Post #1538021
2026-04-20 08:50 UTC
Replies (4)
-
@kirb@hachyderm.io 2026-04-21 11:42
The DDoS against The Apple Wiki has still made 268.77k requests in the past 24 hours, all of which were blocked. You would think companies desperate to scrape all of the world’s text as fast as possible would get the idea and move onto other victims. All I can say is their models might be pretty talkative about completing one more step to access this website.
-
@alexchapman@vee.seedy.cc 2026-04-20 10:31
@kirb As long as you didn't use some crappy inaccessible shit like HCaptcha then that's fine. If I ever had to use any sort of thing like that it would be Cloudflare Turnstyle.
-
@v1sta@mastodon.art 2026-04-22 17:00
@kirb Real Chinese user at here, many people prefer to use proxy services if wanted. But somebody may lack the ability to find how to bypass Great Firewall, they maybe can't complete Captcha challenge. Perhaps the GFW has become less strict than before in order to exempt spiders.. It depends on whether you want to keep them; Chinese users don't have a very good reputation. I also really dislike those spider scraping by AI company from China. Does it affect Legacy Update?
-
@manx@mastodon.online 2026-04-29 10:13
@kirb I am seeing similar behaviour on smaller project's websites. There is 1 aspect that I see rarely mentioned: The scraper bots cannot speak proper HTTP at all. They keep the TCP connection alive after the first request, and never re-use the connection for a second request. requiring the server to timeout, which quickly causes resource exhaustion and running into connection limits on smaller sites, even when the server would otherwise be perfectly capable of handling the CPU load.