Privacy Search Engines

Big search engines like Google track your activity and sell your data for ads — which is why a lot of us don’t search freely anymore. I’ll share a few alternatives I’ve found, and would love to hear what others are using too! DuckDuckGo: Best for untracked searching Startpage: Best for unprofiled browsing MyAllSearch: Clean interface and an emphasis on user privacy. Qwant: GDPR-protected searching Startpage: Combines the top results of multiple engines, primarily Gigablast and Yandex.

does anybody else feel like duckduckgo has been extremely enshittified recently? the search results are worse than before plus it's pushing ai... anyways startpage up top

Most of the time I ask AI to search and compile a list of results with summaries.

i wanna throw wibby into the ring here

>>88746 youre not the only one, ever since they started using AI summaries instead of showing a short text preview of the websites on their search results page i knew it was only gonna be downhill from there

>>88746 They stopped being trustworthy when they openly and proudly admitted to censoring their search results on the Ukraine proxy war.

Searx redirector that sends your query to a bunch of random SearX instances: https://searx.neocities.org/ This has some problems but I still use it in spit of often needing to search multiple times on it before I get useful results. It needs to be updated, a couple of the instances are dead and some others vomit up useless dogshit from Bing by default. Also I have to block instances that "protect" their website with Anubis malware before they get the chance to crash my browser.

I don't care that much about my data and I don't really see adds anyway, but the results of all search engines have been getting really unreliable due to all the SEO and AI crap. At this point I mostly look up wiki pages and subreddits related to what I'm after and work my way from there.

>>88744 for me, it's 4get.ca made by lolcat the bestest search website on this planet

I understand people complaining about DDG Implementing AI as it's basically shoving AI down throats. I use a multitude of different search engines based on what I'm searching for, but I have in fairness found the DDG AI Useful in certain applications.

For a long time, starting December 23rd, 2023, modern versions of Firefox have been redirecting all Google searches to https://www.google.com/httpservice/retry/enablejs when JavaScript is disabled. This was an A/B test and it hasn't been rolled out to all users that time, so tabloids have noticed only when the A/B test had ended in January, and the “anti-feature” was applied universally to everyone. Some users have noticed it still worked in Lynx. Indeed, changing the user agent to Lynx had made searches working again, although the resulting page markup was quite poor. To get proper modern markup without triggering the JavaScript requirement, you simply had to set your User-Agent to a sufficiently old version of Firefox (no later than version 23). So I configured my browser to selectively present the desired user agent only for Google search using an extension, this has allowing me to continue using search. I refrained from publishing this method publicly to prolong its usefulness. However, all of that is now a thing of the past. Google websites now require JavaScript to be enabled even when presented with an old Firefox user agent. User agents like Lynx/2.9.2 libwww-FM/2.14 SSL-MM/1.4.1 GNUTLS/3.8.5 now receive the following message: >Update your browser >Your browser isn't supported any more. To continue your search, upgrade to a recent version. [Learn more](https://support.google.com/websearch/answer/16515119) Therefore, I have to stop using the Google search engine, just as I have stopped using Bing in 2020, which returned empty search results for any non-trivial query. Now it returns not empty search results, but fake ones. Likely an anti-scraper measure.

P.S. >December 23rd, 2023 Not 2023, but 2024

>>88777 If you really want to use Google, try 4get, as mentioned above. It's a metascraper -- returns search results for one of the 20 different engines at your choise, including Google and Google CSE. JS is not required.

>>88786 It uses the same HTML I used to be using: https://git.lolcat.ca/lolcat/4get/src/branch/master/scraper/google.php Quick test has returned >This scraper returned an error: >Google fuarrrked some soykaf up again. The noJS endpoint i was using got nuked. I have another endpoint, so its not completely over... But yeah, when they patch that soykaf, then I will have no other option but to use their API. So let's do a bit of trolling and abuse the free tier for their API. They give 100 free searches per day, so if every1 gives me API keys I can rotate through them and do some epic trolling. Sign up at https://developers.google.com/custom-search/v1/overview and email me your keys at will@lolcat.ca - make sure to use a different Google account for each key!! Ok thanks love you all xoxo

>>88791 Are you truing to self-host? The main instance just works.

Thanks for the info. And big thanks to the author of the tool. I tried again - and it worked. It turned out Google's JS uses the same endpoint for HTML results and the API and the author of the tool has reverse engineered Google's API (results are returnd in obfuscated form, so he has to write the deobfuscator lib: https://git.lolcat.ca/lolcat/4get/src/branch/master/lib/fuarrrkhtml.php ) and utilized it. But I'm not going to use that engine. Instead I'll comprehend the source and make a browser addon using the same technique.

>>88822 There is also https://leta.mullvad.net/ No JS required and it's available over .onion and from 4get.