{ "schema_version": 1, "stage": "visual gold authored before any new extractor output", "source": { "path": "ecofactor-google-document-15-nonconfidential-joint-appendix.pdf", "sha256": "b11e78413fee46e151eb441556afca793ad3378ea76debd7bc3811bc9f988d49", "physical_pages": 373, "provenance": "https://www.courtlistener.com/docket/68048163/15/ecofactor-inc-v-google-llc/" }, "selection": { "method": "fixed systematic positions prescribed before rendering or inspecting extraction output", "physical_pages": [ 1, 20, 39, 58, 77, 96, 115, 134, 153, 172, 191, 210, 229, 248, 267, 286, 305, 324, 343, 362, 373 ], "sample_pages": 21, "eligible_pages": 373, "page_exclusions": [], "claim_scope": "Sampled checks only; not a random-sample confidence estimate, not percent of whole corpus correct.", "phrase_rule": "First meaningful body passage and a passage near the vertical middle of meaningful body; on multicolumn pages use the left body column for both. On figures/tables/notices select visible text in the corresponding regions; record each substitution. Neither target is selected by extractor output." }, "matching_contract": { "primary": "Case-sensitive literal phrase after Unicode NFC and whitespace collapse. No spelling, number or punctuation repair. Explicit wrap-hyphen alternatives are listed per check; primary phrase may dehyphenate only that declared wrap.", "recall_denominator": "All positive phrase checks, including figure labels, notices and needs_OCR pages. Also report prose-only and needs_OCR strata without deleting any failures.", "negative_denominator": "Each declared forbidden interleaving; both boundary anchors must be located in order or result is unassessable, not pass.", "furniture_denominator": "One page-level contamination check per sampled page: any exact declared filing/footer string remaining in body fails. Metadata channel is allowed. This measures clean-body output, not lossless archival export.", "metadata_denominator": "Each named page label; distinguish body presence from structured metadata. No inference that a missing structured metadata field is absent from the PDF.", "ocr": "A body missing from the native layer must be explicitly flagged needs_OCR or unsupported by a deterministic extractor; missing text remains a phrase miss even if correctly flagged." }, "pages": [ { "physical_page": 1, "structure": "Title page with centred case caption and two separate counsel columns", "page_labels": [ "Page: 1" ], "furniture": [ "Case: 23-1101", "Document: 15", "Filed: 05/09/2023" ], "adjustments": [ "No running body prose; first target is court title, middle target is appeal caption. Counsel contact columns are not prose continuation." ], "phrases": [ { "id": "p001-a", "position": "first", "original_line_breaks": "United States Court of Appeals\nfor the Federal Circuit", "phrase": "United States Court of Appeals for the Federal Circuit" }, { "id": "p001-b", "position": "middle", "original_line_breaks": "On Appeal from the United States District Court Western District of Texas\nCase No. 6:20-cv-00075-ADA, Judge Alan D. Albright", "phrase": "On Appeal from the United States District Court Western District of Texas Case No. 6:20-cv-00075-ADA, Judge Alan D. Albright" } ], "negative_assertions": [], "evidence": { "render": "renders/p001.jpg", "render_sha256": "405ef47b1d6f620d5f7a9683b39ce2b8dc6d461d17a7edf1cd41ab9cc7a661a4", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 1087, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p001.txt", "phrase_presence": [ { "id": "p001-a", "literal_phrase_present": true }, { "id": "p001-b", "literal_phrase_present": true } ], "class": "sampled_text_present", "reason": "Selected visual phrases are present in this native control. This does not establish whole-page fidelity.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": false, "furniture_check_id": "p001-furniture", "metadata_check_id": "p001-metadata" }, { "physical_page": 20, "structure": "Single-column protective-order text; overlapping outer and inner filing stamps", "page_labels": [ "10", "x" ], "furniture": [ "Case: 23-1101", "Document: 15", "Filed: 05/09/2023" ], "adjustments": [], "phrases": [ { "id": "p020-a", "position": "first", "original_line_breaks": "other recordkeeping purposes, but may destroy such material (including electronic media\ncontaining such information)", "phrase": "other recordkeeping purposes, but may destroy such material (including electronic media containing such information)" }, { "id": "p020-b", "position": "middle", "original_line_breaks": "Each nonparty supplier shall be provided a copy of this order by the party seeking\ninformation from said supplier.", "phrase": "Each nonparty supplier shall be provided a copy of this order by the party seeking information from said supplier." } ], "negative_assertions": [], "evidence": { "render": "renders/p020.jpg", "render_sha256": "29132330b931220aea78d60536cd8d0c64fccd4574d07534b6854c179a0b815d", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 1893, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p020.txt", "phrase_presence": [ { "id": "p020-a", "literal_phrase_present": true }, { "id": "p020-b", "literal_phrase_present": true } ], "class": "sampled_text_present", "reason": "Selected visual phrases are present in this native control. This does not establish whole-page fidelity.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": false, "furniture_check_id": "p020-furniture", "metadata_check_id": "p020-metadata" }, { "physical_page": 39, "structure": "Single-column jury instruction with nested filing header and two footer labels", "page_labels": [ "2", "Appx4", "Page 2 of 41" ], "furniture": [ "Case: 23-1101", "Document: 15", "Appx4" ], "adjustments": [], "phrases": [ { "id": "p039-a", "position": "first", "original_line_breaks": "It is my duty and responsibility to instruct you on the law you are to apply in this case. The law\ncontained in these instructions is the only law you may follow.", "phrase": "It is my duty and responsibility to instruct you on the law you are to apply in this case. The law contained in these instructions is the only law you may follow." }, { "id": "p039-b", "position": "middle", "original_line_breaks": "The testimony of the witnesses and other exhibits introduced by the parties constitute the evidence.", "phrase": "The testimony of the witnesses and other exhibits introduced by the parties constitute the evidence." } ], "negative_assertions": [], "evidence": { "render": "renders/p039.jpg", "render_sha256": "71c48ca7a8697484eb3d737b4901ba78c61066713936c07aa890500ec0db8982", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 125, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p039.txt", "phrase_presence": [ { "id": "p039-a", "literal_phrase_present": false }, { "id": "p039-b", "literal_phrase_present": false } ], "class": "absent_body", "reason": "Native output contains only filing stamps and appendix label; visible instruction prose absent.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": true, "furniture_check_id": "p039-furniture", "metadata_check_id": "p039-metadata" }, { "physical_page": 58, "structure": "Single-column jury instruction", "page_labels": [ "21", "Appx23", "Page 21 of 41" ], "furniture": [ "Case: 23-1101", "Document: 15", "Appx23" ], "adjustments": [], "phrases": [ { "id": "p058-a", "position": "first", "original_line_breaks": "I will now instruct you how to decide whether or not EcoFactor has proven that Google has\ninfringed the Asserted Claims of the Asserted Patents.", "phrase": "I will now instruct you how to decide whether or not EcoFactor has proven that Google has infringed the Asserted Claims of the Asserted Patents." }, { "id": "p058-b", "position": "middle", "original_line_breaks": "In reaching your decision on infringement, keep in mind that only the claims of a patent can be\ninfringed.", "phrase": "In reaching your decision on infringement, keep in mind that only the claims of a patent can be infringed." } ], "negative_assertions": [], "evidence": { "render": "renders/p058.jpg", "render_sha256": "ae9c945acc7c4e2b8e580c2fc0ee9f3923c116a029daf15de77aeee0d686eca9", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 127, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p058.txt", "phrase_presence": [ { "id": "p058-a", "literal_phrase_present": false }, { "id": "p058-b", "literal_phrase_present": false } ], "class": "absent_body", "reason": "Native output contains only filing stamps and appendix label; visible instruction prose absent.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": true, "furniture_check_id": "p058-furniture", "metadata_check_id": "p058-metadata" }, { "physical_page": 77, "structure": "Single-column jury instruction", "page_labels": [ "40", "Appx42", "Page 40 of 41" ], "furniture": [ "Case: 23-1101", "Document: 15", "Appx42" ], "adjustments": [], "phrases": [ { "id": "p077-a", "position": "first", "original_line_breaks": "It is now your duty to deliberate and to consult with one another in an effort to reach a verdict.", "phrase": "It is now your duty to deliberate and to consult with one another in an effort to reach a verdict." }, { "id": "p077-b", "position": "middle", "original_line_breaks": "When you go into the jury room to deliberate, you may take with you a copy of this charge, the\nexhibits that I have admitted into evidence, and your notes.", "phrase": "When you go into the jury room to deliberate, you may take with you a copy of this charge, the exhibits that I have admitted into evidence, and your notes." } ], "negative_assertions": [], "evidence": { "render": "renders/p077.jpg", "render_sha256": "f0c9b740ca5f4403f8a596301fe388fe4d08ab45f305760c5178da7ecd967a08", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 127, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p077.txt", "phrase_presence": [ { "id": "p077-a", "literal_phrase_present": false }, { "id": "p077-b", "literal_phrase_present": false } ], "class": "absent_body", "reason": "Native output contains only filing stamps and appendix label; visible instruction prose absent.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": true, "furniture_check_id": "p077-furniture", "metadata_check_id": "p077-metadata" }, { "physical_page": 96, "structure": "Patent flowchart, not body prose; arrows and handwritten reference numbers", "page_labels": [ "Sheet 8 of 8", "Appx61" ], "furniture": [ "Case: 23-1101", "Document: 15", "Appx61" ], "adjustments": [ "No prose body; first substantive process node and near-middle process node selected. This does not score graph connectivity." ], "phrases": [ { "id": "p096-a", "position": "first", "original_line_breaks": "Transmit\ncontent data", "phrase": "Transmit content data" }, { "id": "p096-b", "position": "middle", "original_line_breaks": "Match content\ndata to user", "phrase": "Match content data to user" } ], "negative_assertions": [], "evidence": { "render": "renders/p096.jpg", "render_sha256": "065a272363dbb7aead425efad06b1a2eaae65fe3814282b2321f5756b66b5577", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 477, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p096.txt", "phrase_presence": [ { "id": "p096-a", "literal_phrase_present": false }, { "id": "p096-b", "literal_phrase_present": false } ], "class": "noisy_existing_layer", "reason": "Native output has Content where image reads content, an injected / O1 near Transmit, and ls where image reads Is. New recognition or visual reconciliation required for literal fidelity; body is not wholly absent.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": true, "furniture_check_id": "p096-furniture", "metadata_check_id": "p096-metadata" }, { "physical_page": 115, "structure": "Two-column patent body with central line-number gutter; left column must precede right column", "page_labels": [ "1", "2", "Appx80" ], "furniture": [ "Case: 23-1101", "Document: 15", "Appx80" ], "adjustments": [], "phrases": [ { "id": "p115-a", "position": "first-left", "original_line_breaks": "This application is a continuation of U.S. patent applica-\ntion Ser. No. 13/037,162", "phrase": "This application is a continuation of U.S. patent application Ser. No. 13/037,162", "wrap_hyphen_alternative": "This application is a continuation of U.S. patent applica- tion Ser. No. 13/037,162" }, { "id": "p115-b", "position": "middle-left", "original_line_breaks": "Climate control systems such as heating and cooling sys-\ntems for buildings", "phrase": "Climate control systems such as heating and cooling systems for buildings", "wrap_hyphen_alternative": "Climate control systems such as heating and cooling sys- tems for buildings" } ], "negative_assertions": [ { "id": "p115-gutter", "kind": "forbidden_interleaving", "before": "cooling sys-", "after": "tems for buildings", "forbidden": "30", "reason": "30 is a gutter line number, not a word between systems and for buildings." } ], "evidence": { "render": "renders/p115.jpg", "render_sha256": "c736965845d8df3106f3ae69a1229ad65cba403f56767d55c91cc38516ef59b7", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 7674, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p115.txt", "phrase_presence": [ { "id": "p115-a", "literal_phrase_present": false }, { "id": "p115-b", "literal_phrase_present": false } ], "class": "noisy_existing_layer", "reason": "Native output reads 137037,162 where source visibly reads 13/037,162 and splits applica tion without its visible wrap hyphen. New recognition or visual reconciliation required for literal fidelity.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": true, "furniture_check_id": "p115-furniture", "metadata_check_id": "p115-metadata" }, { "physical_page": 134, "structure": "Two-column patent body with central gutter numbers", "page_labels": [ "3", "4", "Appx99" ], "furniture": [ "Case: 23-1101", "Document: 15", "Appx99" ], "adjustments": [ "First passage is a continuation fragment, including visually printed 'munication may'; no grammatical repair." ], "phrases": [ { "id": "p134-a", "position": "first-left", "original_line_breaks": "munication may in the form of a signal sent over the power\nlines themselves, or a wireless communication over a data", "phrase": "munication may in the form of a signal sent over the power lines themselves, or a wireless communication over a data" }, { "id": "p134-b", "position": "middle-left", "original_line_breaks": "relationship between the inside temperature and the outside\ntemperature over time to derive a first prediction for the rate of", "phrase": "relationship between the inside temperature and the outside temperature over time to derive a first prediction for the rate of" } ], "negative_assertions": [ { "id": "p134-columns", "kind": "forbidden_interleaving", "before": "over the power", "after": "lines themselves", "forbidden": "By using these multiple data streams", "reason": "Right-column opening cannot interrupt the left-column sentence." } ], "evidence": { "render": "renders/p134.jpg", "render_sha256": "5c061d3ddc0317424e5132ba96735b8133c95877eb7590d64c4a583250a6fc78", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 7267, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p134.txt", "phrase_presence": [ { "id": "p134-a", "literal_phrase_present": true }, { "id": "p134-b", "literal_phrase_present": true } ], "class": "sampled_text_present", "reason": "Selected visual phrases are present in this native control. This does not establish whole-page fidelity.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": false, "furniture_check_id": "p134-furniture", "metadata_check_id": "p134-metadata" }, { "physical_page": 153, "structure": "Two-column patent body with central gutter numbers", "page_labels": [ "3", "4", "Appx118" ], "furniture": [ "Case: 23-1101", "Document: 15", "Appx118" ], "adjustments": [], "phrases": [ { "id": "p153-a", "position": "first-left", "original_line_breaks": "in the average home today would have limited value because\nthere are likely to be many places one or more people could", "phrase": "in the average home today would have limited value because there are likely to be many places one or more people could" }, { "id": "p153-b", "position": "middle-left", "original_line_breaks": "home in which the thermostat is located. The server logs the\nambient temperature sensed by each thermostat vs. time and", "phrase": "home in which the thermostat is located. The server logs the ambient temperature sensed by each thermostat vs. time and" } ], "negative_assertions": [ { "id": "p153-columns", "kind": "forbidden_interleaving", "before": "limited value because", "after": "there are likely", "forbidden": "FIG. 2 shows a high-level illustration", "reason": "Right-column figure description cannot interrupt left-column prose." } ], "evidence": { "render": "renders/p153.jpg", "render_sha256": "8438148c487f562c5595a6c868704dcc3fa43d1fd8a85679b8922847121daea8", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 63, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p153.txt", "phrase_presence": [ { "id": "p153-a", "literal_phrase_present": false }, { "id": "p153-b", "literal_phrase_present": false } ], "class": "absent_body", "reason": "Native output contains only outer filing stamp and appendix label; both visible patent columns absent.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": true, "furniture_check_id": "p153-furniture", "metadata_check_id": "p153-metadata" }, { "physical_page": 172, "structure": "Docket table: date, entry number, narrative; preserve row association and within-cell reading order", "page_labels": [ "Page 16 of 40", "Appx137" ], "furniture": [ "Case: 23-1101", "Document: 15", "Appx137", "Centralized CM/ECF LIVE - U.S. District Court:txwd" ], "adjustments": [ "First and middle body passages selected from narrative cells, not date or docket-number columns." ], "phrases": [ { "id": "p172-a", "position": "first", "original_line_breaks": "REPLY to Response to Motion, filed by Google LLC, re 64 Opposed MOTION to\nAmend/Correct for Leave to Amend Invalidity Contentions", "phrase": "REPLY to Response to Motion, filed by Google LLC, re 64 Opposed MOTION to Amend/Correct for Leave to Amend Invalidity Contentions" }, { "id": "p172-b", "position": "middle", "original_line_breaks": "MOTION to Withdraw as Attorney C. Jay Chung by EcoFactor, Inc.. (Attachments: #\n1 Proposed Order)", "phrase": "MOTION to Withdraw as Attorney C. Jay Chung by EcoFactor, Inc.. (Attachments: # 1 Proposed Order)" } ], "negative_assertions": [], "evidence": { "render": "renders/p172.jpg", "render_sha256": "66e78f0df215177864b0dc1042831e2fd64709ecac574caefd59fcaac9a93fa0", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 2815, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p172.txt", "phrase_presence": [ { "id": "p172-a", "literal_phrase_present": true }, { "id": "p172-b", "literal_phrase_present": true } ], "class": "sampled_text_present", "reason": "Selected visual phrases are present in this native control. This does not establish whole-page fidelity.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": false, "furniture_check_id": "p172-furniture", "metadata_check_id": "p172-metadata" }, { "physical_page": 191, "structure": "Docket table with a row continued from previous physical page", "page_labels": [ "Page 35 of 40", "Appx156" ], "furniture": [ "Case: 23-1101", "Document: 15", "Appx156", "Centralized CM/ECF LIVE - U.S. District Court:txwd" ], "adjustments": [ "First passage begins inside continuing narrative cell; it is not a missing title." ], "phrases": [ { "id": "p191-a", "position": "first", "original_line_breaks": "purchased from the court reporter or viewed at the clerk's office public terminal. If\nredaction is necessary, a Notice of Redaction Request must be filed within 21 days.", "phrase": "purchased from the court reporter or viewed at the clerk's office public terminal. If redaction is necessary, a Notice of Redaction Request must be filed within 21 days." }, { "id": "p191-b", "position": "middle", "original_line_breaks": "NOTICE of Attorney Appearance by Shaun William Hassett on behalf of Google LLC", "phrase": "NOTICE of Attorney Appearance by Shaun William Hassett on behalf of Google LLC" } ], "negative_assertions": [], "evidence": { "render": "renders/p191.jpg", "render_sha256": "9fa289c00d092ae8c51fbe738830a5b739553dbb7c95650bf9b687b90ddd9d75", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 3445, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p191.txt", "phrase_presence": [ { "id": "p191-a", "literal_phrase_present": true }, { "id": "p191-b", "literal_phrase_present": true } ], "class": "sampled_text_present", "reason": "Selected visual phrases are present in this native control. This does not establish whole-page fidelity.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": false, "furniture_check_id": "p191-furniture", "metadata_check_id": "p191-metadata" }, { "physical_page": 210, "structure": "Deposition transcript body; left line-number gutter 1–25, right timestamps, Q/A speaker labels; retain speech but not gutters in clean prose", "page_labels": [ "Page 131", "Appx1167", "Page 7 of 16" ], "furniture": [ "Case: 23-1101", "Document: 15", "Appx1167", "Veritext Legal Solutions", "866 299-5127" ], "adjustments": [], "phrases": [ { "id": "p210-a", "position": "first", "original_line_breaks": "place. There was no operational software, but in my\nhead, I was able to do that on a\none-house-by-one-house basis.", "phrase": "place. There was no operational software, but in my head, I was able to do that on a one-house-by-one-house basis." }, { "id": "p210-b", "position": "middle", "original_line_breaks": "the hardware. It was a modification to the software\nthat was actually running on the hardware.", "phrase": "the hardware. It was a modification to the software that was actually running on the hardware." } ], "negative_assertions": [ { "id": "p210-gutter", "kind": "forbidden_interleaving", "before": "but in my", "after": "head, I was able", "forbidden": "2", "reason": "Line number 2 is not part of the witness's sentence." }, { "id": "p210-timestamp", "kind": "forbidden_interleaving", "before": "but in my", "after": "head, I was able", "forbidden": "14:21:52", "reason": "Timestamp is outside the spoken text." } ], "evidence": { "render": "renders/p210.jpg", "render_sha256": "cbec30feec0e216fd504f3ba9c0d30943774c018a40997048af9a710943de07a", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 1417, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p210.txt", "phrase_presence": [ { "id": "p210-a", "literal_phrase_present": false }, { "id": "p210-b", "literal_phrase_present": false } ], "class": "sampled_text_present", "reason": "Selected visual phrases are present in this native control. This does not establish whole-page fidelity.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": false, "furniture_check_id": "p210-furniture", "metadata_check_id": "p210-metadata" }, { "physical_page": 229, "structure": "Mathematical declaration with pasted two-page extract: equations, internal running header and internal page labels within outer page body", "page_labels": [ "1", "2", "15", "Appx1192", "Page 4 of 5" ], "furniture": [ "Case: 23-1101", "Document: 15", "Appx1192", "RESTRICTED – CONFIDENTIAL SOURCE CODE", "NEWTON'S LAW OF COOLING OR HEATING" ], "adjustments": [ "Displayed equations precede first prose; target first meaningful prose rather than treating equations as linear words. Second target begins after inner running header; equations are not scored." ], "phrases": [ { "id": "p229-a", "position": "first-prose", "original_line_breaks": "Now here is where we need to be careful. We want to drop the absolute value signs to", "phrase": "Now here is where we need to be careful. We want to drop the absolute value signs to" }, { "id": "p229-b", "position": "middle-prose", "original_line_breaks": "Since the object is cooling down to the surrounding temperature, T will always be greater\nthan M so A will be a positive value.", "phrase": "Since the object is cooling down to the surrounding temperature, T will always be greater than M so A will be a positive value." } ], "negative_assertions": [], "evidence": { "render": "renders/p229.jpg", "render_sha256": "60f1b2cb1cfe19d82c6318d09169d7b71f01d60afc5f1e645d2e78a52afa5d9f", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 635, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p229.txt", "phrase_presence": [ { "id": "p229-a", "literal_phrase_present": false }, { "id": "p229-b", "literal_phrase_present": false } ], "class": "partial_body", "reason": "Native output retains declaration paragraph37 but omits pasted mathematical source pages and their prose.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": true, "furniture_check_id": "p229-furniture", "metadata_check_id": "p229-metadata" }, { "physical_page": 248, "structure": "Single-column judicial prose with section heading and block quotation; redacted running header above body", "page_labels": [ "423", "Appx1919" ], "furniture": [ "Case: 23-1101", "Document: 15", "Appx1919", "PUBLIC VERSION" ], "adjustments": [], "phrases": [ { "id": "p248-a", "position": "first", "original_line_breaks": "These claims are expressly directed to technical improvements to then-existing\ntechnology, namely those associated with electronically programmable thermostats and other", "phrase": "These claims are expressly directed to technical improvements to then-existing technology, namely those associated with electronically programmable thermostats and other" }, { "id": "p248-b", "position": "middle", "original_line_breaks": "thermostat HVAC control system at the time. For instance, the inventions use a networked\nthermostat HVAC control system", "phrase": "thermostat HVAC control system at the time. For instance, the inventions use a networked thermostat HVAC control system" } ], "negative_assertions": [], "evidence": { "render": "renders/p248.jpg", "render_sha256": "823d438f8bb9ed8f3719884545dda2f959194aa37f3a6e5970a83bb05198d5d6", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 2032, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p248.txt", "phrase_presence": [ { "id": "p248-a", "literal_phrase_present": true }, { "id": "p248-b", "literal_phrase_present": true } ], "class": "sampled_text_present", "reason": "Selected visual phrases are present in this native control. This does not establish whole-page fidelity.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": false, "furniture_check_id": "p248-furniture", "metadata_check_id": "p248-metadata" }, { "physical_page": 267, "structure": "Sparse certificate of service; inner filing stamp appears around page middle", "page_labels": [ "Page 4 of 4", "Appx2253" ], "furniture": [ "Case: 23-1101", "Document: 15", "Appx2253", "Case 6:20-cv-00075-ADA", "Document 186" ], "adjustments": [ "Only one prose paragraph; split into two nonoverlapping phrases instead of pretending two independent passages exist." ], "phrases": [ { "id": "p267-a", "position": "first", "original_line_breaks": "The undersigned hereby certifies that on January 26, 2022, the foregoing was served on", "phrase": "The undersigned hereby certifies that on January 26, 2022, the foregoing was served on" }, { "id": "p267-b", "position": "middle-same-paragraph", "original_line_breaks": "all counsel of record by e-mail.", "phrase": "all counsel of record by e-mail." } ], "negative_assertions": [], "evidence": { "render": "renders/p267.jpg", "render_sha256": "3e5262dc56b2d596a0fa81307457bc283e5ac8bed6be252bd96098fb4d44caec", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 293, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p267.txt", "phrase_presence": [ { "id": "p267-a", "literal_phrase_present": true }, { "id": "p267-b", "literal_phrase_present": true } ], "class": "sampled_text_present", "reason": "Selected visual phrases are present in this native control. This does not establish whole-page fidelity.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": false, "furniture_check_id": "p267-furniture", "metadata_check_id": "p267-metadata" }, { "physical_page": 286, "structure": "Three-column claim-construction table; each cell has its own text stream", "page_labels": [ "2", "Appx6752", "Page 3 of 7" ], "furniture": [ "Case: 23-1101", "Document: 15", "Appx6752" ], "adjustments": [ "First target from first term cell; middle target from defendant proposal cell nearest vertical midpoint. Table interpretation/association is documented but not reduced to prose recall." ], "phrases": [ { "id": "p286-a", "position": "first-cell", "original_line_breaks": "rate of change in inside\ntemperature", "phrase": "rate of change in inside temperature" }, { "id": "p286-b", "position": "middle-cell", "original_line_breaks": "Indefinite due to lack of\nessential structural\nconnections, under In re", "phrase": "Indefinite due to lack of essential structural connections, under In re" } ], "negative_assertions": [ { "id": "p286-cells", "kind": "forbidden_interleaving", "before": "rate of change in inside", "after": "temperature", "forbidden": "the difference between", "reason": "Proposed-construction text in another cell must not interrupt the term phrase." } ], "evidence": { "render": "renders/p286.jpg", "render_sha256": "ed846cd46c9ba98a5ea459b0f2a0e6739efd1ca32a23a26c9ae3475375ec73f5", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 2423, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p286.txt", "phrase_presence": [ { "id": "p286-a", "literal_phrase_present": true }, { "id": "p286-b", "literal_phrase_present": true } ], "class": "sampled_text_present", "reason": "Selected visual phrases are present in this native control. This does not establish whole-page fidelity.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": false, "furniture_check_id": "p286-furniture", "metadata_check_id": "p286-metadata" }, { "physical_page": 305, "structure": "Patent drawings: form UI, consumption graph and dropdown; no running body prose", "page_labels": [ "Sheet 18 of 18", "DTX0219, Page 19 of 50", "Appx10124", "DC_PRIOR_ART_0000421" ], "furniture": [ "Case: 23-1101", "Document: 15", "Appx10124", "DTX0219, Page 19 of 50", "DC_PRIOR_ART_0000421" ], "adjustments": [ "Figure labels substitute for unavailable body prose; does not score graph numeric readings.", "Unusual comma spacing verified in enlarged visual crop and retained literally." ], "phrases": [ { "id": "p305-a", "position": "first-figure", "original_line_breaks": "Program Definitions", "phrase": "Program Definitions" }, { "id": "p305-b", "position": "middle-figure", "original_line_breaks": "Energy Consumption for Tuesday , March 18 ,2003", "phrase": "Energy Consumption for Tuesday , March 18 ,2003", "detail_render": "renders/p305-title-detail.png" } ], "negative_assertions": [], "evidence": { "render": "renders/p305.jpg", "render_sha256": "0756a14e0fb90c9474417181fbd996c412eaf803615d26faa38f8920cf0c74b0", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 929, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p305.txt", "phrase_presence": [ { "id": "p305-a", "literal_phrase_present": true }, { "id": "p305-b", "literal_phrase_present": true } ], "class": "noisy_existing_layer", "reason": "Native layer is present but figure controls contain visible/native mismatches; title spacing also differs. Not classified as absent body.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": true }, "needs_OCR": true, "furniture_check_id": "p305-furniture", "metadata_check_id": "p305-metadata" }, { "physical_page": 324, "structure": "Two-column patent body; paragraph numbers are content, not gutter furniture", "page_labels": [ "19", "Appx10143", "DTX0219, Page 38 of 50", "DC_PRIOR_ART_0000440" ], "furniture": [ "Case: 23-1101", "Document: 15", "Appx10143", "DTX0219, Page 38 of 50", "DC_PRIOR_ART_0000440" ], "adjustments": [ "Preserve printed 'simplest from' rather than correcting it to 'simplest form'." ], "phrases": [ { "id": "p324-a", "position": "first-left", "original_line_breaks": "should be noted here that the set point of the system 3.08 was\nset at a fixed point for the entire day and the use of humidity", "phrase": "should be noted here that the set point of the system 3.08 was set at a fixed point for the entire day and the use of humidity" }, { "id": "p324-b", "position": "middle-left", "original_line_breaks": "While this example would in its simplest from provide a\nselection of 10 options", "phrase": "While this example would in its simplest from provide a selection of 10 options" } ], "negative_assertions": [ { "id": "p324-columns", "kind": "forbidden_interleaving", "before": "system 3.08 was", "after": "set at a fixed point", "forbidden": "For the maximum savings setting", "reason": "Right-column continuation must not interrupt the left-column sentence." } ], "evidence": { "render": "renders/p324.jpg", "render_sha256": "99c5d160c5206a2a228cacca801705a270835895f748b39be386b61804d012c0", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 7824, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p324.txt", "phrase_presence": [ { "id": "p324-a", "literal_phrase_present": false }, { "id": "p324-b", "literal_phrase_present": false } ], "class": "noisy_existing_layer", "reason": "Native output changes visible lowercase set, sensing, so to Set, Sensing, So. Literal fidelity not established by nonzero native count.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": true, "furniture_check_id": "p324-furniture", "metadata_check_id": "p324-metadata" }, { "physical_page": 343, "structure": "Confidential-material removal notice, not blank and not missing OCR", "page_labels": [ "APPX10364", "APPX10367" ], "furniture": [ "Case: 23-1101", "Document: 15", "Filed: 05/09/2023" ], "adjustments": [ "No substantive exhibit remains. Score notice text, not unavailable removed material. APPX references are body content and must not be removed as footers." ], "phrases": [ { "id": "p343-a", "position": "first-notice", "original_line_breaks": "Pages APPX10364;\nAPPX10367", "phrase": "Pages APPX10364; APPX10367" }, { "id": "p343-b", "position": "middle-notice", "original_line_breaks": "Removed Due to Confidential Material", "phrase": "Removed Due to Confidential Material" } ], "negative_assertions": [], "evidence": { "render": "renders/p343.jpg", "render_sha256": "2f543a1d4405186b2b7b85976eb1bb1425ae9e63ea30f9cc9d72ca22a8ada9db", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 121, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p343.txt", "phrase_presence": [ { "id": "p343-a", "literal_phrase_present": true }, { "id": "p343-b", "literal_phrase_present": true } ], "class": "sampled_text_present", "reason": "Selected visual phrases are present in this native control. This does not establish whole-page fidelity.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": false, "furniture_check_id": "p343-furniture", "metadata_check_id": "p343-metadata" }, { "physical_page": 362, "structure": "Confidential-material removal notice, not blank and not missing OCR", "page_labels": [ "APPX10439", "APPX10467" ], "furniture": [ "Case: 23-1101", "Document: 15", "Filed: 05/09/2023" ], "adjustments": [ "No substantive exhibit remains. Score notice text, not unavailable removed material. APPX references are body content and must not be removed as footers." ], "phrases": [ { "id": "p362-a", "position": "first-notice", "original_line_breaks": "Pages APPX10439;\nAPPX10467", "phrase": "Pages APPX10439; APPX10467" }, { "id": "p362-b", "position": "middle-notice", "original_line_breaks": "Removed Due to Confidential Material", "phrase": "Removed Due to Confidential Material" } ], "negative_assertions": [], "evidence": { "render": "renders/p362.jpg", "render_sha256": "f984cb4011fcd1d5832bb4e4d04bc35391dfc2bbe118b250f8b811b895d5b5e0", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 121, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p362.txt", "phrase_presence": [ { "id": "p362-a", "literal_phrase_present": true }, { "id": "p362-b", "literal_phrase_present": true } ], "class": "sampled_text_present", "reason": "Selected visual phrases are present in this native control. This does not establish whole-page fidelity.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": false, "furniture_check_id": "p362-furniture", "metadata_check_id": "p362-metadata" }, { "physical_page": 373, "structure": "Webpage screenshot/exhibit with running prose, bold questions and cropped graphic", "page_labels": [ "PTX-0263-0003", "Appx10894", "EF_0895725" ], "furniture": [ "Case: 23-1101", "Document: 15", "Appx10894", "PTX-0263-0003", "EF_0895725" ], "adjustments": [], "phrases": [ { "id": "p373-a", "position": "first", "original_line_breaks": "we’re taking steps to make that possible for most systems in Canada and the U.S.\nwith HVAC monitoring, which is rolling out today to all Nest thermostats in those regions.", "phrase": "we’re taking steps to make that possible for most systems in Canada and the U.S. with HVAC monitoring, which is rolling out today to all Nest thermostats in those regions." }, { "id": "p373-b", "position": "middle", "original_line_breaks": "Ramya: Cloud computing advancements, definitely. We used to run a lot of algorithms on\nthe device, that’s what got Nest started.", "phrase": "Ramya: Cloud computing advancements, definitely. We used to run a lot of algorithms on the device, that’s what got Nest started." } ], "negative_assertions": [], "evidence": { "render": "renders/p373.jpg", "render_sha256": "6144bbea6b341fb75165c11f9e0e345d4e11b5613250f3d2c6a0db903f10fe18", "read_directly": true, "independent_visual_recheck": "claude-opus-5", "coordinate_system": "physical page, 1-based; original image pixel dimensions in render-manifest.json" }, "native_layer_diagnostic": { "characters": 2114, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p373.txt", "phrase_presence": [ { "id": "p373-a", "literal_phrase_present": true }, { "id": "p373-b", "literal_phrase_present": true } ], "class": "sampled_text_present", "reason": "Selected visual phrases are present in this native control. This does not establish whole-page fidelity.", "phase": "Pre-freeze diagnostic control, not source of visual gold.", "phrase_presence_is_pre_reconciliation": false }, "needs_OCR": false, "furniture_check_id": "p373-furniture", "metadata_check_id": "p373-metadata" } ], "illustrations": [ { "physical_page": 115, "purpose": "Genuine two-column patent body and gutter numbers", "sample_overlap": true, "scope": "Purposive display role only; no additional weight in sample denominator.", "render": "renders/p115.jpg", "render_sha256": "c736965845d8df3106f3ae69a1229ad65cba403f56767d55c91cc38516ef59b7" }, { "physical_page": 210, "purpose": "Genuine line-numbered transcript body, Q/A and timestamps", "sample_overlap": true, "scope": "Purposive display role only; no additional weight in sample denominator.", "render": "renders/p210.jpg", "render_sha256": "cbec30feec0e216fd504f3ba9c0d30943774c018a40997048af9a710943de07a" }, { "physical_page": 202, "purpose": "Ordinary brief with two footnotes 5 and 6, body subheadings and nested filing stamps", "sample_overlap": false, "page_labels": [ "14", "Appx1151", "Page 18 of 27" ], "positive_phrases": [ { "id": "demo202-a", "original_line_breaks": "By only reciting generic components performing conventional functions, the claims fail\nto specify how any of claimed elements", "phrase": "By only reciting generic components performing conventional functions, the claims fail to specify how any of claimed elements" }, { "id": "demo202-footnote", "original_line_breaks": "As discussed above, the ’488 and ’327 patents share a specification and have many overlapping\nclaim elements.", "phrase": "As discussed above, the ’488 and ’327 patents share a specification and have many overlapping claim elements." } ], "scope": "Demonstration only; do not merge into fixed sample denominator.", "render": "renders/p202.jpg", "render_sha256": "e76fa59f62e8a64448dff7205619105882106084bd3f8c67ce95444c01fd7be2", "native_layer_diagnostic": { "characters": 2423, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p202.txt", "phrase_presence": [ { "id": "demo202-a", "literal_phrase_present": true }, { "id": "demo202-footnote", "literal_phrase_present": true } ] }, "needs_OCR": false, "evidence": "Both primary visual reader and independent Opus vision recheck confirmed all listed phrases and labels." }, { "physical_page": 229, "purpose": "Pasted mathematical source pages with internal running header and multiple pagination layers", "sample_overlap": true, "scope": "Purposive display role only; no additional weight in sample denominator.", "render": "renders/p229.jpg", "render_sha256": "60f1b2cb1cfe19d82c6318d09169d7b71f01d60afc5f1e645d2e78a52afa5d9f" }, { "physical_page": 75, "purpose": "Known challenge page: visible jury instruction must not be confused with nonzero filing-stamp text", "sample_overlap": false, "page_labels": [ "38", "Appx40", "Page 38 of 41" ], "positive_phrases": [ { "id": "demo075-a", "original_line_breaks": "Any amount you find as of damages must be based on the value attributable to the patented\ninvention, as distinct from unpatented features", "phrase": "Any amount you find as of damages must be based on the value attributable to the patented invention, as distinct from unpatented features" }, { "id": "demo075-b", "original_line_breaks": "Apportionment can be addressed in a variety of ways, including by careful selection of the\nroyalty base", "phrase": "Apportionment can be addressed in a variety of ways, including by careful selection of the royalty base" } ], "scope": "Known purposive challenge; not part of unbiased sample denominator.", "render": "renders/p075.jpg", "render_sha256": "b7c91c1d6609397156334c6a06dcf90a11d565ee538538cba8f24757f58306c8", "native_layer_diagnostic": { "characters": 127, "source": "pypdfium2 get_text_range; diagnostic only, not visual truth", "path": "native-control/p075.txt", "phrase_presence": [ { "id": "demo075-a", "literal_phrase_present": false }, { "id": "demo075-b", "literal_phrase_present": false } ] }, "needs_OCR": true, "evidence": "Both primary visual reader and independent Opus vision recheck confirmed all listed phrases and labels." } ], "frozen_at": "2026-09-12T12:33:32.535216+00:00", "blindness": "No new extractor code or output inspected before freeze. Native PDFium control examined only after visual phrases were authored; raw diagnostic disagreements did not determine page or phrase selection.", "reconciliation_file": "provenance/visual-reconciliation.json", "needs_OCR_contract": "True on absent/partial bodies and on demonstrably noisy existing layers. Always stratify absent_body, partial_body, noisy_existing_layer; do not treat these as one missing-body rate. Flag detection score should separately report absent/partial and noisy-layer cases.", "counts": { "sample_pages": 21, "sample_phrase_checks": 42, "sample_negative_checks": 7, "sample_furniture_page_checks": 21, "sample_metadata_label_checks": 57, "purposive_extra_pages": 2, "purposive_extra_phrase_checks": 4 }, "original_private_gold_sha256": "4e13d0727b80b78305b5cc61d61ef82678a53846d9cb49bd3caacefa275f8394", "public_derivative_note": "Only machine-local evidence paths have been replaced by portable paths; all check IDs, phrases, line breaks, expected labels and selection remain unchanged. The original private artifact remains frozen." }