Once Upon a Socket" 📖✨ A beautifully illustrated love story between Client and Server, from awkward HTTP polling to happily-ever-after with WebSockets. By the author of https://t.co/6Qamd4PX7j (yes, really, me 😜) https://t.co/NAyYbcHbwe #WebSockets #RealtimeWeb

@JoshuaWohle @usemonologue @ablyrealtime @WisprFlow @tankots If you're a founder/builder who hasn't tried voice-first: Try @usemonologue for a day. As your PRIMARY input. Report back. (Changed my workflow in 48hrs. Never looked back)

@JoshuaWohle @usemonologue @ablyrealtime Questions: Killer tips? Local or cloud? (I chose cloud - convenience > privacy) Tried @WisprFlow? (@tankots raised $30M) Want to compare workflows 👇

@JoshuaWohle @usemonologue @ablyrealtime @JoshuaWohle's key insight: decouple transcription from the tool Monologue: ✅ Understands context ✅ Corrects rambles ✅ Remembers vocabulary You don't need to be precious about how you speak

@JoshuaWohle @usemonologue @ablyrealtime Real talk: still awkward in public (but I do it anyway) Yesterday: 8hr flight, decided to type It felt SLOW That's when I knew it was permanent

@JoshuaWohle @usemonologue @ablyrealtime Everything transformed: Coding (Ghostty, iTerm2) AI work (Claude) Slack Presentations Work on 3 things at once now. Talk to each, let them run, move on.

@JoshuaWohle @usemonologue Killer use case: doc review at @ablyrealtime 20-page decision records: 1+ hours → 5 minutes Read + dictate feedback simultaneously Game changer.

@JoshuaWohle @usemonologue Speed is obvious (3x faster) Real advantage: you think clearer when speaking Monologue corrects rambles, understands intention, outputs clean instructions No formatting overhead. Just talk.

Coding (@ghostty, @zeddotdev ) AI work (Claude, ChatGPT) Slack Presentations Work on 3 things at once now. Talk to each, let them run, move on.

2 weeks ago @JoshuaWohle told me: "move to voice" Tried @usemonologue afternoon 15 min later: converted 🤫 Now rank ⬆️#7 globally. 150k words. 35 hours saved. I HATE phone calls. But this is different. What changed: 🧵

I am really disliking the new UI in OS 26 Tahoe. The "thinking" behind the UX is fundamentally broken. @Apple is very sadly losing touch. This overhaul is a step backwards, and the liquid effect is a gimmick that masquerades the lack of innovation and refinement.

I’m at #SportsProAI today. Biggest theme so far? How to keep fans connected and engaged in realtime. Who's also here? DM me—let’s grab a coffee.

Our Head of Sales on this new role 👉

🚀 We're hiring AEs and SEs at Ably! Join the team powering realtime for billions of users worldwide. We handle 2 trillion+ API ops monthly, power brands you use daily, and our platform never sleeps (99.999% uptime for 5+ years). Ready to sell/architect the future of realtime?

Software development's biggest shift isn't AI writing code—it's the death of handoffs. When designers build in production and engineers design in code, the entire process collapses into continuous creation. The new skill: knowing when to ship & iterate. https://t.co/40xmEUrY8S

Loving Claude CLI @AnthropicAI, Cursor has been firmly relegated. But the persistent "Error during compaction: Conversation too long" is beyond frustrating because compaction IS needed because the conversation is too long 🙃

@ablyrealtime 6/ This isn't theoretical. It's running now. The future of fan engagement isn't just AI-powered insights. It's AI-powered insights delivered at the speed of live sports. What do you think? 👇

5/ Technical stack: - @ablyrealtime data streaming for instant updates - AI processing at the edge - WebSocket delivery to browsers - Scales to millions without breaking Code:

4/ The demo shows a better way: - Stream live match events - Process in realtime - Deliver insights instantly Watch any live match. See the AI keep up.

3/ Traditional approach: - Poll for updates ❌ - Process in batches ❌ - Deliver when ready ❌ Result: AI insights arrive after the moment has passed.