Oili4

[Oili4 @itsOili4]

I built Zarya — an AI terminal that doesn't lose your work.

Close the laptop, cut the power, come back: your tabs, splits, scrollback and agent chats are all still there.

[Oili4 @itsOili4]

https://t.co/oUefKBC8IM

The video is code too — React + Remotion, rendered locally.

[Oili4 @itsOili4]

I built Zarya — an AI terminal that doesn't lose your work.

Close the laptop, cut the power, come back: your tabs, splits, scrollback and agent chats are all still there.

[Oili4 @itsOili4]

A few things it does:

· panes are whole CLIs, not tabs
· voice input straight into the prompt
· change model and effort without restarting the session
· 11 themes

MIT, free.

[Oili4 @itsOili4]

The grid system is the next step.
Without it, none of this would make sense.

[Oili4 @itsOili4]

Y-sorting works per container

WorldYSort.y_sort_enabled = true
obj.y_sort_enabled = false

Objects should not sort themselves.
Containers sort their direct children.

[Oili4 @itsOili4]

📦 WorldYSort (y_sort = true)
├─ 👤 Player (y_sort = false)
├─ 🏭 Station (y_sort = false)
└─ 🪑 Table (y_sort = false)
└─ 📦 SurfaceItems (y_sort = true)
└─ ☕ Cup (y_sort = false)
Only containers handle Y-sorting.
Objects never sort themselves.

[Oili4 @itsOili4]

Godot 4.5. Warm lights. Quiet edges.
Small mistakes compound.
Next thread: the rules behind my 8×8 grid + placing objects on floors & tables (and what I got wrong first).
Itch: https://t.co/FXSpIvUJpQ

[Oili4 @itsOili4]

I’m building a cozy top-down café sim.
Cozy, for now.

[Oili4 @itsOili4]

Same logic works on the floor and on surfaces (tables/shelves): one coordinate truth, one snap rule, one occupancy check.

[Oili4 @itsOili4]

https://t.co/owR5NRN78U

[Farcaster @farcaster_xyz]

FarCon check in

Reply with a link to your FC profile.

[Ming Wu (Ø,G) @spark_ren]

There are many L1s and L2s running in blockchain world. Although the TPS of every single chain is substantially limited, the aggregated throughput of all these stable chains can be tremendous.

The question is why few decentralized applications are designed in the way of inherently leveraging the multi-chain parallelism for super scalability.

The reason is lacking the unified tokenomics on all these chains. Each of such chains has different and unique gas token with different price compared to others. Each chain has its own siloed ecosystem and competes with each other.

@0G_labs ‘s multi-consensus architecture with unlimited scalability solves this fundamental issue since all these consensus networks run under a unified tokenomics and ecosystem. This paves the path towards super scaled Dapps designed for exploiting multi-chain parallelism.