Canvas v0.1 — the local canvas computer launches

Why this product exists

Most visual tools ask people to accept the wrong tradeoff too early. Before the work is real, you are asked to create an account, choose a plan, think about collaboration, or bend your idea into a template. Canvas starts from the opposite direction.

The first job is to make composition feel immediate. Everything else should come later.

Desktop first, for a reason

Canvas is meant to feel like a small creative machine on your Mac: pan, zoom, drag, resize, arrange, revise, export. The local workspace is not a fallback mode. It is the product.

That means persistence should be restart-safe, export should be native, and the interface should behave like a real desktop tool instead of a browser demo.

TARX is underneath, not in front

TARX matters because it powers the local AI layer. It should make the product feel faster, more private, and less metered. But the user should still feel like they are using Canvas, not managing infrastructure.

That is the balance we want to keep: Canvas is the product; TARX is the reason the product can do more locally.

Publishing comes later

v0.1 is not built around a storefront integration story. Publishing targets matter, but only after a layout deserves to live somewhere. The local composition loop has to stand on its own first.

That is why export matters from day one and why optional integrations should enter later, not as the first-run promise.

What we care about next

We want the product to feel more trustworthy, more fluid, and more obviously local. Better starters, cleaner AI critique, stronger history, and smoother publishing routes all matter. But they only matter if the local surface already feels like home.

That is the v0.1 line in the sand: the local canvas computer ships first.