Hacker News Reader

Hacker News

Comment by AlexTWithBeard | Hacker News Reader

Full thread

AlexTWithBeard

TL, DR:

Push journalism is what's happening now:

- The reporter has a hunch about a story

- The reporter checks with a couple of sources and their editor

- The story is published

- Then everyone in the organization PUSHES LIKE HELL to get the story in front of as many people as possible

Pull journalism would be the alternative (better) approach:

- Have a hunch about a problem that people might be experiencing

- Interview those people – perhaps a dozen of them – and listen for signals that your hunch is correct

- Also listen for signals that your hunch is incorrect

- Formulate a hypothesis based on these conversations about what might address the problem

- Go back to those dozen people, if possible, and present your work. Rinse. Repeat.

- Publish it, and then present it to all of the people that you’ve spoken with and ask them, if they believe it’s useful, to let people know.

View on HN

One issue that immediately comes to mind is that the people you're interviewing often don't want the story covered, or they want it covered in a way that promotes their angle on the issue. The "go back and present your work" part collapses there because they're going to be actively fighting against publishing, not promoting it. I'm not sure what adversarial journalism looks like under the pull model.

That's handled by going to multiple people. If you are seeing fighting by all or some people, or if you have conflicting information, that's a story in itself.

The important part is going to those people in the first place, whether they cooperate or not. You generally get a good idea of what's going on when you present later collected facts to people and get their reaction.

Treat it like a police investigation, you don't just interview a witness once, you interview them, go interview people from their statement, then come back and interview them again to see if everything matches up with what everyone said.

> Treat it like a police investigation, you don't just interview a witness once, you interview them, go interview people from their statement, then come back and interview them again to see if everything matches up with what everyone said.

Lying to the police can lead to charges like obstruction of justice, filing a false police report, lying to the FBI, etc.

Journalists don't have this weapon in their quiver. Unless you're going to propose giving journalists subpoena power, your "come back and interview them again" approach simply isn't going to work.

The article is the charge. Like I said, if they don't comment then note that in the article, and ask others.

If noone comments then note that in the article.

"Treat it like a police investigation" without giving journalists police powers (which they rely heavily upon to complete an investigation) results in a pretty pointless process.

Since when is verifying facts a pointless process?

There are many other ways to verify facts versus an interview with a reluctant/belligerent subject of a piece.

Such as?

Push journalism is how Woodward and Bernstein reported on Watergate.

Pull journalism is a much more respectful model, but, as it turned out, the people involved in Watergate didn't deserve respect nearly as much as the people in the entire rest of the country deserved to know about Watergate.

Most things that matter are not watergate. Most things that matter happen over the course of years or decades, not weeks or months, and absolutely do require more depth and consensus-building to be anything other than partisan noise. Pull journalism is more respectful, but it's not more respectful to it's subjects, it's more respectful to its readers by not treating them like children at a circus.

This "pull journalism" doesn't sound like journalism at all. It sounds more like public policy analysis done by a subordinate for his superiors, which is a totally different thing.

> Go back to those dozen people, if possible, and present your work. Rinse. Repeat.

This is a bad idea. If you do this, your notional dozen people (or company/government PR departments) will try to manipulate the story in their favour.

I agree.

How about: you keep repeating the cycle until all respondents are equally mad at you for you not siding with them?

Then you put your hard hat on and publish.

Due to the revenue structure that exists in the journalism industry, the pull method is becoming less often used or only used in cases where the directors feel they have a solid footing.

I would argue that this is a society issue since people are less inclined to want to and pay for material that has more insight. Society wants quick, flashy snippets of what is happening or what someone think is happening. This creates bad incentives that have led to quite a bit of damage to society so far.

Pull journalism, doesn't really feel like news reporting, sounds like it would be expensive - and not necessarily very interesting.

Pull journalism is what news reporting used to be.

It's very interesting to have all the facts, it's not interesting to wonder if what you're watching is true or not. It makes everything questionable and makes it not even worth watching.

It doesn't really sound like the journalism at least as it existed back in the 80s and 90s when I was a journalist.

Outlets that aren't optimizing for eyeballs / ad revenue (e.g., NPR) seem to operate more on the "pull" journal model.

Access Journalism is also what's happening now:

- Uncritically report what powerful people say so that they will talk to you again tomorrow

Absolutely.

In the political sphere: the current system allows both politicians and news publications to play access games for favourable coverage as you've described.

One system design that is worth exploring: randomised access by accredited journalists / press outfits.

In this system interested parties register their interest to ask a question (from an accredited list of journalists / news publications).

That list is used to randomly select `n` questioners to put their question to be answered.

Reply on news.ycombinator.com