dschnurr

Nice! Something must be in the air – last week I built a very similar project using the historical archive of all-in podcast episodes: https://allin-predictions.pages.dev/

Hi folks – I work at OpenAI and helped build this page, awesome to see it on here! Heads up that it's a bit out of date as GPT4 has a different tokenizer than GPT3. I'd recommend checking out tiktoken (https://github.com/openai/tiktoken) or this other excellent app that a community member made (https://tiktokenizer.vercel.app)

I gave this a try: https://twitter.com/_dschnurr/status/1514673929209614337

OpenAI engineer here. Cloning this project and running it locally with your own API key does not violate any policies. However the way this project is configured, publishing it to the web would expose your API key in the client-side source code, which violates our policies since it would allow your account to be compromised.

Sure there will be middlemen trying to profit due to asymmetry of information / technical knowledge. But as technology marches along the barriers to entry of launching new things trends to zero. Smart creators are increasingly harnessing new technologies to cut out existing middlemen. Podcasts are a great example of this, anyone with a microphone can start an entertainment business now and collect money from direct relationships with advertisers.

The creator is almost always the artist. Provenance is what gives NFTs their value. The value of a Beeple NFT minted by a random user is $0.

Some artists choose to attach "perks" to their NFTs – i.e. meet and greets, access to private communities. They could choose not to honor these perks for NFTs held by contracts. They could even choose to re-mint the work if the new owner is doing nefarious things with the old one (though many would see this as a dubious action)

The people receiving royalties are the artists in these cases – instead of the record companies, instagrams, and etsys of the world. I fail to see how this won't be a net positive for wealth equality when we're shifting the advantage to those who have historically been taken advantage of.

Gov-supported digital currencies would mostly streamline things and replace existing bloated and slow programs. Imagine the current stimulus situation – instead of sending checks to people they could just press a button and immediately send credit to all citizens that have a certain account threshold or income rate. Currency units could be earmarked for certain retailers (i.e. food stamps, education, etc).

This is cool! What tech/service are you using to handle accepting payments & paying out to event creators? To my knowledge you can't use stripe for things like this (though maybe I'm wrong)

going public at any point after nov '19 probably would have been a disaster for most employees due to lockup period. Let's say I have $1M of RSUs at IPO – fed+state gov wants 450-500k for taxes total. When I went through Uber IPO they only withheld ~30%, which leads to a $200k tax bill at the end of the year in this scenario. If stock declines 75% due to COVID before I can sell at expiration of lockup period, I'm left with $175k ($700k post-withholding * 25%) and $200k tax bill. Not sure about you but I'd prefer the non-IPO scenario.

> "human genome" sized project to create an accurate physics engine and a chemistry engine

bingo. this would insanely impactful long term.

If they end up using Rome internally that should be a good sign that it will be supported for years to come. React has been very well maintained for the past 5+ years.

Starting a new JS project these days can be a huge hurdle for many users because they have to manually set up all the compilation/linting/testing/formatting etc infrastructure, each of which have multiple possible libraries, versions, and non-trivial configurations. You could spend a week configuring webpack alone to get code splitting, css compilation, etc to work. Projects like create-react-app and next.js have helped remove some of this fatigue, and you can see by their adoption just how much of an impact this has. Seems like Rome should be similarly beneficial without being inherently tied to an application framework like create-react-app or next.js – the challenge will be avoiding a confusing mess of 1000 configuration options to fine-tune each subsystem.

I'm building a no-code web app development platform. There are numerous existing tools in this space, but I don't think anyone has really hit the sweet spot of having the same composability/reusability/conditional benefits of code in a WYSIWYG interface that drastically increases development speed.

Also happen to be looking for a cofounder as well. I'm a developer myself but I'd love to partner with another product-minded dev or a product person with significant industry experience. Let me know if you're interested.

Naval Ravikant on Joe Rogan. Probably the highest wisdom per minute piece of media I've ever consumed.

One that I heard about last year and have been following – https://www.smallrobotcompany.com/

Hopefully nobody uses an article like this to really inform their decision on which to use.

Last month there was a reddit discussion that outlined some really compelling accessibility-related reasons to use tabs, worth a read: https://www.reddit.com/r/javascript/comments/c8drjo/nobody_t...

Reasons to prefer tabs: https://www.reddit.com/r/javascript/comments/c8drjo/nobody_t...

Good tips. How are you killing remaining zombie Chrome processes? Just polling the processlist and running "kill" commands?

I'm using dumb-init for the main node process in my docker container (node in-turn spawns chrome instances via puppeteer as requests come in). Is there a way to use dumb-init for each individual chrome instance as well? Or are you suggesting we make chrome the main container process and start new containers for each session?

Hey Joel–awesome service and post. Your tip about parallelizing via browser instances instead of tabs resonated with me as I've had a lot of headaches with one bad request taking down a full instance. That said, it does seem like the constant launch/close of puppeteer browser instances does have some sort of memory leak – Here's what happened when I switched from parallelizing via pages to browsers: https://imgur.com/a/rCz3oow

Have you seen any similar behavior in your system? I initially thought I was forgetting to close the browser instance but triple checked my code and it does seem like I'm calling close() everywhere.

No – the services will likely own the cars (much like Waymo owns these chrysler pacificas).