Find an area of interest or general framing question[1] and follow that.
I've been online since the 1980s. I've come to feel that there's a general hierarchy of informational quality by medium. From highest to lowest:
- Books. Particularly Great Books. The list at the end of Mortimer Adler's How to Read a Book is an excellent start.[2] Your library, Project Gutenberg, and the Internet Archive offer lawful access to a tremendous set of titles. Library Genesis and ZLibrary extend that but thumb noses at copyright. IMO in a truly justified sense. I tend not to follow recently-published works closely, it takes about a decade at least for actual value to surface. There are specific exceptions to this, of course, but in general (and as a theme for what follows), the odds that the most useful and compelling work on some topic has been published recently is ... low. Audiobooks are an acceptable substitute or adjunct to reading text. Both fiction and nonfiction have value, though I read far more of the latter. Syllabii from academic courses on a specific topic are an excellent curation tool. I also follow references and bibliographies, and stalk specific authors of interest. Writing authors with specific questions can be productive, but don't abuse the privilege.
- Articles and essays in traditional publications, both academic and popular.
- High-quality produced & edited podcasts or academic or structured lectures / discussions. I'm listening to a law-school podcast episode at the moment. I've listened most to several philosophy podcasts (many of our current questions and problems ... are not especially new. Even where earlier philosophical discussion is wrong, it has often anticipated many present questions and discussions (and the realisation of this can be amusing, frustrating, and/or illuminating, variously). It's also a remarkable tour through just plain wrong results which can be arrived at through many centuries of mislead rational thought. I follow several foreign-language podcasts (mostly as a language-learning aid), and a few topical podcasts. The less these focus on present news and politics, the better. Ezra Klein is the principle exception to that set. Long-form interviews can be quite good. The New Books Network offers a huge list of channels and a tremendous back-catalogue of academic books, though the interviewer and production quality are both highly variable. It's an excellent guide to what's coming out of academic presses, and tends to be eclectic. Not all books are worth reading, or even listening to authors talk about. London School of Economics has an excellent lecture series. There are several university press podcasts, some extant, some defunct, though again, back-catalogues are useful. Some YouTube videos approach this, though these tend to be lectures or presentations, occasionally conversations. The less advertising, the higher the content quality (more below).
- Wikipedia and several related wiki-type sites. Wikipedia and RationalWiki are amongst my favourites. Wikipedia has become my preferred option for reading up on / understanding current news, particularly complex and developing stories. I'd first come to this realisation during the 2004 Boxing Day Indian Ocean Tsunami and Earthquake, in which I watched the article develop from a first mention of a strong quake to the present multi-page form. Very few news organisations can even come close, and Brad Plummer's coverage of the Oroville Dam failure (whilst he was at Vox) is among the few favourable comparisons I can make. Wikis digest multiple sources into a single, usually coherent, generally current, whole. Note that not all wikis are created equal, though the major mainstream ones tend to be quite good.
- Reputable news sources. The less frequently updated (e.g., quarterly, monthly, weekly) the better. Time resolves many early question and filters much churn. Finding several sources from several locations is quite useful. If you're interested in learning or improving a foreign language, reading or listening to news in another language can be handy.
- Specific authors' blogs or article archives. Keep in mind that an authors' best work is typically what they've published, and many good writers have really poor blogs, websites, or far more often, Twitter or other social media feeds. There are exceptions, and those are what I'm pointing at. Note that this is five notches below that primary content: books.
- Broadcast news. Daily news has some of the least durable value of any information. Being aware of a five-minute headline summary is almost always sufficient. I typically rate radio over television and noncommercial / public broadcasting over private. I've never owned a television and haven't watched anything in years. I've all but entirely curtailed radio listening.
- Generally: avoid media which include or are supported by advertising. Ads have a reverse Midas touch: they turn everything to shit. See Hamilton Holt's 1909 book Commercialism and Journalism for an early succinct argument as to why. <https://archive.org/details/commercialismjou00holtuoft>. There's a larger literature: <https://old.reddit.com/r/dredmorbius/comments/7k7l4m/media_a...>.
- Online discussions. Most are poor. A very few are modestly useful. HN is among the best I know presently. I did some informal research in 2015 identifying where more substantive discussions might be found, with some interesting findings. See "Tracking the Conversation" <https://old.reddit.com/r/dredmorbius/comments/3hp41w/trackin...>. A similar methodology utilising other indicia (I've done some initial exploration based on philosophers) might be interesting or useful.
- Large-scale social media. Conversation scales poorly, and larger social media fail in many, many regards. Even intelligent people are often highly incoherent and/or generate far too many / off-topic posts or comments. Facebook is also all but entirely external-search opaque. I will occasionally use Nitter to either catch up on specific Twitter profiles (often via RSS), or to search some present topic of interest. Replies all but entirely detract from the overall informational value in most cases. Facebook, Twitter, TikTok, etc. Reddit is virtually always useless, though there are occasional exceptions. The site's dynamics are actively anti-conversation-generative. Subreddits > 10k subscribers tend to fall tremendously in quality. Conversation scales poorly.
- Tabloid / outrage / promotional / SEO press. Crap used to either spread propaganda or sell ads, but I repeat myself. Block at your router. E.g., Daily Mail.
Once you've found useful information, find where those are being discussed significantly, intelligently, and non-polemically. You'll find in general that this is exceedingly rare.
There's a recurring trope in fiction and myth of the guru on the mountaintop. There's a reason for that.
If you want to talk to a guru, first sort out who that is. Then find their mountain. It's usually their book.
________________________________
Notes:
1. My own for over a decade has been "what are the big problems". Exploring potential candidates, what those are, what they're founded in, and how they interrelate could occupy many lifetimes.
2. <https://www.worldcat.org/title/300152756>