5 ms·

Useful study. UK-based. The "authenticity" thing of podcasters is only meaningful if the podcaster was there. Sometimes that happens, and those are the good on

by Animats 7mo ago

Useful study. UK-based. The "authenticity" thing of podcasters is only meaningful if the podcaster was there. Sometimes that happens, and those are the good ones. There are good protest videos. Not many war videos. Secondary sources are just pundits, of which we have too many. It's easy to be an influencer who covers entertainment - entertainment wants to be watched. It's hard to be an influencer who covers, say, unemployment. It's possible, but you have to go and talk live to people who just got laid off. That's reporting. It's not the delivery system. It's whether the source goes out and pulls in news. Most don't. “Whatever a patron desires to get published is advertising; whatever he wants to keep out of the paper is news." - City Editor of a Chicago newspaper, 1918. Look at a news story and ask "did this begin with a press release or a speech?". If so, it's publicity. HN had an article from a few days ago about "CEO says" journalism. It's worse on the political front. Democracy requires that a sizable fraction of voters know what's really happening. This is a big problem. Influencers can be controlled. Dubai has cracked down on war reporting by the large number of influencers there.[1] Right now, Iran claims a missile hit on an Oracle data center in Dubai. The UAE denies this. Did anybody in Dubai drive over and take pictures? Call up Oracle and ask? Nah. [1] https://www.theatlantic.com/national-security/2026/03/dubai-influencers-war-messaging/686329/ https://www.theatlantic.com/national-security/2026/03/dubai-...

rectang 7mo ago
I would argue that reading discussion of "CEO says" "journalism" on HN will often better inform you than reading a mainstream journalist puff-piece interview of a CEO. Many journalists will not provide adversarial viewpoints, because to do so would stanch the flow of interview subjects.
- i80and 7mo ago
  "Access" is the filthy dirty word here. Can't be anything other than a stenographer because you might lose your ACCESS, with which you can do MORE stenography. It's a sickness.
rdevilla 7mo ago
> Did anybody in Dubai drive over and take pictures? Call up Oracle and ask? Nah. Why is any of that necessary? LLMs can just synthesize a story for the press de novo, without reference to prior developments or indeed any 20th century style on-the-ground reporting. Reporters should in fact be pleased that this meaningless drudgery has been automated out of the profession. Besides, whatever "facts" are presented will be labelled as fake news by its detractors, who will not have their own internal narratives swayed in the slightest. The rest will rest easy in their confirmation bias, it now being confirmed once more.
- Animats 7mo ago
  Sadly, that's not a joke. There are multiple news articles repeating the IRGC claim of a successful attack. Some have pictures of missiles and Oracle buildings. Most of the same building stock shots, including some of Oracle HQ in San Mateo, CA.[1] There are also denials from the UAE press office calling this "fake news". Oracle doesn't seem to have said anything. It's not even clear that they have a major data center in Dubai. They have two in Saudi Arabia (Riydah and Jeddah), but the Oracle status page lists none in Dubai.[2] They have some offices in Dubai, all on upper floors of buildings and probably not large data centers. Do press reports indicate a call to the Dubai office? (Phone: 971(4)3909000) [3] Nah. [1] https://www.youtube.com/watch?v=yPo4vJjq-3M&t=17718s https://www.youtube.com/watch?v=yPo4vJjq-3M&t=17718s [2] https://ocistatus.oraclecloud.com/#/ https://ocistatus.oraclecloud.com/#/ [3] https://ae.bizdirlib.com/node/924412 https://ae.bizdirlib.com/node/924412
  - rdevilla 7mo ago
    I am not sure how to navigate this world any more.
    - Animats 7mo ago
      It's like intelligence analysis. You get various bits of intel. Some are true, some are false, some are biased, some are fake. Intel is evaluated in terms of source reliability and information credibility, which are separate. Bracketing helps. If China Daily and Breitbart News agree on something, that's reasonably solid. But check to see if they're getting their info from the same place. Distinguish between two independent sources and two paths from the same source. When both sides are too far apart, digging for more data is needed. It's often available, and best obtained from sources close to the situation. Not pundits. Recognize that you're not going to be sure. Accept that and realize that the goal is to get good enough quality information to make a decision. Being too sure is an error of its own. Here's a basic article on the subject from the CIA.[1] Much of this is about the problems of being too sure. All of this is way too much work for most people. (This is not something I've done professionally. I have worked with people who did.) [1] https://www.cia.gov/resources/csi/static/Article-Principles-of-Intelligence-Analysis-Studies65-4-Dec2021.pdf https://www.cia.gov/resources/csi/static/Article-Principles-...

by Animats 7mo ago

rectang 7mo ago

I would argue that reading discussion of "CEO says" "journalism" on HN will often better inform you than reading a mainstream journalist puff-piece interview of a CEO. Many journalists will not provide adversarial viewpoints, because to do so would stanch the flow of interview subjects.

i80and 7mo ago
"Access" is the filthy dirty word here. Can't be anything other than a stenographer because you might lose your ACCESS, with which you can do MORE stenography. It's a sickness.

rdevilla 7mo ago

> Did anybody in Dubai drive over and take pictures? Call up Oracle and ask? Nah. Why is any of that necessary? LLMs can just synthesize a story for the press de novo, without reference to prior developments or indeed any 20th century style on-the-ground reporting. Reporters should in fact be pleased that this meaningless drudgery has been automated out of the profession. Besides, whatever "facts" are presented will be labelled as fake news by its detractors, who will not have their own internal narratives swayed in the slightest. The rest will rest easy in their confirmation bias, it now being confirmed once more.

Animats 7mo ago
Sadly, that's not a joke. There are multiple news articles repeating the IRGC claim of a successful attack. Some have pictures of missiles and Oracle buildings. Most of the same building stock shots, including some of Oracle HQ in San Mateo, CA.[1] There are also denials from the UAE press office calling this "fake news". Oracle doesn't seem to have said anything. It's not even clear that they have a major data center in Dubai. They have two in Saudi Arabia (Riydah and Jeddah), but the Oracle status page lists none in Dubai.[2] They have some offices in Dubai, all on upper floors of buildings and probably not large data centers. Do press reports indicate a call to the Dubai office? (Phone: 971(4)3909000) [3] Nah. [1] https://www.youtube.com/watch?v=yPo4vJjq-3M&t=17718s https://www.youtube.com/watch?v=yPo4vJjq-3M&t=17718s [2] https://ocistatus.oraclecloud.com/#/ https://ocistatus.oraclecloud.com/#/ [3] https://ae.bizdirlib.com/node/924412 https://ae.bizdirlib.com/node/924412
- rdevilla 7mo ago
  I am not sure how to navigate this world any more.
  - Animats 7mo ago
    It's like intelligence analysis. You get various bits of intel. Some are true, some are false, some are biased, some are fake. Intel is evaluated in terms of source reliability and information credibility, which are separate. Bracketing helps. If China Daily and Breitbart News agree on something, that's reasonably solid. But check to see if they're getting their info from the same place. Distinguish between two independent sources and two paths from the same source. When both sides are too far apart, digging for more data is needed. It's often available, and best obtained from sources close to the situation. Not pundits. Recognize that you're not going to be sure. Accept that and realize that the goal is to get good enough quality information to make a decision. Being too sure is an error of its own. Here's a basic article on the subject from the CIA.[1] Much of this is about the problems of being too sure. All of this is way too much work for most people. (This is not something I've done professionally. I have worked with people who did.) [1] https://www.cia.gov/resources/csi/static/Article-Principles-of-Intelligence-Analysis-Studies65-4-Dec2021.pdf https://www.cia.gov/resources/csi/static/Article-Principles-...