Distantly.<p>I'll scan headlines a few times per day, and read the odd story. One problem I'm encountering with websites, aside from the ads and annoyances, is that <i>much of the news seems curiously irrelevant</i>.<p>There's a list of text-only news sites which ... slightly ... improves the situation, but only just, as few media outlets bother to <i>organise</i> their text-only offerings. What's generated is typically an unordered jumble of clickbaitish headlines. That said, the listing:<p><<a href="https://greycoder.com/a-list-of-text-only-new-sites/" rel="nofollow">https://greycoder.com/a-list-of-text-only-new-sites/</a>><p>Of those I'll hit NPR, CNN, and the Christian Science Monitor most often.<p>CNN annoyed me so much that I looked under the hood ... and found that the URLs encode both a news topic and date along with a title, e.g., from current headlines:<p><pre><code> https://lite.cnn.com/2025/03/08/us/long-island-fire-suffolk-county/index.html
</code></pre>
Link: <<a href="https://lite.cnn.com/2025/03/08/us/long-island-fire-suffolk-county/index.html" rel="nofollow">https://lite.cnn.com/2025/03/08/us/long-island-fire-suffolk-...</a>><p>So, that's an 8 Mar 2025 article in US News about fires on Long Island.<p>Turns out that it's a SMOP[1] to extract URLs and headlines from the published Web page, and regenerate a far more useful page grouped by subject heading. Another interesting side effect is that I've accumulated an archive of > 4,400 stories published since mid-December 2024 and can easily run statistics on the story distribution. There are 30 sections, which I've aggregated to a smaller set of five, represented as follows:<p><pre><code> US: 1298 (29.29%)
World: 917 (20.70%)
Business: 638 (14.40%)
Science: 371 (8.37%)
Trivia: 1207 (27.24%)
----------------------------------------
Total: 4431 / 4431 [0] (sections crosscheck)
</code></pre>
Note that nearly a third of the "news" is trivia topics (Travel, Food, Homes, Entertainment, Sport, Style, CNN). A deeper look at both US and World news shows that about 2/3 of the former is relatively minor nonsystematic crime, and the latter is entertainment / gossip / personality. It's not that CNN leans one way or the other, it's just mostly <i>irrelevant</i>.<p>(I'd started the project in part to quantify how much this was the case, and to dump all the trivia in one place so I didn't have to parse individual headlines to suss out what was what.)<p>Other sites are less amenable to this as they use random strings in URLs (e.g., <i>Financial Times</i> and NPR). Most website, <i>even the so-called "lite" or "text" versions</i>, read quite poorly in a terminal-based browser (I favour w3m, but others differ little).<p>I'll listen to foreign news services a fair bit, including CBC, BBC, ABC (Australia), Deutschlandfunk, and more. I find the quality of foreign-service news seems a bit better than most US sources. Most of these are via podcast.<p>I've been listening to <i>The World</i>, out of WBUR Boston. Unlike virtually all audio/video news these days, the programme is pre-recorded <i>and edited</i>. Its focus tends to reduce US-centric news, which I prefer, and what I find is that the editing and pre-recording make the segments (reports and interviews largely) far less disjoint than, say, NPR, BBC, or the PBS News Hour, which I find all but unlistenable.<p>Among other podcasts are those focusing heavily on philosophy and history, which I find tend to <i>inform</i> on current events without <i>wallowing</i> in them. Peter Adamson's <i>History of Philosophy</i>, David Runciman's <i>History of Ideas</i> (and his several earlier podcasts), Stephen West's <i>Philosophize This</i>, and Nigel Warberton and David Edmunds's <i>Philosophy Bites</i> are all high on my list. Among those dealing with contemporary politics, Ezra Klein tends to be insightful without being overly grating or outrage-inducing, while also including a range of guests across the political spectrum (though he quite admittedly leans left), while also managing to generally keep them away from thought-stopping cliches and stale talking points by pushing back gently but effectively in a way that few interviewers seem capable of. Runciman also has this skill, though it's less often evidenced, see his <i>Intelligence Squared</i> (UK) appearances for good examples.<p>For those interested, my CNN category breakdown is:<p><pre><code> 489 (11.04%) US News
809 (18.26%) US Politics
76 (1.72%) World
111 (2.51%) World -- Americas
62 (1.40%) World -- Africa
112 (2.53%) World -- Asia
29 (0.65%) World -- Australia
37 (0.84%) World -- China
17 (0.38%) World -- India
260 (5.87%) World -- Europe
183 (4.13%) World -- MidEast
30 (0.68%) World -- UK
91 (2.05%) Economy
369 (8.33%) Business
1 (0.02%) Markets
85 (1.92%) Tech
2 (0.05%) Cars
11 (0.25%) Investing
79 (1.78%) Media
112 (2.53%) Science
170 (3.84%) Health
48 (1.08%) Weather
41 (0.93%) Climate
163 (3.68%) Travel
26 (0.59%) Food
3 (0.07%) Homes
496 (11.19%) Sport
405 (9.14%) Entertainment
109 (2.46%) Style
5 (0.11%) CNN 10 Minutes
</code></pre>
That runs ~12 Dec 2024 through today, and is de-duplicated, though it's also based on when I've happened to request the latest CNN headlines, and may miss some stories. Ordering is roughly by my level of interest and/or what seems to be logical classification to me. I've picked up a few additional classifications (e.g., "CNN 10 Minutes" and "Cars") after my initial run, there is some logic to make previously unknown categories jump out.<p>________________________________<p>Notes:<p>1. <<a href="https://en.wikipedia.org/wiki/Small_matter_of_programming" rel="nofollow">https://en.wikipedia.org/wiki/Small_matter_of_programming</a>>