If you're going to debunk an article about something like a lack of vaccine safety and need to retract a couple of sentences later, but the overall conclusion is the same, then the conspiracists will be right on you with a diff of the changes from the archive trying to piece together what it all means and who paid you off.

Preventing people from seeing the change log of a page because "conspiracy theorists might use it" is dangerous thinking. Saying that it's okay to forego transparency because it could be used to bolster people who disagree with you is backwards. That's precisely why transparency should be retained - when you're the person disagreeing, it benefits you.

Later in the day, they changed the headline to be something like "Insurgents invade Capital building", but did not change the 10AM timestamp up top. One had to read the article to find the update timestamp.

Newsmaxx, and thus many if its viewers looked at the headline and headline timestamp and trumpeted "how did they know?! It's a conspiracy!"

I had to look into the wayback machine to see the original headline to show to a relative who'd bought the Newsmaxx narrative. I'm not sure they totally believed or even understood what I was telling them, but I am certain it raised some doubt.

All of that would was unnecessary, if only NPR had done proper change management on its articles.

News websites really need to stop continuously rewriting their articles. It's a PITA if you want to cite them as a source, then all the sudden what you cited isn't there.

I've had NY Times articles get completely rewritten while I read them (admittedly with some long pauses between when I started and finished), and when I went back to quote something, it wasn't there anymore.

I don't think this is nefarious or anything, it's just them trying to adapt to expectations that they'll always be up to date with the latest stuff. You have the same article about some rapidly changing story hanging around their app/website all day (and "most popular articles" facets), and I suppose most people expect those things to be up to date.

I understand mistakes happen, but you should be always very clear on your corrections and if you are a web publication, have revision history links.

Source: Journalism degree holder that is tired of the MSM. I advise all the people I know to stop reading/watching it. So tired of the outrage driven news cycle that is often highly misleading.

Not the person you were replying to, but I would love this, yeah. That changelog proposal being implemented would be more than enough to resolve most of the issues I have with publications that silently alter contents of their articles after they were originally published (without indicating that an alteration took place).

If they want to be an encyclopedia then be one.

You're misreading the situation. The media doesn't have a crystal ball any more than you or I do. They can be "committed to it being an accurate story at the time of publish," but then new stuff happens after that time, or they have the time to go into more depth. The news isn't just about accuracy, it's also about timeliness.

Sure, but the media is not writing encyclopedia entries. They make hyperbolic claims about how they break stories before anyone else, and how accurate they are.

But since they are human, sometimes mistakes are made, and I think it's pretty dishonest to quietly change an article after it's been published. Publishing a story should mean that you back its content 100%, and it's the best, most accurate story you could have written with all available information. If new information comes out later that changes things, write a new article that references the mistakes of the original. There were still decisions made to include the content that was later shown to be incorrect, and I would _much_ rather see a follow up explaining what they got wrong than "correcting" the article after publication.

Well, breaking news stories is their job an no one else's.

> If new information comes out later that changes things, write a new article that references the mistakes of the original. There were still decisions made to include the content that was later shown to be incorrect, and I would _much_ rather see a follow up explaining what they got wrong than "correcting" the article after publication.

I'm sure a front page of mea culpas and minutiae about the process of reporting itself would scratch some people's ideological itches, but it wouldn't actually be very useful as an information source about what the current state of things are, which is what people are actually going there to get.

Personally I'd prefer they just issue a new version of the story with a kind of version history (or at least leave the old version up somewhere), at least for major revisions, but I also kind of empathize with them not wanting to do that, because that would probably be a feature almost exclusively used by annoying pests trying to stir up BS outrage (e.g. like the kind of people who think that software should never have bugs).

By having a fluid changing story, two people won’t even have the same reference point if they both read the same NYTimes article.

I was trying to find some information on the previous activity of a cop who recently became infamous in my area for beating and choking a man who was pleading for his life, to see if he had ever been in the news before. So I used the time window feature on Google, only to get tons and tons of articles published that same day that were backdated years because of bad change management.

See for yourself on the link to a google search below, results are filtered between 2018 and January 2021, but every single article for the first five pages is about the incident that occurred two weeks ago, mistakenly backdated years.

https://bit.ly/3xEpCGA

Is that the news website's fault, or Google's fault?

The first two links are to what look like keyword/topic pages with many articles, and it wouldn't be surprising of Google pulled a different date than what you're looking for out of such a page. Some of the other hits appear to be old articles with a "new stories" bar (not sure exactly what you call it, here's an example: https://boingboing.net/2020/11/15/trump-admits-biden-won-ele...).

Personally, I don't consider Google date searching to be very reliable, IMHO. They're making a promise with it they can't really keep.

It might work better if they searched on first-indexed date, rather than server modification timestamp headers or dates embedded within the page.

But that assumes they index things in a timely matter, which is likely not true a huge amount of the time.

And I don't think that would help in this case, because in many of the cases it appears they got the article date right, but the article page is dynamic and now includes snippets of stuff that post-dates the article.

Knowing when an article was written is a useful piece of information that is being destroyed by these always evolving articles. It might not be full on Orwell yet, but these always-mutating-articles are opening the door to an Orwellian future.

We need to be able to go back and look at the past unfiltered and unmodified.

It doesn't. That was my point. I wasn't praising or criticizing NPR. I was saying that conspiracy theorists shouldn't be a reason either for or against transparency.

Just saying I can see how you could rationalise this decision, even if it’s a terrible idea.

Sadly if you do retractions, you are not qualified for the name "fact checker". Because "facts" are not retractable.

And fact-checkers are just the tip of the iceberg, and if you care to look, there are many things that are peddled as being "fact" when there is not enough credible evidence to say so..

For example the statement "Vaccines don't cause autism". Any one with two brain cells will know that this NOT a fact, simply because it does not say "Which vaccines".

But that does not prevent people from saying that as a FACT.

So it is not only them. It is you as well.

That said, fact checks are typically written by journalists or institutions claiming to serve a journalist function; and so are fully governable by common journalist practices such as retractions.

Retractions are nice because they acknowledge that there was a human error somewhere in the organization’s processes, memorialize the error but then correct the record with supplementary and reviewable matter. Doing this also provides a signal to readers that even though this particular institution filled with these imperfect humans made a mistake on the record, they sufficiently value their reputation to own the mistake and correct it on the record.

Good faith and honesty can take you far.