Comment by voodooEntity | Hacker News Reader

Comment by voodooEntity | Hacker News Reader

So first of all its a very cool project even tho i don't code ruby.

But in case the author of the page is in here, please read the following:

Some years ago i had a similar idea, just instead of ruby-cons i went for defcon videos. I build a page that indexed all defcon talks and even tried to categorize them in terms of topics and difficulty. Added a nice search, and put it out as "defcon.video". It didn't take google long to message me and tell me i had to take it down because its actually forbidden to host an alternative search index if you are using the youtube API. You are not allowed to store the results of youtube API (longer than 24hours if i remember) and as stated before not allowed to offer an alternate search.

While you might argue: why didn't you just state that you don't use the api to grab the data and shutoff the token usage : well because google is not really known to be friendly and i expected them to just close down my whole developer account if i don't follow.

So - in case not all the videos are hand inserted (and even then they might come at you) google definatly will reach out to you and try to force you down. Back when it happend to me i didn't have the time and patience to fight them back so defcon.video was shutdown.

Best of luck tho

> well because google is not really known to be friendly and i expected them to just close down my whole developer account if i don't follow.

and if their algorithm is in a bad mood that day, they'll shut down your personal account(s) as well, and possibly adjacent accounts. Hope you don't rely on personal email or Google Photos or Android if that happens.

This nuke-somebodys-life-from-orbit capability of such a gigantic company is IMHO one of the best arguments for regulatory limits on company size/scope. That's obviously easier said than done and comes with a whole column of cons, but the power imbalance is so unbelievably great and the reach so deep that in some cases I'd rather be on the wrong side of the state than on the wrong side of Google.

Well that's the reason why im running my private mailserver and the only thing im using google for is actual google services.

I always wondered why people would just give all their personal information to google in terms of mail and files (convenience i know but still..).

But than you see the posts of "google just shutdown my account without any reasoning and i lost everything help!!" here on hackernews and you know exactly why you don't use google for anything but their services.

I tell people I left Google because of all the weird amount of data harvesting, and they always argue Apple's not any better, but I rarely hear of someone losing their iCloud to the level I hear of people losing their Google Account which is glued to too many damn things.

At this point I assume the only way Google will ever fix their ridiculous ban process is if someone passes a law that forces them to separate these comingled services from losing you access to your email, and essentially in some cases your personal business.

The talks are all indexed manually and are stored in yaml files, see: https://github.com/adrienpoly/rubyvideo/tree/main/data.

We are just embedding the video using vlite.js and the YouTube Video ID (which is just embedding an official iframe in the end). So there isn't really an account to "shut down", unless I'm missing something.

That sounds good :) tbh i didn't have the time to check your source. So lets hope google understands it and does not herras you :)!

Oh...

I am currently working on building a curated collection of YouTube channels, with custom search index. Thank you for sharing this, I guess I can't continue. It's a shame.

What is different about https://pyvideo.org/ ? That one has been running for many years.

Appears to be user-curated instead of sourced from the YouTube API.

I guess give every API response a LLM laundering and there you go.

Reply on news.ycombinator.com