[
    {
        "role": "user",
        "content": "Create%20a%20presentation%20I%20can%20use%20from%20the%20description%20below.%20I%20do%20not%20need%20doucments%20included.%20The%20goal%20of%20this%20story%20is%20to%20demonstrate%20how%20a%20screen%20reader%20user%20would%20access%20information%20in%20a%20PDF%20(both%20tagged%20and%20untagged)%20and%20a%20PowerPoint%20presentation.%20Please%20show%20how%20screen%20readers%20navigate%20these%20types%20of%20documents%20and%20be%20prepared%20to%20talk%20about%20how%20accessing%20information%20is%20different%20than%20a%20web%20page%20that%20uses%20HTML.%20The%20team%20should%20learn%20how%20they%20can%20use%20screen%20readers%20to%20test%20these%20types%20of%20documents%20to%20uncover%20potential%20accessibility%20barriers."
    },
    {
        "role": "assistant",
        "content": "Here\u2019s a slide-by-slide presentation outline you can use to teach your team *how screen reader users access information* in **PDFs** (tagged vs untagged) and **PowerPoint**, contrast with web pages (HTML), and show how testing reveals barriers. You can paste this into PowerPoint, Google Slides, etc.  \nUse each \u201cSlide #\u201d as a slide title.\n\n---\n\n**Slide 1: Title Slide**  \n- Title: \u201cHow Screen Reader Users Access PDFs & PowerPoints vs Web Pages\u201d  \n- Subtitle: \u201cUnderstanding tagged/untagged PDFs, PowerPoint navigation, and HTML differences\u201d  \n- Your name / date / team name  \n\n---\n\n**Slide 2: Why Accessibility Matters**  \n- Everyone deserves equal access to information \u2014 legal, ethical, usability reasons  \n- PDFs and PowerPoints are common in reports, training, sharing material  \n- If not accessible, people who are blind or have low vision may be blocked  \n\n---\n\n**Slide 3: What Screen Readers Do**  \n- Software that reads text aloud or sends to refreshable Braille display  \n- Navigate via structure: headings, lists, tables, links  \n- Use keyboard commands and special navigation shortcuts (e.g. \u201cprev heading\u201d, \u201cnext heading\u201d)  \n\n---\n\n**Slide 4: Tagged vs Untagged PDFs \u2013 What Are They?**  \n- Tagged PDF: includes hidden structure-tags (headings, paragraphs, tables, alt text) that define reading order and relationships between elements ([accessible-docs.com](https://accessible-docs.com/tagging-accessible-pdf/?utm_source=openai))  \n- Untagged PDF: just visual appearance; no semantic info; screen reader must infer, often poorly ([barrierefreies-webdesign.de](https://www.barrierefreies-webdesign.de/knowhow/pdf-screenreader/tagged-pdf.html?utm_source=openai))  \n\n---\n\n**Slide 5: How a Screen Reader Accesses a Tagged PDF**  \n- Announces document title & language if metadata present ([binghamton.edu](https://www.binghamton.edu/ssd/accessibility/accessibility-features.pdf?utm_source=openai))  \n- Allows navigation by headings (headings hierarchy), lists, tables, links ([accessible-docs.com](https://accessible-docs.com/tagging-accessible-pdf/?utm_source=openai))  \n- Can skip sections, jump to specific parts quickly  \n\n---\n\n**Slide 6: How a Screen Reader Handles an Untagged PDF**  \n- Reads almost everything in visual order (page order), often as one long stream of text without heading landmarks ([barrierefreies-webdesign.de](https://www.barrierefreies-webdesign.de/knowhow/pdf-screenreader/tagged-pdf.html?utm_source=openai))  \n- Tables read like linear text\u2014no row/column context ([barrierefreies-webdesign.de](https://www.barrierefreies-webdesign.de/knowhow/pdf-screenreader/tagged-pdf.html?utm_source=openai))  \n- Only certain tools or readers may offer \u201creading order\u201d panels, but with limited reliability  \n\n---\n\n**Slide 7: Demonstration Scenario \u2013 PDF (Tagged vs Untagged)**  \n- Use JAWS, NVDA, or VoiceOver to open two PDFs: one tagged, one untagged with same content  \n- Ask:  \n  \u2022 Can you list all headings?  \n  \u2022 Can you use the shortcut to jump to the next heading?  \n  \u2022 Can you move through a table, reading header info automatically?  \n  \u2022 How is decorative image handled?  \n\n---\n\n**Slide 8: PowerPoint Documents \u2013 Common Structure**  \n- Slides with *titles*, *placeholders*, *bulleted lists*, *images/charts*  \n- Objects layered (z-order), placeholders have internal roles  \n- Built-in layout vs manually drawn text boxes  \n\n---\n\n**Slide 9: How Screen Reader Users Navigate PowerPoint**  \n- In editing mode or reading mode: can navigate via slide titles, outline view or object selection pane ([mass.gov](https://www.mass.gov/info-details/powerpoint-accessibility?utm_source=openai))  \n- Use Reading Order Pane (Review \u2192 Check Accessibility \u2192 Reading Order) to see order of content on each slide ([section508.gov](https://www.section508.gov/blog/accessibility-bytes/presentation-reading-order/?utm_source=openai))  \n- Alt text on images, marking decorative items as decorative so screen reader ignores them ([colorado.edu](https://www.colorado.edu/digital-accessibility/resources/understanding-powerpoint-accessibility?utm_source=openai))  \n\n---\n\n**Slide 10: Common Barriers in PowerPoint**  \n- Decorative elements not marked \u201cdecorative\u201d\u2014screen reader reads them unnecessarily  \n- Incorrect reading/order sequencing: visuals read in wrong order  \n- Missing alt text or poor descriptions for visuals ([colorado.edu](https://www.colorado.edu/digital-accessibility/resources/understanding-powerpoint-accessibility?utm_source=openai))  \n- Complex graphics without long descriptions  \n\n---\n\n**Slide 11: Web Pages (HTML) vs PDFs & PowerPoint**  \n- HTML pages inherently semantically structured: headings (h1-h6), lists, links, table markup, ARIA, etc.  \n- Screen readers and browsers collaborate via accessibility APIs; navigation by landmarks, semantic tags, ARIA ([mathtech.org](https://mathtech.org/2025/04/25/screen-reader.html?utm_source=openai))  \n- Live content updates and responsive layouts often easier to adapt in HTML  \n\n---\n\n**Slide 12: Comparison Table (HTML vs Tagged PDF vs Untagged PDF vs PPT)**  \n\n| Feature | HTML | Tagged PDF | Untagged PDF | PowerPoint |\n|---|---|---|---|---|\n| Headings navigation | Strong support | Strong, if tags correct | Very weak / none | Titles act like headings, if titles consistent |\n| Reading order reliability | High | High (if tags correct) | Low | Variable; must adjust reading order pane |\n| Alt text for images | Built in HTML image alt or aria | Alt tags inside figure tags | Rarely present | Alt text or alt dropdown; decorative option |\n| Tables properly associated | <table>, <th>, etc. | Tagged table structure | Linear text, headers lost | Use table objects; screen reader reads table |\n| Live updates / dynamic content | Excellent with ARIA | No dynamic content unless embedded | N/A | Can embed, but less responsive |\n\n---\n\n**Slide 13: How to Test Using Screen Readers**  \n- Tools to try: JAWS, NVDA, VoiceOver (built-in on macOS/iOS), Narrator on Windows ([support.microsoft.com](https://support.microsoft.com/en-us/office/use-a-screen-reader-to-show-your-presentation-with-powerpoint-7d9983fd-753d-474e-8214-c760e0bf75f2?utm_source=openai))  \n- Test workflows:\n  \u2022 Use keyboard only: tabbing, arrow keys, skipping headings  \n  \u2022 Try to navigate via headings, lists, tables  \n  \u2022 Turn off visual view; use outline or reading order pane in PowerPoint  \n  \u2022 Export PowerPoint to PDF with tags option and test in PDF reader  \n\n---\n\n**Slide 14: Tips for Creators to Improve Accessibility**  \n- In PDFs: ensure \u201cDocument Structure Tags for Accessibility\u201d option is checked when exporting; check and correct tags manually ([it.tamus.edu](https://it.tamus.edu/accessibility/accessibility-guides/powerpoint/?utm_source=openai))  \n- In PowerPoint: use built-in layouts; assign titles; use Reading Order Pane; alt text; mark decorative content ([mass.gov](https://www.mass.gov/info-details/powerpoint-accessibility?utm_source=openai))  \n- Always define logical reading order (z-order) that matches the visual flow  \n\n---\n\n**Slide 15: Summary \u2013 Key Takeaways**  \n- Tagged PDFs provide structure; untagged PDFs are very challenging for screen reader users  \n- PowerPoint can be accessible, but only if best practices are followed  \n- HTML offers inherent advantages for navigation and semantics  \n- Testing with screen readers is essential \u2014 you\u2019ll uncover barriers only when trying  \n\n---\n\n**Slide 16: Questions & Discussion**  \n- Invite team to brainstorm recent documents: which were tagged or not?  \n- Discuss any accessibility problems they\u2019ve heard from users  \n- Consider whether the team\u2019s current document templates require improvements  \n\n---\n\nIf you want, I can give you slide designs or speaker\u2010notes (what to say under each slide) too."
    }
]