Ask HN: How do you find credible information online?

Websites which strictly reporting events and stick to the 5 Ws (What, Who, When, Where, Why).

They don't even try to make predictions and only talk about the past, not the future

Picking good sources is hard, though. Generally I find it has to be the case that specialists writing about their specialism are the people who knowledge originates with, even when it isn't lay people.

Where curation is worth it is in reviews. The London Review of Books will introduce a wide array of new literature to me and prepare the ground for approaching those books by explaining the content and critically analysing it.

Statistical literacy also helps - for this I've read David Spiegelhalter's Art of Statistics, which is for the lay reader.

News-wise, single generally credible sources are hard to come by. The Guardian does good live blogs, especially on UK parliamentary debates. The New York Times has a solid but anglocentric newsgathering operation, but tragically partisan opinion. Le Monde Diplomatique is good for international stuff from a francophone perspective.

tl;dr, navigate institutions, not search engines. Institutions are in effect communities with a culture that guards against rot algorithms can't perceive.

We desperately sought help for our child who was reading at pre-k reading levels while in the 3rd grade. In many US school systems, after 3rd grade, if you haven't acquired basic reading skill, then you are completely screwed: the curriculum moves on. Finding useful information was astonishingly hard -- what we learned is that within the US, an entire industry has evolved around dogmas that are completely uninformed by reproducible research. If you search for reading help, you find appeals to emotion, "this-one-trick-worked-for-me" type books, self-help conferences and such. We also learned that professional teaching programs are built around dogma, and publishing "research" in the field of education requires strict adherence, and does not meaningfully apply hypothesis testing or reproducibility.

The practice of reading pedagogy lives in this sweet spot where it kinda-sorta works for just enough students that school districts and admins can assert that kids for whom the system does not work should adjust their expectations in life.

We were lucky to discover that reading problems are well understood through neuroscience (not education), and there are decades of reproduced research supporting useful ideas. We were also lucky that my wife happens to be a good teacher (I could never do this!), the two of them get along well, she left her paid job, started homeschooling our child full time, and our child now reads far above grade level.

The return on time spent is horrible and I also realize that more often than not, I don't have the time and knowledge knowledge to sift through and evaluate the information that I find.

1. Credible sources. 2. Tools and methods for searching and matching.

In my opinion:

1. It's very subjective. Pepople just believe in different things. It's hard and time consuming. There is no the single point of truth. 2. For this purpose I wrote https://github.com/livelace/gosquito and I extract different keywords (tech, people, orgs etc.) from information flows. It frees my time heavily.