ifyouknewone

3.1 has made running the API so cheap that this doesn't really matter.

Who in their right might thinks it's a good idea to use something they pay a NAMED SUBCRIPTION FOR as a secondary engine in another tool?

Like, it's hilarious some of you guys think it's OC's fault for this.

It's open source software, with extensive documentation that anything you do with it being at your own risk.

It's no one's fault but the people plugging their oauth into this thing like complete MORONS lol

when was the last time you saw anything inside the government work as intended? any paperwork submitted to a federal agency can be FOIA'd so that would be... the IRS

If you've got access to chatgpt o1 pro, which carries a Massive compute - you can ingest a SHIT LOAD of public data to get this. I'm releasing a lead magnet that has most of this information on my twitter here in a day or so, @makershelpdesk

A few months ago, I caught Gumroad trying to roll out MOR (Merchant Operator of Record) on Twitter and immediately called them out on it. They were attempting to launch this feature before it was even completed.

They were internally rolling it out globally, even for parts of the world without a taxation system to collect on. Nevertheless, Gumroad was still planning to apply these charges. This was despite the fact that there are countless existing and affordable platforms, 100,000 times more developed—that could seamlessly integrate with Gumroad and already have this system in place. However, they decided to build it themselves.

Why? I’m not entirely sure, but I’m willing to bet it has something to do with Sahil taking a position at DOGE and perhaps an effort to obscure payment activity.

Why else? All the other scaled and fully built systems are likely to face heavy scrutiny over the next 24 months. And if a system is "still being built" or "unfinished," I guess it’s much harder to comply with regulators or provide data in response to FOIA requests, isn’t it?

things that make you go hmmmmm.

fairly certain I dethroned all content tools.

What started as a meme has unexpectedly turned into a project worth sharing. After showing it around to a few people, I’ve gained some users, and I’m beginning to believe I might be on to something here.

I’m a marketer with 20 years of experience, highly tech-enabled but not a coder by any stretch of the imagination. While I can’t write a single line of code, I deeply understand how applications work and the challenges involved in building them. Likely thanks to my ADHD and some advanced pattern recognition skills.

One thing I’ve always hated is trying to create good content or blog copy using traditional big LLMs. Yes, they’ve come a long way, but the output always feels too "annunciated" and "exaggerated.” Even when you use brand voice documents to guide them, the result is rarely closer than 65% to something actually shippable. That still leaves a ton of manual editing on my end.

So naturally, I started asking myself: "What if I built my own solution?"