Decorative Images… - CourtneyR.dev's 2nd Brain

Decorative Images… ![rw-book-cover](https://adrianroselli.com/wp-content/uploads/2026/10/decorative-300x300.png) 🔗 **Source:** [Decorative Images…](https://adrianroselli.com/2026/10/decorative-images.html) > [!note]+ Document Notes > tags: [accessibility, alt-text, decorative-images, wcag, web-content] > author: Adrian Roselli > source: adrianroselli.com > reference: https://adrianroselli.com/2026/10/decorative-images.html > date created: 2026-10-19 15:30 > >  > > > [!info] Generated analysis — not the author's words. Sections below are machine-written > > from the captured source. Only text inside `> [!quote]` blocks is verbatim from the author. > > Do not quote or attribute any other line to the named author. > > # 🏷️ Title: Decorative Images… >  ## 📝 Summary - **What it is.** A short opinion post by Adrian Roselli on adrianroselli.com, numbered here by blank-line splits of the captured text (¶1 to ¶19, headings and list items each count). The capture starts mid-title: ¶1 reads as the end of a sentence about decorative images, and the page title is truncated to "Decorative Images…" in the metadata given for this task. - **The argument.** ¶2 calls the position that decorative images don't exist an absolutist take, adopted on purpose to make an author justify each image. ¶3 limits scope to images in narrative content that are neither iconography nor part of interactive controls. ¶4 says an image added for mood or theme gives a person who can't see it something worth knowing. ¶5 says an image added to fill white space points to a layout gap that should be fixed in layout. - **The method.** Two over-simplified tests (¶7 to ¶10): would any reader care if the image were removed, and would the author care if it were swapped for something unwanted. A yes to either means the image isn't decorative. ¶11 to ¶12 introduce a decision tree adapted from the WAI alt text decision tree with more steps. - **What the capture does not include.** The decision tree itself (no image, alt text or description appears in the captured text), any alt text guidance beyond the two tests, and any WCAG success criterion. The text names none, so none is named here. ¶18 links a WCAG 3 issue about the definition of decorative images, which is the only standards reference. - **Bottom line.** Advocacy for treating "decorative" as a conclusion reached by questioning, not a default. ¶13 to ¶19 are a link list of seven related posts and a W3C issue, all with live anchors. ## 📖 Main content - **¶1 to ¶3, framing.** The post opens by saying decorative images are not a thing as far as people are concerned, concedes the claim is absolutist, and scopes it to narrative images. - **¶4 to ¶5, two reasons images get added.** Mood or theme (the non-visual reader benefits from knowing it) and filling space (a design gap that sends an unintended message to sighted readers if left). - **¶6 to ¶10, the two tests.** Removal test and substitution test. Both end in the same rule: if nobody would care, remove the image; if someone would, it isn't decorative. - **¶11 to ¶12, decision tree.** Heading plus one sentence; the tree is the WAI one, extended. - **¶13 to ¶19, related reading.** Eric Bailey (17 June 2021), three posts by Roselli (13 August 2023, 20 April 2024, 28 May 2024), Nicolas Steenhout (28 September 2026), WCAG 3 issue #878 (29 September 2026), Jens Oliver Meiert (8 October 2026). ## 📚 References All seven entries in the capture's link list match a ¶13 to ¶19 anchor, plus the WAI link in ¶12. Primary sources: - WAI alt text decision tree: https://www.w3.org/WAI/tutorials/images/decision-tree/ - Your Image Is Probably Not Decorative: https://www.smashingmagazine.com/2021/06/img-alt-attribute-alternate-description-decorative/ - An alt Decision Tree Using Only :has(): https://adrianroselli.com/2023/08/an-alt-decision-tree-using-only-has.html - Long Alt: https://adrianroselli.com/2024/04/long-alt.html - My Approach to Alt Text: https://adrianroselli.com/2024/05/my-approach-to-alt-text.html - It's time to retire the concept of decorative images: https://nicolas-steenhout.com/retire-decorative-images/ - WCAG 3 issue #878: https://github.com/w3c/wcag3/issues/878 - On the Decision Underlying Decorative Images: https://meiert.com/blog/decorative-images/ The two tests in ¶7 to ¶10 and the claims in ¶4 to ¶5 have no link of their own in the captured markup. ## ✅ Action items - **Anyone writing or reviewing alt text.** Run the two tests in ¶7 to ¶10 before marking an image as decorative. - **Designers.** Treat an image added only to fill space as a layout problem (¶5). - **DevRel, OSPO, Community, The WPOCC.** No direct action from this source. ## ❓ Questions - What do the extra steps in the adapted decision tree (¶12) change compared with the WAI original? - How does the WCAG 3 issue linked in ¶18 define decorative images, and does it match the position in ¶2? - Where does the post draw the line between a decorative image and one that needs only a short alt (the capture is silent)? ## ✅ Logical validity #### Strong reasoning - **¶2 states its own method.** The post labels the stance absolutist and says why: to make the author justify the choice. That separates a rhetorical device from a factual claim. - **¶7 to ¶10 give checkable tests.** Each ends in an action (remove, or treat as informative). #### Fallacies - **Overstated headline claim (¶1 to ¶2).** The claim that decorative images aren't a thing applies "as far as people are concerned," but ¶3 then excludes iconography and controls. Fix: state the scope in the first sentence. - **Unsupported assumption (¶9 to ¶10).** The substitution test assumes author indifference means the image has no value to a reader. No source is cited for that link in the captured text. - **Missing standards mapping.** The post gives no WCAG success criterion for when an image may carry empty alt. Fix: cite the criterion or W3C technique the tree implements. #### Tone of voice Direct and instructional, with one openly provocative stance in ¶2. ## 🎯 Conclusion A compact argument that "decorative" should be the result of two questions, not a starting label. The captured text carries the reasoning and a reading list but not the decision tree, so use the linked WAI tree and the author's earlier posts for the actual alt text rules. ## 🔄 Refactor Appendix — Decorative Images… > Density: light-touch note on a short opinion post; three literature atoms, no permanent candidates. ### 📚 Literature Split Pad #### [H1] Scope is narrative images, not icons or controls > [!quote] [¶3] > Broadly, I’m talking about images in narrative content that are not also iconography nor used as part of interactive controls. **Context & use**: - ¶3. Sets the boundary for every later claim in the post. **Tags**: #xref  #### [H2] A mood image helps a person who can't see it > [!quote] [¶4] > Maybe the author wants to convey a mood or connect it to a theme. **Context & use**: - ¶4. The reason the post gives for describing images that look purely atmospheric. **Tags**: #xref  #### [H3] A space-filling image is a design gap > [!quote] [¶5] > Maybe the author wants to fill white space. In which case, that’s a design gap. **Context & use**: - ¶5. Moves the fix from alt text to layout. **Tags**: #xref #share  ### 🧠 Atomic Split Pad No permanent candidates: the post is a short argument whose rules live in the linked decision trees. ### ✏️ Fleeting Queue - F1: Read the WAI decision tree next to the adapted version and note which steps were added. ### → Things 3 (max 2 actions) - [ ] Compare Roselli's adapted decision tree with the WAI original before using it in alt text guidance — accessibility #things  📄 **Full text:** [[Full Document Contents/Articles/Decorative Images….md|Decorative Images… by Adrian Roselli]] ## Highlights - Broadly, I’m talking about images in narrative content that are not also iconography nor used as part of interactive controls. ([View Highlight](https://read.readwise.io/read/01m5atjxfbrp3tng4gpq56e45s)) - Note: [H1] Scope is narrative images, not icons or controls - Tags: atom/literature/h1 - Adding an image to a design, email, document, page, app, etc. is done for a reason. Maybe the author wants to convey a mood or connect it to a theme. In which case, the person who can’t see the image would likely benefit from knowing that mood, having that connection made. ([View Highlight](https://read.readwise.io/read/01m5atjzp5z3dz34ax3j0x0nyz)) - Note: [H2] A mood image helps a person who can't see it - Tags: atom/literature/h2 - Sometimes the reason is far more pedestrian. Maybe the author wants to fill white space. In which case, that’s a design gap. It should be addressed in layout, otherwise the author is unintentionally sending a message to those who can see it. ([View Highlight](https://read.readwise.io/read/01m5atk20rzfz5nmpc0q2tesjt)) - Note: [H3] A space-filling image is a design gap - Tags: atom/literature/h3