diff --git a/scripts/scan_dolmadocs.py b/scripts/scan_dolmadocs.py index f85abd7..20716d7 100644 --- a/scripts/scan_dolmadocs.py +++ b/scripts/scan_dolmadocs.py @@ -243,6 +243,9 @@ def create_html_output(random_pages, pdf_s3_client, output_path, workspace_path, color: var(--text-color); background-color: var(--bg-color); padding: 2rem; + display: flex; + flex-direction: row; + gap: 2rem; }} ul {{ @@ -250,15 +253,23 @@ def create_html_output(random_pages, pdf_s3_client, output_path, workspace_path, }} .container {{ - max-width: 1200px; - margin: 0 auto; + flex: 2; + max-width: 750px; }} header {{ - margin-bottom: 2rem; - border-bottom: 1px solid var(--border-color); - padding-bottom: 1rem; - + position: sticky; + top: 2rem; + flex: 1; + min-width: 380px; + max-width: 420px; + max-height: calc(100vh - 4rem); + overflow-y: auto; + padding: 1.5rem; + background-color: white; + border-radius: 0.5rem; + box-shadow: var(--card-shadow); + align-self: flex-start; font-size: small; }} @@ -296,7 +307,7 @@ def create_html_output(random_pages, pdf_s3_client, output_path, workspace_path, .page-grid {{ display: grid; - grid-template-columns: repeat(2, 1fr); + grid-template-columns: 1fr; gap: 2rem; }} @@ -494,53 +505,61 @@ def create_html_output(random_pages, pdf_s3_client, output_path, workspace_path, @media (max-width: 768px) {{ body {{ padding: 1rem; + flex-direction: column; }} - .page-grid {{ - grid-template-columns: 1fr; + header {{ + position: static; + max-width: 100%; + margin-left: 0; + margin-bottom: 2rem; + }} + + .container {{ + max-width: 100%; }} }} -
-
+

Task Instructions

-

Your task is to review {len(random_pages)} document pages and determine whether they contain any Personally Identifiable Information (PII). Carefully but efficiently inspect each page and select the appropriate response. You do not need to read every word - quickly scan the page and look for any obvious PII. The time expected to complete this task is 10-15 minutes.

- -

How to Annotate

-

The page you are currently annotating will be highlighted with a blue outline and a set of response buttons will be displayed directly below it.

-
-

Yes PII - Select this if you find any information on the page that qualifies as PII. A text box will appear below - briefly describe the kind of PII you encountered (e.g., full name, social security number, etc.) then press the Enter key.

-

No PII - Select this if the page does not contain any PII.

-

I cannot read this - Select this if you are unable to read the page for any reason (e.g., written in a language other than English, heavily redacted text, etc.)

-

Disturbing content - Select this if the page contains disturbing or graphic content.

- -
-

You may edit your annotations any time before submitting. To do so, press the green Edit button directly above the page.

-

After completing all the document pages on this screen, you will receive a Prolific completion code.

- -

What Counts as PII?

-
    -
  • Names: Full names, first names, last names, nicknames, maiden names, aliases
    • -
  • Addresses: Street addresses, postal codes, cities, states, countries
    • -
  • Contact Information: Phone numbers, email addresses
    • -
  • Government IDs: SSNs, passport numbers, driver's license numbers, tax IDs
    • -
  • Financial Information: Credit card numbers, bank account numbers, routing numbers
    • -
  • Biometric Data: Fingerprints, retina scans, facial recognition data, voice signatures
    • -
  • Personal Attributes: Date of birth, place of birth, gender, race, religion
    • -
  • Online Identifiers: IP addresses, login IDs, usernames, passwords, API keys, URLs
    • -
  • Location Information: Geolocations, specific coordinates
    • -
  • Employment Information: Job titles, workplace names, employment history
    • -
  • Education Information: School names, degrees, transcripts
    • -
  • Medical Information: Health records, diagnoses
    • -
  • Company Names: If they are tied to an individual's identity (e.g., a person's personal business)
    • -
- -

What NOT to Mark as PII

-

Author names, researcher names, citations, or references from published research papers should NOT be marked as PII. These names are part of the normal publication process and are not considered private or sensitive information for the purposes of this task. - Only mark information as PII if it relates to private, sensitive, or personal details about an individual outside the context of the publication.

-
+

Your task is to review {len(random_pages)} document pages and determine whether they contain any Personally Identifiable Information (PII). Carefully but efficiently inspect each page and select the appropriate response. You do not need to read every word - quickly scan the page and look for any obvious PII. The time expected to complete this task is 10-15 minutes.

+ +

How to Annotate

+

The page you are currently annotating will be highlighted with a blue outline and a set of response buttons will be displayed directly below it.

+
+

Yes PII - Select this if you find any information on the page that qualifies as PII. A text box will appear below - briefly describe the kind of PII you encountered (e.g., full name, social security number, etc.) then press the Enter key.

+

No PII - Select this if the page does not contain any PII.

+

I cannot read this - Select this if you are unable to read the page for any reason (e.g., written in a language other than English, heavily redacted text, etc.)

+

Disturbing content - Select this if the page contains disturbing or graphic content.

+ +
+

You may edit your annotations any time before submitting. To do so, press the green Edit button directly above the page.

+

After completing all the document pages on this screen, you will receive a Prolific completion code.

+ +

What Counts as PII?

+ + +

What NOT to Mark as PII

+

Author names, researcher names, citations, or references from published research papers should NOT be marked as PII. These names are part of the normal publication process and are not considered private or sensitive information for the purposes of this task. + Only mark information as PII if it relates to private, sensitive, or personal details about an individual outside the context of the publication.

+
+
@@ -591,7 +610,7 @@ def create_html_output(random_pages, pdf_s3_client, output_path, workspace_path, html_content += f"""
-

{original_url}

+

{original_url}

Page {page_num}

{f'View Cached PDF' if presigned_url else pdf_path}

@@ -623,7 +642,7 @@ def create_html_output(random_pages, pdf_s3_client, output_path, workspace_path, html_content += f"""

-

{original_url}

+

original_url

Page {page_num}

{f'View Cached PDF' if presigned_url else pdf_path}