>>/10548/
> The motivation is obvious given the many recent changes to that site.
What I've been doing for like one month now: automatically grabbing booru.org subdomains because it's shutting down and the greedy pig site owner cares very little about archiving and boorus that don't make him money.

 >>/10551/
> Unfortunately a few months back almost all the free nitter instances stopped working
I think I remember that. Maybe .onion or .i2p ones work better now compared to clearweb ones.
> seems that a lot of people were using the instances for mass tweet scraping
I'm curious about a breakdown on this: "ai" training, social ops/drama, personal archiving, public archiving, ad-related/data brokers maybe.
> recommended approach was to set up your own instance to do that and to leave the public ones for everyone else's convenience
I was thinking of that when I was writing  >>/10550/ - perhaps I will see it that private instance method still works.

> gallery-dl
I didn't know it worked for that.

 >>/10545/
> search results page is now worse
Other thing I noticed:
> https://archive.4plebs.org/pol/search/text/trans%20contradiction
> 245 results found.
But that page only shows 25 results and not more 245! That's another website which became worse: only shows a small amount of search results. /mlpol/ search then and now:
Thousands">https://web.archive.org/web/20231210064132/https://archive.4plebs.org/pol/search/text/%22my%20little%20pony%22[/code]Thousands of posts across >15 pages. (Code markup because Endchan will mess up the link otherwise.)
SERP">https://web.archive.org/web/20240625151322/https://archive.4plebs.org/pol/search/text/%22my%20little%20pony%22[/code]SERP shows only one page with 25 posts.