n7\nActionable research conclusions\nThe site functions less like a passive comment wall than a live rebuttal arena. Consider reporting top-level comments\nversus replies, issue-specific threads, civil versus abusive content, and stance among confidently hand-coded\nsubstantive comments as separate measures. Preserve comment creation timestamps and publication dates of articles in\nany future exports. To test spike causes, compare on-site visits, referrals, social shares and post publication dates in the\nsame timezone.\n\n--- Page 8 ---\nWIJDR | Published comments research | Oct 2026\n8\nReproducibility and external source notes\nDataset: WIJDR-all-comments-20261030-130300.xml; root export generated 2026-10-30T13:03:00+00:00. Only comment\nelements with an explicit published status were included. All dates in the numeric analysis derive from each published\ncomment\u2019s created_at timestamp, not updated_at. Replies are entries with nonempty parent_id; root trees are reconstructed\nusing parent links among published items. If a parent is outside the published dataset, tree reconstruction cannot reconstruct\nthat unpublished context.\nSource comparison (independently verified URLs, accessed October 30, 2026):\nAP/Washington Post, campus protests, Oct 6\nhttps://www.washingtonpost.com/national/2026/10/05/cornell-university-rape-investigation/8b072562-c0dc-11f1-8170-681419af1cc9_story.html\nWashington Post, sealing order, Oct 9\nhttps://www.washingtonpost.com/education/2026/10/09/judge-seals-documents-cornell-rape-case-citing-doxing-unrelated-people/\nReuters, Cornell fraternity party ban, Oct 13\nhttps://www.reuters.com/world/us/cornell-bans-fraternity-parties-rest-of-academic-year-after-alleged-gang-rape-2026-10-13/\nNew York Times republished by GV Wire, trustees, Oct 17\nhttps://gvwire.com/2026/10/17/cornell-trustees-present-a-united-front-amid-crisis/\nCBS, pseudonymity ruling, Oct 23\nhttps://www.cbsnews.com/news/jane-doe-can-remain-anonymous-cornell-university-sexual-assault-lawsuit/\nNetflix Tudum, documentary, Oct 28/29\nhttps://www.netflix.com/tudum/articles/instadocs-jane-doe-v-cornell-release-date-news\nWIJDR, response to court order, Oct 12\nhttps://www.lucycraig.blog/article/jane-doe-meet-jane-doe-we-re-coming-to-court\nWIJDR, Netflix article, Oct 29\nhttps://lucycraig.blog/article/watch-netflix-releases-jane-doe-v-cornell-documentary-today\nSource hierarchy: counts and thread descriptions are from the supplied published XML; event dates are based on the listed journalistic,\norganizational or site publications. Self-published WIJDR assertions are treated as evidence of the site\u2019s own editorial activity, not independent\nproof of disputed facts.","text":"--- Page 1 ---\nWIJDR | Published comments research | Oct 2026\n1\nWIJDR | PUBLISHED COMMENT ANALYSIS\nDeep-dive research report | October 6\u201330, 2026 (UTC)\nPrepared from a published-only XML export generated October 30, 2026 at 13:03 UTC. This report analyzes comment\ncontent, timestamps, public reply structure, and reaction totals. No commenter identifying attributes, network addresses,\naccount identifiers, or unpublished entries are used.\nExecutive findings\n3,286 published comments across 174 page keys. 2,274 (69.2%) are replies and 1,012 (30.8%) are top-level comments.\nThe archive is unusually concentrated: 2,717 (82.7%) appeared October 8\u201317. The top day, October 10, produced 656\n(20.0%) comments. October 9\u201314 together account for 2,058 (62.6%).\nBoth pro-Jane Doe and anti-Jane Doe arguments are present, and the loudest conflicts are interactive. Supportive\narguments emphasize intoxication, consent, alleged group coordination, institutional failures, and victim-blaming. Critical\narguments question credibility, subsequent conduct, motivations, evidence, and procedural fairness. Opposing commenters\nrepeatedly criticize the site itself. These are positions expressed by commenters, not adjudicated facts.\nImportant limitation: the automatic stance scan identified 169 strong anti-position signals, 51 strong pro-position signals, 33\nmixed signals, and 3,033 not confidently classified. Those numbers are not a valid pro/anti percentage of the conversation.\nMuch of the debate uses sarcasm, replies, quotations, indirect disagreement, or short language. Reporting 169:51 as\naudience preference would materially misrepresent the record.\nVolume over time\n10-06\n10-10\n10-13\n10-17\n10-22\n10-29\nDaily published comments (UTC); peak = 656 on Oct 10\nDate (UTC)\nComments\nReplies\nInterpretation\nOct 9\n327\n198\nRapid buildup and active exchanges\nOct 10\n656\n421\nPrimary spike; broad sitewide activity\nOct 11\n304\n223\nContinuation of major arguments\nOct 12\n283\n221\nDebate sustained mainly through replies\nOct 13\n287\n202\nInstitutional actions and contested evidence\nOct 17\n235\n202\nSecond, strongly reply-driven episode\n\n--- Page 2 ---\nWIJDR | Published comments research | Oct 2026\n2\nDate (UTC)\nComments\nReplies\nInterpretation\nOct 29\n44\n27\nDocumentary-linked discussion, not mass-volume resurgence\nHow sentiment and stance were handled\nStance means a substantive position toward Jane Doe and/or her allegations; it is distinct from positive or negative emotional\ntone. A hostile comment aimed at the website may support Jane Doe. A legal criticism of the court may be neutral about the\nsexual-assault allegations. Comments that condemn harassment but question evidence can be mixed. The automated scan\nused explicit phrase patterns only, deliberately abstaining from most classifications rather than guessing. No claim of hand\nvalidation or exhaustive machine-learning classification is made.\nConservative high-signal counts: 169 anti (5.1% of all comments), 51 pro (1.6%), 33 ambiguous/mixed (1.0%), 3,033\nunclassified (92.3%). These are lower-bound counts of selected explicit expressions, not prevalence estimates. The anti\nphrases cover more common short insults than the pro phrases, producing an asymmetric detector; comparing the two counts\nas an unbiased ratio is inappropriate.\n\n--- Page 3 ---\nWIJDR | Published comments research | Oct 2026\n3\nArgument map: supportive and critical positions\nPro-Jane Doe / critical of anti-Jane-Doe narrative\nAnti-Jane Doe / skeptical of allegations\nIntoxication and capacity: accepting some activity does not\nnecessarily establish consent to other activity or continued\nparticipation.\nConsent and subsequent conduct: some commenters argue earlier\nexpressions of agreement or later communications contradict the\nallegations.\nAlleged coordination: several commenters emphasize claims of\ngroup messaging, others joining, and unequal knowledge.\nCredibility and motive: allegations characterized by some as regret,\nreputational concern, money-seeking, or changing accounts.\nInstitutional response: criticism of Cornell, fraternities, campus\nenforcement and the handling of sexual-assault reports.\nDue process: objections to reputational consequences, the\npresumption of guilt, sealing practices, and media narratives.\nPrivacy and abuse: complaints that identification efforts,\nsexualized ridicule, and the website itself amplify harassment.\nSpeech and publication: arguments that public records, court\ntransparency, and criticism remain legitimate topics.\nInterpretive caution: consent is conduct-specific and context-dependent. A person can consent to one act and not another;\nsubsequent communications do not, by themselves, settle what happened. Likewise, a lawsuit is not proof of the pleaded\nallegations, and the absence of charges is not proof the allegation was false. The comments repeatedly collapse these\ndistinctions.\nTopic fingerprint: conservative keyword matches\nTheme\nComments with match\nShare of corpus\nconsent/intoxication\n578\n17.6%\ncredibility/evidence\n437\n13.3%\ncourts/due process\n346\n10.5%\nmedia/censorship\n400\n12.2%\ncornell/fraternities\n266\n8.1%\nclass/wealth\n169\n5.1%\nprivacy/identity/doxxing\n143\n4.4%\nprotest/institutional\n90\n2.7%\ninsult/sexual shaming\n441\n13.4%\ndocumentary\n21\n0.6%\nTheme categories overlap; a single comment can match several. These are reproducible word/phrase matches, not\ncomplete semantic counts. In particular, some references to \u201crape\u201d are rebuttals or quotations rather than endorsements of\nan assertion. The 441 insult/sexual-shaming matches indicate a material amount of aggressive rhetoric; they do not mean\neach matched comment attacks the same side.\nConversation mechanics\nReplies dominate overall (69.2%) and particularly the Oct 17 wave (202 of 235, or 86.0%). That indicates a few contested\nconversations can generate many posts without an equally large increase in new opening statements. Oct 10 was both the\n\n--- Page 4 ---\nWIJDR | Published comments research | Oct 2026\n4\nbiggest absolute reply day (421) and a significant top-level-comment day (235).\nLargest reply trees by total comments including the initial comment: 59 (Oct 11; defense of Jane Doe and criticism of\nmisogynistic attacks), 45 (Oct 13; off-topic university-access argument), 33 (Oct 8; sexually explicit harassment), 29 (Oct 9;\ndispute over ability to overpower the accused), 25 (Oct 10; criticism of the blog and reference to the claimant\u2019s legal team),\nand 22 (Oct 17; dispute over identity claims). These thread summaries are anonymized and do not validate any participant\u2019s\nclaimed firsthand knowledge.\nThe second-largest tree is largely off-topic, demonstrating why \u201cmost-commented\u201d does not equal \u201cmost relevant.\u201d The top\nsupportive tree and the prominent blog-critical tree demonstrate that pro-Jane-Doe voices were not absent from the published\narchive.\n\n--- Page 5 ---\nWIJDR | Published comments research | Oct 2026\n5\nEvents and temporal alignment\nDate\nVerified external or site event\nCommentary relationship / confidence\nOct 5\u20136\nCornell campus protests and Day Hall property\ndamage reported by AP/Washington Post.\nThe export starts Oct 6 with just 3 comments. An\nopening-stage news climate is established, but no\nwithin-export before/after baseline exists.\nOct 8\u201310\nOct 8 reporting on a party safety monitor; Oct 9\njudicial sealing order following misidentification and\nthreats.\nComments jump 164 \u2192 327 \u2192 656. Legal secrecy, free\nspeech, privacy, identity, consent and credibility appear.\nStrong temporal coincidence; causal attribution unproven.\nOct 12\u201314\nOct 12 WIJDR article challenging the Oct 9 order; Oct\n13 Cornell fraternity-party suspension; Oct 14\ninstitutional leadership debate.\n283 \u2192 287 \u2192 201. Discussion includes court process and\nCornell institutional issues. No clean single-event attribution.\nOct 16\u201317\nTrustees met; protesters demonstrated; New York\nTimes report Oct 17.\n115 \u2192 235. Of Oct 17 comments, 202 are replies; this points\nto renewed disputes as well as news attention.\nOct 22\u201323\nWIJDR redaction article on Oct 22; judge grants Jane\nDoe pseudonymity Oct 23.\n45 \u2192 20. No large fresh volume spike; important events are\nnot necessarily reflected in sitewide posting volume.\nOct 28\u201330\nNetflix announced the Instadocs installment Oct 28,\nreleased Oct 29; WIJDR published a same-day\narticle.\n31 \u2192 44 \u2192 22 (Oct 30 partial). Clear topical linkage within\nseveral Oct 29 posts, modest total-volume bump.\nTiming is expressed in UTC, following the XML timestamps. A comment near midnight UTC may fall on the preceding local\ncalendar date in New York or Bogot\u00e1. The October 30 data end at the export timestamp, so that day is incomplete. External\narticles were independently checked; their publication dates do not establish when each commenter first encountered them.\nDistinguishing correlation from causation\nFor Oct 10, the document shows an unusually broad increase on the home page (205 comments), discussion landing page\n(110), an article page (40), and multiple individual discussion topics. The evidence fits wider awareness of the website, highly\nvisible controversy and self-reinforcing argument chains. It does not isolate whether a specific article, repost, search-ranking\nchange, or judicial action drove the traffic. No page-view/referral history, first-publication timestamps for every post, or\nplatform reach series were supplied.\nFor Oct 17, the high reply share and large existing-topic contributions suggest conversational reactivation, possibly alongside\nexternal reporting. For Oct 29, topical mentions of Netflix show a more direct content link even though aggregate comments\nremained modest.\n\n--- Page 6 ---\nWIJDR | Published comments research | Oct 2026\n6\nReader reaction system: what it can and cannot prove\nReaction\nTotal recorded\nInterpretation\nThumbs up\n6,886\nPotential approval, not verified stance\nTrain / clown\n5,026\nSite-specific symbol; meaning may vary by context\nLaugh\n3,184\nCan be agreement or mockery\nMiddle finger\n2,588\nHostility to comment, speaker, or subject unclear\nGold star\n1,327\nPotential approval\nTrophy\n1,204\nPotential approval\nTotal recorded reactions: 20,215. This is a snapshot of reaction counters at export time, not reactions created on each\ncomment\u2019s posting date. They cannot be dated to news events or reliably reclassified as pro/anti sentiment. The\nhighest-reacted items include a site-feature announcement, a terse criticism of the site, support/hostility concerning\nsexual-assault claims, and profanity. Reaction popularity does not imply factual accuracy.\nNarrative evolution\nOpening phase (Oct 6\u20138): direct accusations about dishonesty and sexual conduct coexist with a lengthy pro-Jane-Doe\nargument about intoxication, alleged group coordination and delayed trauma recognition. Off-topic link drops are also evident\nearly in the dataset.\nMass-engagement phase (Oct 9\u201313): the debate widens from the specific incident to rape law, evidence, credibility,\nfeminism/misogyny, press freedom, anonymous speech, judicial sealing, political commentary and Cornell leadership.\nComments critiquing the site itself become an important counter-narrative. This phase produces most of the archive.\nRebuttal phase (Oct 14\u201317): argument depth exceeds opening-message growth. Fights over sources, identity assertions\nand the meaning of consent persist. Do not infer that a commenter claiming personal acquaintance has verified any identity.\nLower-volume phase (Oct 18\u201328): overall output contracts to tens of comments per day; legal/privacy and class concerns\nrecur, but there is no second October 10-scale rise.\nDocumentary phase (Oct 29\u201330 partial): discussion touches documentary production, Netflix coverage, alleged uneven\naccess or inclusion, and the blog\u2019s editorial response. The dated film release makes this link relatively direct.\nQuality, risks and methodological boundaries\nThis is a self-selected, contentious, site-specific audience\u2014not a representative poll of Cornell students, the public, or either\nparty\u2019s supporters. Raw entries are not unique persons; the analysis deliberately does not attempt identity matching,\ndeduplication by technical metadata, or coordinated-behavior attribution. The report excludes all nonpublished statuses\nentirely. Violent fantasies, sexually derogatory remarks and unsupported allegations appear in published comments; they are\nanalyzed as rhetoric, not repeated as factual claims.\nA reliable final population-wide pro/anti split would require blind, human-reviewed annotation of all 3,286 comments or a large\nstratified sample, with separate pro, anti, mixed, neutral, off-topic, unclear labels, paired agreement tests and context-aware\nhandling of sarcastic replies. The strong-signal scan included here is a transparent starting point, not that completed\nvalidation.\n\n--- Page 7 ---\nWIJDR | Published comments research | Oct 2026\n7\nActionable research conclusions\nThe site functions less like a passive comment wall than a live rebuttal arena. Consider reporting top-level comments\nversus replies, issue-specific threads, civil versus abusive content, and stance among confidently hand-coded\nsubstantive comments as separate measures. Preserve comment creation timestamps and publication dates of articles in\nany future exports. To test spike causes, compare on-site visits, referrals, social shares and post publication dates in the\nsame timezone.\n\n--- Page 8 ---\nWIJDR | Published comments research | Oct 2026\n8\nReproducibility and external source notes\nDataset: WIJDR-all-comments-20261030-130300.xml; root export generated 2026-10-30T13:03:00+00:00. Only comment\nelements with an explicit published status were included. All dates in the numeric analysis derive from each published\ncomment\u2019s created_at timestamp, not updated_at. Replies are entries with nonempty parent_id; root trees are reconstructed\nusing parent links among published items. If a parent is outside the published dataset, tree reconstruction cannot reconstruct\nthat unpublished context.\nSource comparison (independently verified URLs, accessed October 30, 2026):\nAP/Washington Post, campus protests, Oct 6\nhttps://www.washingtonpost.com/national/2026/10/05/cornell-university-rape-investigation/8b072562-c0dc-11f1-8170-681419af1cc9_story.html\nWashington Post, sealing order, Oct 9\nhttps://www.washingtonpost.com/education/2026/10/09/judge-seals-documents-cornell-rape-case-citing-doxing-unrelated-people/\nReuters, Cornell fraternity party ban, Oct 13\nhttps://www.reuters.com/world/us/cornell-bans-fraternity-parties-rest-of-academic-year-after-alleged-gang-rape-2026-10-13/\nNew York Times republished by GV Wire, trustees, Oct 17\nhttps://gvwire.com/2026/10/17/cornell-trustees-present-a-united-front-amid-crisis/\nCBS, pseudonymity ruling, Oct 23\nhttps://www.cbsnews.com/news/jane-doe-can-remain-anonymous-cornell-university-sexual-assault-lawsuit/\nNetflix Tudum, documentary, Oct 28/29\nhttps://www.netflix.com/tudum/articles/instadocs-jane-doe-v-cornell-release-date-news\nWIJDR, response to court order, Oct 12\nhttps://www.lucycraig.blog/article/jane-doe-meet-jane-doe-we-re-coming-to-court\nWIJDR, Netflix article, Oct 29\nhttps://lucycraig.blog/article/watch-netflix-releases-jane-doe-v-cornell-documentary-today\nSource hierarchy: counts and thread descriptions are from the supplied published XML; event dates are based on the listed journalistic,\norganizational or site publications. Self-published WIJDR assertions are treated as evidence of the site\u2019s own editorial activity, not independent\nproof of disputed facts.","thumbnailUrl":"https://lucycraig.blog/media/689/thumb.webp","transcriptStatus":"not_applicable","transcriptText":"","url":"https://lucycraig.blog/media/689"}