top | item 33926746

(no title)

nickvincent | 3 years ago

Super cool idea, thanks for building and sharing!

One possible suggestion for future iteration: I think it's a bit of a shame that everyone's been sharing ChatGPT outputs almost exclusively through screenshots. I threw together a (very quick and rudimentary) browser extension I've been using to save some of my more exciting transcripts in JSON files. I think something along these lines could be really great for sharing purposes, especially if people want to study these outputs more systematically (e.g. for research, or just a kind of crowd-sourced audit).

Here's the barebones extension, just as an example: https://github.com/nickmvincent/chatgpt-exploration/tree/mai...

There's also a longer discussion to be had about the best way to do this: ideally, we might save some of the formatting information too and/or more metadata (exact timestamps for responses, etc.), but I think the json with plaintext is adequate (at least for personal retrieval use).

discuss

order

matt1|3 years ago

Thanks, totally agree about sharing screenshots. I originally started building the site intending people to share text-only examples, but it gets complicated with the formatted responses that ChatGPT generates. And since everyone is sharing screenshots right now anyway, it made sense to just support that for this v1. I do want to add support for sharing text responses so screenshots aren't necessary in all cases. And maybe a Chrome extension could make the process easier, including working with formatted output. Thanks for sharing yours; it could be a good starting point for one build for LearnGPT.

ctchocula|3 years ago

Not working for me on Firefox 107.0.1 / MacOS. I see the save button and the ChatGPT screen has a blue edge. However, when I hit "Save Conversation" the chatgpt_2022-12-10.json file that I get only shows [] and doesn't have the conversation.

nickvincent|3 years ago

Thanks! Looks like a bunch of changes were made to page structure and CSS (in a way that makes scraping a bit more inconvenient). Filed an issue for myself, but looks like the other suggestions in this thread may be better approaches to avoid this kind of fragility.

aftbit|3 years ago

Someone needs to OCR all of the GPT screenshots on twitter and bootstrap them into one of these databases.