Hacker News Reader

Comment by rekabis | Hacker News Reader

> it doesn’t have the same modern security tech stack reddit.com has so we need to tighten security on old reddit to keep it viable.

This is how you justify it to anyone who doesn’t know web dev.

For those who do, it’s an idiotic statement. The only difference between new Reddit and old Reddit is the visual styling and how content is presented. Both have all the same security.

View on HN

They don’t mean websec, they mean textsec. The old site doesn’t have the level of text scraper defenses that the new site does, without which Reddit can’t monetize the written output of their userbase for AI training. Putting old behind a login wall solves that.

Having previously recommended on HN that blogs put up a simple basicauth that takes any password to stop AI theft dead in its tracks, I definitely agree with their reasoning; if nothing else, it’ll result in an honest accounting of users left using it.

Potentially related is yt-dlp getting a 403 lately when downloading from Reddit. The extractor uses old.reddit.com. Hopefully they'll find a way to use the regular Reddit gateway instead.

But I can't stand new Reddit so maybe I'll take this as an opportunity to give it up entirely.

Yeah, that would align; this probably also cuts off most of the “download Reddit videos” websites. More textsec, though I guess that’s more like contentsec? CONSEC there we go.

>a simple basicauth that takes any password to stop AI theft dead in its tracks

https://www.youtube.com/shorts/GWu2fY9Pu-c

I’m having trouble opening this link; could you please reply in your own words instead?

in my own words: if something is publicly visible, it will get scrapped. you could rate limit your website to 1 request per minute with a custom Turing test upon each request, and it would still, still get fully scrapped. the Internet is getting scrapped by hundreds of players, and at least some of them will bother to investigate and bypass the countermeasures even for a tiny obscure blog.

Ah. I’m not talking about a perfect and total block like Reddit wants. I just want the vast bulk of scrapers (whether AI or not) to hit a 403 auth-required wall and give up, which is what they’ll mostly do. Good enough for me :)

If I did want a near-perfect block I’d post to a telnet BBS and have forums accessible only by QWKmail. Which would of course drive some jerk to synthesize a republish just to troll me. Isn’t humanity wonderful?

Reply on news.ycombinator.com