Sidebar Concept

We had a concept. What if the command interface could turn into a sidebar that could be docked to your applications?

So we sketched...

And we sketched better...

But it turns out we were wrong.

The Sidebar Trap: Why Docking Would Have Killed Our Vision

The sidebar concept seemed logical at first, familiar, organized, predictable. But the more we explored it, the more we realized we were about to build another productivity prison.

Here's what would have gone wrong:

1. Screen Space Theft

Sidebars are space thieves. They push your actual work into a cramped corner, making everything smaller and harder to read. Your code editor becomes a narrow column. Your spreadsheets get squeezed. Your browser tabs fight for pixels. We would have been solving the app-switching problem by creating a screen-space problem.

2. The Docking Delusion

"Docking" sounds organized, but it's actually restrictive. Which app gets the sidebar? What happens when you switch between applications? Do you need multiple sidebars? The complexity multiplies instantly. We would have created a management nightmare disguised as a solution.

3. Fixed Position = Fixed Thinking

Sidebars force linear, top-to-bottom thinking. But AI assistance isn't linear, it's contextual, spatial, and dynamic. Sometimes you need help with something at the top-right of your screen, sometimes bottom-left. A sidebar locks you into one interaction pattern, killing the natural flow of work.

4. The Cognitive Load Mistake

When you dock something permanently, your brain starts treating it like furniture, always there, always demanding attention. This creates constant cognitive load. Your peripheral vision is always processing the sidebar, even when you don't need it. It becomes visual noise instead of helpful signal.

5. Breaking the Mental Model

Users already understand floating windows. Every app uses them, dialogs, notifications, tooltips. But docked sidebars? That's application-specific behavior. We would have forced users to learn a new interaction paradigm instead of leveraging existing mental models.

Why Our Current Approach Works

Spatial Freedom

Our floating interface goes wherever your attention is. Working on a design in the top-right corner? The AI floats there. Debugging code at the bottom? It follows. This isn't just convenience, it's cognitive alignment with how humans actually work.

On-Demand Presence

The most powerful feature isn't what our AI can do, it's what it can not do. It can disappear completely. No sidebar can truly disappear without breaking the interface. Our floating assistant exists only when summoned, then vanishes without a trace.

Universal Compatibility

A floating interface works with every application ever made and every application that will be made. Sidebars require cooperation from other developers. We built something that works everywhere, immediately, without asking permission.

Context Preservation

When you summon our floating assistant, it appears exactly where you need it without rearranging anything. Your work stays exactly as it was. No layout shifts, no window resizing, no visual disruption. The context remains perfect.

Natural Interaction Flow

Humans don't think in sidebars. We think in spatial relationships. "I need help with this thing right here" is how people actually request assistance. Our floating approach honors this natural instinct instead of forcing artificial constraints.

The Moment We Realized

The breakthrough came when we asked ourselves: "What if AI assistance felt like having a friend looking over your shoulder, helpful when needed, invisible when not?"

Sidebars feel like having a supervisor permanently watching your work.

Floating feels like having a genius friend who appears exactly when and where you need them, then politely steps away.

Why This Matters for the Future

We're not just building an AI assistant, we're defining how humans will interact with AI in their existing workflows.

The sidebar approach would have perpetuated the current problem: forcing users to adapt to tools instead of tools adapting to users. It would have been another interface to learn, another cognitive burden to carry, another constraint on natural work patterns.

Our floating approach does the opposite. It disappears into your existing workflow, appearing only when beneficial and vanishing when not needed. It works with your spatial thinking, your attention patterns, and your natural interaction instincts.

The Lesson

Sometimes the "obvious" solution is obviously wrong. Sidebars felt familiar because they're everywhere in software. But familiarity isn't the same as correctness.

The best interfaces don't announce themselves. They fade into the background and amplify human capability without demanding attention.

That's exactly what our floating AI assistant does, and exactly what a sidebar never could.