choraria

That's helpful. I think I need to beef up profanity keywords A LOT more than I'd originally anticipated. On it!

No disagreements there. My goal is to make this THE BEST list (of all, if any) out there. As much as I see AI taking over handling variations, it'll still need an exhaustive source of truth about what's what and I really think I'd be able to provide that with this API.

Right on all counts. Twitter is a rather simplified example. I see it as something that literally every platform can use. Say, ProductHunt, other platforms that offer product launches, link-in-bio tools etc. etc. I'm a bit bullish around the market because, regardless of me knowing all of 'em, the challenge of using usernames exists in general.

On the static list, yes. Me too. But I keep updating mine as well. For ex: on day 1, "apple" was just a dictionary word. On day 2, it was also classified as a brand. Also, every quarter, half-yearly or yearly, there are newer companies, public figures whose usernames keep getting to be significant. Currently, though manually, I intend to maintain this list for the long run.

As for a better, permanent solution, on another comment, I came across using an LLM/classifer for this (based on my understanding, that's not just asking OpenAI but building an LLM of my own) where I have the "best" source of truth and the LLM handles all variations. I think it actually is solvable to an extent now. Though, I'm not sure what the final solution looks. I WILL SOLVE THIS THOUGH :D

YOU NAILED IT! That's part of what I want to explore next too. One of the other common thing I've heard is to check other social media platforms to check if a username is reserved. But, namevine.com already did that. Maybe it doesn't do it as accurately anymore. Almost all platforms are super protective of their API usage in that way. In any case, that's definitely the next challenge to solve for.

I see. Will re-run through the categories and the datasets from which I've adopted the names and categories. Maybe either I missed something or it might've not existed in the import in the first place. But noted. Also, thanks :)

Interesting. I think you're right (on the API category this falls under). Also love the approach on keeping this API async. Makes so much more sense that way.

Yes; I've gotten that request from another person on LinkedIn too for bulk checking existing usernames. Will work on releasing that shortly too. Thanks for being helpful and constructive all the way throughout the convo :)

Hey @sampli — was there some kind of bundling that you were looking for?

Interesting! Didn't think about it that way. Currently, it's a super dumb system. There's a list of ~1.7 million records and the API simply looks-up against that. Super lazy approach. Was avoid running an API across OpenAI or other model but didn't think about hosting a classifier/LLM myself. Might consider it in the future.

Full disclosure: I'm not a developer. I understand tech architectures well. Can code (have coded in JS pre-AI too) BUT will figure this out as I go along. Thanks and truly appreciate the input.

Edit note: added million next to 1.7. fml!

I think there's a general aversion to subscriptions at the moment so wanted to offer this on a usage-based pricing to begin with. While I may have to (hope not to though) will switch to subscription if that's what most users end up asking for. Thanks for the note and comment though. Much appreciated.

True. I've tried to add language where possible. I think currently, it's only on dictionary words so if the username is a dictionary word in another name, it would be flagged. It may or may not show-up under the "restricted" category though.