Right off the gate, while we do have libraries where i live, they are mostly outdated or don’t have books on controversial topics, So how does anyone find reliable information when so much of the internet is slop 🤢 and search engines only give the same top 5 "How to stuff 🤑 " websites.
I have been listening to a couple of podcasts 📻 about debunking popular pop culture misinformed science 🧑🔬 claims(Your wrong about, Maintenance phase, If books could kill) and that got me interested in trying to research my own topics. But where does one start. I understand that there is no universal solution 🧪 , really what it takes, is alot of googling and going beyond the results first page but maybe there are some good sources?
Sharing some of what i found useful here:
-
While libraries in my area are not really an option, looking up books online is a good start. I use goodreads to look up different topics and see what the reviews are saying about the quality and reliability of the book. Its also worth fact the citations of books.
-
Turns out google is very good with finding pdfs 📄 , so add type:pdf to whatever your searching. you can also add site:.gov 🏢 to search only gov websites ,or site:.ca 🇨🇦 to search websites that label themself Canada in the URL, for example.
-
Use before:, A lot people suggest that you use before:2023 to filter out most AI Slop, date before chatgpt came out, but also i have found interesting stuff when going back further like before:2015.
-
I have tried alot of different search engines and they all just serve google results. the one good one is Marginalia . Its not really reliable like say, google in certain cases is, but alot of the time will serve me new sources, its focus is on non-commercial websites
-
The last one is kinda of mixed bag(and Meta). When you ask people online for suggestions for something, like say some podcasts recommendations, they tend to give you the same five podcasts. Sometimes its something totally different but its a hit or miss most of the time.
P.S: I discovered emoji markdown just now but its kinda of lacking a bit. Still cool though,


Kagi seems well regarded as a search engine, dunno I use my own SearxNG which takes hits from multiple engines (Marginalia is one of my mains). Occasionally I fire up a local LLM and pipe it through SearxNG via vane when I’m fighting slop for current issues. (Slop vs Slop, sigh, but at least I get references I can hand check).
For specific interests there is often a specific forum where knowledgeable people gather, if you find it you’ll get good results, e.g. Endless Sphere for eBikes and batteries and such. It may be reddit unfortunately, before: is your friend where useful. There are high quality but still accessible magazines out there, e.g. Quanta for maths and Knowable for more general science.
With academic papers have a look at the impact factor of the journal, high impact is highly competitive and quality is usually higher. If you’re dealing with heavy jargon, translation is something LLMs are actually sot of good at, use them to see if the info you want is actually there before wading through the jargon yourself to check / understand deeper.
IMO Snopes.com is a good source of truth on popular things that may be BS.
Your last point is an example of a statistical phenomena called a long tail. It also applies to search engine results ironically.
Another thing that comes into play is the XY problem where you think you know what you’re looking for, but it’s actually a specific case of a more general problem, and hunting just keeps turning up bad results. Good to let things rest, or ask a question somewhere.
There are useful tools that can help you as well, I like Obsidian for note taking, but there are many, while Zotero for organizing papers and Calibre for books are pretty invaluable. No good finding stuff out if you just forget it in a month, and once you have a large library running grep over your own stuff can often turn up useful info right at your fingertips.