Heath Dingwell

In the creator economy, the creator *is* part of the content.

@antigravity Not sure what you did, but I can't open the workspace I've been using for six months.

Introducing Antigravity 2.0, a new standalone desktop application that delivers fully on that original glimpse of a truly agent-optimized experience. Rebuilt from the ground up with multi-agent teams, scheduled tasks, native voice and one-click integration with other Google products. Learn how to get started with Antigravity 2.0 👇

I was surprised. Tried it in ChatGPT and it was horrible. Had really low expectations with Codex, but wanted to try. It worked exceedingly well, but I fed it a lot of information about my process and output, had it ask me questions, then worked through the guidelines to follow. Took hours to accomplish, but I'm happy to have a second system. Hoping Gemini can do the same after this week's updates (prior attempts have been disasters)

@HeathDingwell thats surprising to me - i cant seem to get 5.5 to get my vibe enough to write like me

what’s the best model & harness that just gets writing vibe right? i don’t want any content generated, but sparring partner and for it to really get the style im going for so it can critique i mix up trying a lot of md files, examples and such

@thsottiaux @mattlam_ @mweinbach So in other words...you're asking Siri for help. 😉

@mattlam_ @mweinbach We don’t. It’s all just magic. We stare at the screens and pray.

Does anybody know what the root causes were or fixes? Just curious for learnings @mweinbach @thsottiaux

@thsottiaux A canvas feature like in Gemini so I can make edits to the file being built without switching apps.

For those of you living inside the codex app, what should we prioritize among features, reliability or performance?

@thsottiaux @kr0der Free pizza deliveries so we can work in Codex uninterrupted.

@kr0der What would you like to see in a > $1000 plan

who’s gonna pull the trigger on $500/$1000/$2000 AI subscription plans first? everyone who needs those plans is just buying multiple $200 plans anyway 👀

@emollick Discussing scrotum snorkling through a gravy orgie as a class topic is probably a fast track to having a one on one with the Dean.

GPT-5.5 Pro faces its hardest academic challenge: to apply the technique from a paper analyzing which word pairs were funny & why to come up with its own It came up with scrotum snorkel, tuba subpoena, waffle coffin, toad commode, diarrhea tiara, banana tribunal & muffin ruffian

@emollick Does this mean the Magic 8 ball was the wizard of the 1980s?

That explains this startup.

@OfficialLoganK @antigravity Will I be able to cook a pizza with it?

Very excited for all the stuff the @antigravity team has been cooking :)

New Smarter Society AI is out, and the question I kept coming back to writing it is when does information actually become useful? Most of what professionals need to make a good decision exists somewhere already. It's in the record, the file, the journal, the satellite feed. The hard part is getting it in front of them at the moment they need it, in a form they can act on. A probation officer has hours, not days. A doctor has minutes. An emergency manager has the time it takes a flash flood to form. Issue 19 looks at four AI tools that are trying to close that timing gap. Probation case files in California. Child welfare records in Illinois. Clinical decision support in US health systems. Flash flood forecasting in 150 countries. Different problems, same underlying question about whether AI can actually deliver information in time to matter. Here's the link to check it out: smartersocietyai.substack.com/p/in-time-to-m…

@emollick In January 2024, the Freakanomics Radio podcast did a 2 part series on misconduct regarding the research/publishing process. It'd be interesting to see a follow up at this point, focusing on how AI has impacted that.

Making humans responsible for their AI use seems like an incredibly reasonable way to address problems & opportunities in the use of AI for academic research, at least in the short term (autonomous scientific work will require different solutions).

@pixeluibygoogle Liquid glass? 😉

One of the most requested features is finally coming to Pixel phones. Which one do you think it is?

@acadictive I have to run my process through Codex a few more times to see if I get consistently good results. At that point, maybe Claude becomes my #2, but certainly not yet.

@HeathDingwell i am still team claude. :) lets connect, heath.

Night and day difference between the ChatGPT app and codex when doing knowledge work. Same model though. ChatGPT was a disaster trying to build out a newsletter issue. Codex followed my workflow almost perfectly. I may have found an alternative to Claude for my work.

@kepano @obsdmd You really should have spoken with the people in charge of the review process - kick 'em in the rear and get them moving quicker. 😉

The most exciting thing for me is that I can finally get back to making @obsdmd plugins. I too was daunted by waiting in the review queue 😅 Now you can get a plugin into Obsidian within 24 hours.

Expect your feed to look more and more like this in the coming weeks and months.

What will come after AI?

@kinginmotion @obsdmd I'm all for using agents with my vault and they help a lot. But I don't see how I can't heavily be in the loop if it's my knowledge base.

Still using a Karpathy-style @obsdmd vault? I tried going agent-only, but it felt like I was missing the real work... so I put myself back in the loop... Anyone else? Am I missing something?

@pixeluibygoogle @imparkerburton @Android iPhone?

What do you think might be inside it? 🤤 @imparkerburton @Android

Now that the Codex app is close to being the super app. What should the super duper app do?

Personally, I like using Claude (and now Codex) to help me write. Even if it's with a tool, the ability to bounce suggestions/ideas/revisions off of something really helps. There are plenty of times where Claude has made some good points and recommendations, and others where I've had to point out what (and why) suggestions don't work.

This is going to get even worse as people realize that careful tuning in their prompts can make AI writing seem not like AI writing to readers. We expect word counts to align, in some way, with thinking & value. Writing took effort. We are not mentally ready for the alternative.