Testing a visual language built from small signals: time, temperature, and weather. Familiar, useful, and quietly computational.
📓
Most TV interfaces treat browsing like a flat shelf: move left, move right, repeat.
A spatial gallery suggests another possibility. What if shows lived in a navigable field, with depth helping communicate proximity, category, or relevance? The interesting part is not the 3D effect. It is whether a remote could move through that field predictably without losing orientation, focus, or a clear way back.
I am not ready to build it yet. I want to leave the question here so I can return later and ask whether spatial browsing can make television discovery feel more alive without making it harder to use.
Framer Community 
I collaborate with Israel to define problems, test hypotheses, coordinate agents, and render behavior to any screen.
Spotify knows what I played. Netflix knows what I watched. Neither necessarily knows why I chose it—or what I want tonight.
That is why recommendation is hard: behavior is evidence, not intent. A click can mean curiosity, habit, compromise, background noise, or regret. Then systems train on their own suggestions, turning yesterday’s guesses into tomorrow’s profile.
AI can help, but the interface cannot become an interview. Every question spends effort and raises the expectation that the result will improve. If it still misses, the system may lose more trust than if it had made a low-confidence suggestion first. The better shift is selective conversation: ask only when one answer will materially change the slate; otherwise show a diverse starting point, expose the assumptions, and make correction effortless.
Good recommendation may be less prediction and more shared discovery.
trying to make data honest. bare, clear and transparent.
I saw this Apple TV menu and kept thinking about why it feels so clear.
The interesting part isn’t only that the modal is small. It appears directly over the selected show while the surrounding row stays visible. That proximity explains ownership: these actions belong to this item, not the whole screen. The viewer doesn’t have to remember what they selected or reconstruct where they came from.
A modal can still feel disruptive at any size. Keeping it close to the object that triggered it makes the interruption feel local. The remaining test is whether Back dismisses it and restores focus to the same show.
I saw this sponsored video layout on TV and kept thinking about why it worked.
The video stays visible on the left while the product becomes a browsable surface on the right. That separation preserves the story that earned attention, then gives the viewer a clear place to explore without turning the entire screen into a landing page.
It feels less like an interruption and more like a temporary mode. The layout is smart. The real test is whether remote focus, product selection, and the return path are just as clear.
Google just added “Created or edited with AI” disclosures to ads across Search, YouTube, and Discover. But the disclosure lives inside the three-dot My Ad Center panel.
That is technically available, but UX is not only about whether information exists. It is about whether people can see it at the moment it changes their judgment. With EU transparency rules taking effect August 2, AI labels are becoming a real interface pattern. The best version should be visible, understandable, and proportional—not buried, alarming, or vague.
The new ChatGPT desktop UI makes it unnecessarily difficult to find regular chats. Chat, Code, and Agents feel like separate products occupying the same app, but the navigation doesn’t explain that model.
I’d make the modes explicit and keep Projects and Recent conversations visible:
💬 Chat | 💻 Code | 🤖 Agents Search Projects • Atlas • Identity Ledger • Home Power Recent • Finding ChatGPT Chats • Harvester Research • ...
One app, three clear modes—without hiding conversation history or making people troubleshoot where their chats went.
I keep thinking about what happens to design systems when the user is not only a person clicking through an interface, but also an agent trying to understand what it is allowed to do.
Traditional design systems gave teams reusable visual and interaction pieces. Buttons, cards, spacing, typography, states, components. But agents need a different layer of consistency: commands, source boundaries, privacy rules, evidence requirements, tool permissions, output contracts, and stop conditions.
That feels like the next version of a design system to me. Not replacing the visual system, but sitting beside it. A product should be legible to people and operable by agents.
arXiv.org