iiinova — where your digital self keeps evolving

Ohhh I didn't see the find more extensions button! I'll use that again. When I configured AGY to launch /Applications/Polypane.app/Contents/MacOS/polypane it launched Chrome instead. Probably Electron incompatibility nbd ✌🏼

100

Google Antigravity - Build the new way

AGY abstracts away the MCP. It installs a Chrome extension and allows the Chromium executable to be configured. Just an idea in case enabling the extension is easy! antigravity.google/docs/chrome-...

antigravity.google

Google Antigravity

100

And thanks for the link! Just switched from an abandoned alternative developer browser 🧠

Antigravity would require the Chrome extension. I haven't checked if electron browsers work in general yet. Re. MCP I meant I wonder if a separate Polypane MCP would be magically smarter at testing responsive or something (maybe not)

200

@polypane.app Feature idea 💡 Antigravity's browser is configurable. It'd be cool if it could launch Polypane! I wonder if its own MCP might help too

Texas Toland @texastoland.com · 19/11/2025

Now would be the most baller timing to announce this 😁

000

Texas Toland @texastoland.com · 13/09/2025

Gemini personally but I understand if you do a more popular 1 first!

Texas Toland @texastoland.com · 12/09/2025

This sounds like a perfect match 😍 Any plans to provide other agents like Gemini CLI through ACP?

100

Texas Toland @texastoland.com · 26/08/2025

I feel guilty thinking it but good AI art looks as good as human to me youtu.be/Pwy5T6OPGzA?...

youtu.be

Kling's Frames Are Insane & MORE Nano Banana!

YouTube video by Theoretically Media

020

Texas Toland @texastoland.com · 08/08/2025

🥉 LG Exaone 32B kept trying to run a failing Python script without fixing it 🥈 Phi-4 Reasoning and DeepSeek Qwen 8B both rattled verbose academic thought chains and failed at tool calling 🥇 Qwen3 30B (2507 non-reasoning) alone gave succinct correct output identical to my daily driver Gemini Flash!

000

Texas Toland @texastoland.com · 08/08/2025

I couldn't get "OpenGPT" (4-bit quantized) to respond sanely. It was chatty, broke Markdown formatting, and often ended with long chains of gibberish tokens. I used a basic Fibonacci prompt with a custom code execution tool. I tried the following SOTA models instead (spoiler: Qwen3 won) 🧵

Texas Toland @texastoland.com · 07/08/2025

gpt-oss-20b is MoE! Only needs 14 GB unified memory. Trying soon 👀

100

Texas Toland @texastoland.com · 05/08/2025

At least once a day @raycast.com AI integration/MCP does something for me that would've been a chore 🪄

Texas Toland @texastoland.com · 23/07/2025

I'm so late to the party but dev containers and especially IDE support like @webstorm.jetbrains.com is fire ❤️‍🔥

000

Texas Toland @texastoland.com · 05/05/2025

I 🤬 at Gemini so it did it back with emphasis 🤣

000

Texas Toland @texastoland.com · 17/04/2025

This. I avoid politics because I'm anti-🇺🇸 in general. But fuck me if Disney hasn't been too on the nose. Like "news next month" level 💔

000

Texas Toland @texastoland.com · 23/03/2025

With the Augment Code on @jetbrains.com and Junie on the horizon I'm in the process of switching to @webstorm.jetbrains.com. But the biggest surprise of the week was Fleet ❤️‍🔥 I've seen so many bad reviews but it's everything I ever wanted from Code! If VS gets on your nerves it's worth a try

Texas Toland @texastoland.com · 23/03/2025

A week later I still love dropping into Codespaces for random shit like fixes to Obsidian plugins. But I'm surprised how often I lose my terminal connection inside. It also doesn't warn me before accidentally closing my broswer tab vs editor pane

000

Texas Toland @texastoland.com · 14/03/2025

I know it's stupid but sometimes I don't PR just because I don't want to clone etc. Lately I've been opening a repo into GitHub.dev to make easy fixes then jumping straight into Codespace when necessary. And @github.com has made the flow between sites completely seamless. It's 🔥🔥🔥

000

Texas Toland @texastoland.com · 12/03/2025

Now I completely switched from Notion I love it 🥰

000

Texas Toland @texastoland.com · 28/02/2025

Using it all day and sooo fucking good ✨

000

Texas Toland @texastoland.com · 28/02/2025

I've had a challenging past year. Just merged my first (tiny) PR in months. Progress 💪🏼

000

Texas Toland @texastoland.com · 24/02/2025

Just discovered Make.md completely replaces Notion in @obsidian.md while supporting SSG with BrainDB 😍 youtu.be/ISpII6nsego?...

I only tried briefly around the time you posted. I liked it but mostly got frustrated with key bindings between panes. It improved a little since then. Did you end up on Zen then? I switched back to Arc until it's unusable

100

I used it for a year or so but not for me 🥲

100

Not dead but very quiet 🥲 x.com/texastoland/...

x.com

I personally loved SigmaOS. Just concerned because of inactivity basically Arc--. Zen looking solid lately though!

220

Texas Toland @texastoland.com · 18/02/2025

Confirmed but Sigma piggybacked (not a bad thing) on many of Arc's features (esp. re. AI)

000

Excellent analogy 👏🏼 It isn't that I don't care. There's just nothing I can do about the kernel starving my process. Doing the best with my few allocated resources

Texas Toland @texastoland.com · 18/02/2025

I'm torn whether to follow fellow technicians I probably agree with politically but want less politics in my feed 🥲

100

Texas Toland @texastoland.com · 16/02/2025

Correction: The original match between AlphaZero vs Stockfish on unequal hardware was 28W-72D-0L. The equalized rematch ended… 155-839-6 🤯 And these games look utterly alien! youtu.be/8dT6CR9_6l4?...

youtu.be

How AlphaZero Completely CRUSHED Stockfish

YouTube video by GothamChess

000

Texas Toland @texastoland.com · 15/02/2025

I got frustrated with the pricing and switching to LobeChat! It looks like it resolves some UX quirks too. I'll report back soon ✌🏼

Texas Toland @texastoland.com · 15/02/2025

2019 - OpenAI Five wins a Dota 2 tournament and 99.4%of games in a public event 2019 - DeepMind AlphaStar achieves Grandmaster level in StarCraft II above 99.8% of ranked humans 2019 - DeepMind beats 80% of "strong" players with a handicap to simulate human reaction time