Hacker News Reader

Comment by dang | Hacker News Reader

An article about a conversation that hasn't happened yet doesn't pass the "significant new information" test that we use to distinguish which instances of long-running stories count as dupes: https://hn.algolia.com/?sort=byDate&prefix&page=0&dateRange=.... When such submissions make HN's front page, all that happens is that the same discussion from the last N cycles gets repeated, since there's nothing new of significance to talk about. Of course, when something new does happen, the same discussion will mostly repeat then too—but at least there will be some mutation to keep things from getting too generic.

For those who are curious, this test originated after the Snowden deluge of 2013 (https://hn.algolia.com/?sort=byPopularity&prefix&page=0&date...) and has turned out to work pretty well.

View on HN

That's not how news works. The New York Times generally doesn't publish rumors. Multiple trusted sources told famous, celebrated journalists the same thing in independent conversations. If it's a dupe that has nothing to do with rumors.

Your fallacy: Appeal to Authority

https://en.wikipedia.org/wiki/New_York_Times_controversies

And these are just the examples that we know about and gained enough notoriety to be included in a Wikipedia article. If the NY Times treated itself with the same scrutiny as it treats Uber, we'd know more about how reliable it actually is.

Like all other media companies, the NYTimes at the end of the day is made up of humans like you and I and it's entirely possible for them to publish heresay, conjecture, biased opinions, etc. They may have a better track record than most, but they aren't infallible. Even the golden goose occasionally takes a shit.

This article is just presenting the conjecture of supposed insiders as facts. That's not news. That's rumor.

I used "rumor" in the dictionary sense, which doesn't depend on the factors you mention. But you're right that it isn't necessary to make the point. The point is that, be it rumor or celebrated journalism, there's no significant new information here yet, so no harm in waiting until the thing actually happens (actually for HN purposes it's better). There will be no dearth of celebrated journalism then either. So I've replaced "rumor" with "article" above.

Maybe we need to start adding a filter that bans the submission of articles authored by journalists whose material doesn't meet the criteria for submission on HN?

Some journalists writing for mostly respectable media companies have veered so far from journalism and increasingly often publish articles that are just their opinion, conjecture and heresay. Maybe it's time the content aggregators such as HN start penalizing poor quality journalism that leads to low-quality discussion them by banning submission of their work.

I'm also in favor of banning articles with titles and introductions that present one angle that the journalist is promoting and then bury the only evidence refuting that angle in the last paragraph or two.

http://www.slate.com/articles/technology/technology/2013/06/...

Articles from major newspapers are increasingly engineered to go viral at the expense of journalistic integrity and presenting unbiased facts. Content aggregation sites need to stand up and somehow communicate that enough is enough.

Fair enough. Good to know.

What's the long-running article that this is a dupe of?

https://news.ycombinator.com/item?id=14530797

Not article; story. Uber.

Edit: Oops, forgot that this was a dupe in the strict sense (thanks detaro). My point was that a second test ("significant new information") applies during deluges.

The story is 'Uber'? I don't understand: were their a lot of submissions of this article? Was it marked as a dupe because we have a lot of submissions about Uber (on different topics)?

This article was a straight-up dupe of another submission of the same article.

Even if it hadn't been, though, the 'significant new information' test—i.e. the lack thereof—would have made it a dupe. Otherwise we end up with a front-page of Uber Uber Uber the same way we had Snowden Snowden Snowden in 2013, with the discussions always the same.

> a straight-up dupe of another submission of the same article

At work, when we have two or more tickets like this, we close one as a Duplicate and link them together with a 'duplicated by/duplicate of' relationship, so that people who end up on the ticket we're not going to work on can find the active one. Do you think this sort of system would work for dupes on HN?

I think Hacker News tries to avoid news about people and prefers news about ideas and businesses. Don't think they really want to be gawker.

Reply on news.ycombinator.com