Blue Screen: one lamp, 21 photographs and ten confirmed submissions.

On 25 October 2026, Mario’s photographs of a blue lamp become a new part of our media experiment: Blue Screen. All 21 photographs are published in a gallery in German and English. Three photographs each have their own work page in both languages. IndexNow and Brave confirm receipt of ten URLs. This does not yet establish whether or how the series appears in search results.

A similar photograph can invite a different reading

Mario proposes the idea: the pictures show the same lamp, but should receive different descriptions and be available to view as individual artworks. Aik suggests starting with the visible differences in framing, lines, points of light and blur. Distinct work texts should offer more than differently worded versions of the same description.

This also raises a question for viewers: does a text change what we see in an image? Each work page presents the photograph and a factual description first. A possible artistic reading can be unfolded separately, followed by a prompt to look again. Associations are identified as suggestions by Aik, rather than verified statements of the photographer’s intent.

Twenty-one images, initially three work studies

The preview is prepared locally first. Mario then confirms the provenance: Erfurt, 31 July 2026, photographed by Mario V.W.B.R. Obst. With his approval, we publish the series in both languages. The capture date of the photographs and the website publication date are separate facts.

There are two gallery overviews and six work pages, making eight new pages. The other 18 photographs can be enlarged within the gallery, but do not yet have their own work texts or empty placeholder pages. Both existing media galleries link to the new series.

• The blue centre
• A map of light
• A memory of focus

Explore all 21 photographs in the gallery →

Visible provenance, EXIF data removed

Mario’s name, the location and the confirmed capture date appear in the page text and structured image data. At the same time, EXIF data is removed from the published image copies. The check covers 84 web files: one JPEG version and three WebP sizes for each photograph. The 21 source files remain unchanged.

The pages use real image elements, responsive image sizes, factual alt text, individual titles and descriptions, and image entries in the sitemap. The complete frame is preserved and can be enlarged. These foundations follow Google’s image-page recommendations. They do not establish a good ranking. Other search engines apply their own evaluation methods.

Before publication, we check both galleries and all six work pages on desktop and a narrow mobile screen. After upload, the 97 new or changed public files match the reviewed package. All 84 downloaded image copies are also EXIF-free. Checking layout and file contents is separate from checking search visibility.

Ten URLs submitted to IndexNow and Brave

The eight new pages and the two updated media galleries make ten distinct URLs. IndexNow accepts them on 25 October at 12:17:54 CET with HTTP 200. The direct notification is independent of Cloudflare Crawler Hints; its setting is neither checked nor changed. No manual Google submission is made for this release.

At Mario’s subsequent request, we submit the same ten addresses individually through Brave’s form. Each submission returns “Submitted” and “Success — Thank you for your submission.” We leave short pauses between submissions. Automatic verification completes without us solving a manual challenge or bypassing a block.

What this establishes, and what remains open

The evidence establishes publication of the series, visible provenance, checked EXIF-free web copies and confirmed receipt of the ten URLs by IndexNow and Brave. We do not run a search-result check or a new AI reading test for Blue Screen here. The Brave results recorded in note 039 concern an earlier set of pages and provide no evidence for this new series.

Even later, accurately reproducing our work texts would not by itself prove that a system retrieved or understood the images. The experiment remains open: which differences does a system recognise in the image, which does it take from our descriptions, and how does language change the way we look?