i’ve had a huge amount of trauma in my life and i find myself using chat gpt as kind of a cheater coach thing where i know i’m feeling a certain way, i know it’s irrational, and i don’t really need to reflect on why it’s happening or how i can fix it, and i think for that it’s perfect.

a lot of people use therapists as sounding boards, which actually isn’t the best use of therapy imo.

Probably comes down to what issue people have. For example, if you have anxiety or/and OCD, having a „therapist“ always at your disposal is more likely to be damaging than beneficial. Especially considering how basically all models easily tip over and confirm anything you throw at them

Your use-case is very different from someone selling you ChatGPT as a therapist and/or telling you that it's a substitute for other interventions

What's a "cheater coach thing"?

A/ i love this distinction.

B/ my brother used to use "fetcher" as a non-swear for "fucker"

Did you tell him to stop trying to make fetcher happen?

Very funny. Now let's hear Paul Allen's joke.

He picked up that habit in Balmora.

Fetcher? Damn near killed'er!

wow thanks for the heads up! no idea this was a thing.

It’s not a thing.

i don’t understand. is the serialized write global lock a thing or no?

If you’re going to use a tool. Make an attempt to use it properly. If you do dumb things. Dumb things will happen. Example. This blog post. This is not a problem for everyone using LISTEN/NOTIFY. It’s a problem with not knowing how to use it, then spreading bad info.

i knew anyone who loves ocaml that much couldn’t be trusted.

this is the most non sequitur press statement ever.

you are a saint.

<3

Do we think that cyborg coding is the right word? I've been struggling how to explain when i have 4 claude code instances going working on stuff from specs / fixing bugs / updating docs / etc.

Note that Claude overuses tokens. I guess they get paid by the token. It is cliched in this way, sort of like if a programmer were to get paid by the line. Other LLMs can produce clean results with far fewer tokens than Claude.

I believe we already have the correct word (agent).

Agentic coding.

To me this implies that it's more agent than person... like when you use claude in github actions vs using it in your ide.

Agentic coding works via the IDE, GitHub Issues, and command-line tools like OpenAI Codex and Claude CLI. Agentic coding is AI-based feedback-driven multi-step coding. In general, an agent implies that the agent is working for someone, which in this case is a person.

I'm into it, i have a quest pro (which i think is still higher res than Q3 but idk if it's powerful enough) and would be very interested in trying it out.

Not enough gaussian splatting research imo.

It’s powerful enough for this application! Mostly because here, faces have way less vertices than the gaussian splat scenes that you see in research papers.

Totally agree that there’s not enough research, and honestly not enough people in VR making applications on top of it

holy shit so cool!

thank you!

that’s the next step… we have a structured approach to prompting too that we think will help people build better prompts too.