17 ms·

Looking at the list of removed stories makes me really happy with the moderators here. They're all sensationalist, advertising for some company, clickbait, way

by a12k 3y ago

Looking at the list of removed stories makes me really happy with the moderators here. They're all sensationalist, advertising for some company, clickbait, way off topic, or some combination of above. In fact, I don't see a single story that I personally feel should not have been removed. Thanks, mods.

hubraumhugo 3y ago
dang has been doing a fantastic job for years now. How big is his team? What kind of tools are they using? Would love to read a writeup sometime, but I guess there are good reasons to keep this secret.
- giraffe_lady 3y ago
  https://www.newyorker.com/news/letter-from-silicon-valley/the-lonely-work-of-moderating-hacker-news https://www.newyorker.com/news/letter-from-silicon-valley/th...
  - next_xibalba 3y ago
    Tldr the answer is “2”. Not sure how posting a link where this info is buried is helpful.
    - giraffe_lady 3y ago
      Then downvote it, that's what the button is for! The other commenter said they were interested in a write up with more information than that one question. But hey, can't please everyone.
    - hk__2 3y ago
      > Tldr the answer is “2”. Not sure how posting a link where this info is buried is helpful. The answer to OP’s first question, but there was a second one: > How big is his team? What kind of tools are they using?
  - dang 3y ago
    sctb stopped working on HN in fall 2019, alas!
zug_zug 3y ago
Agree, and to be clear, that's why I upvoted this submission -- it really is an endorsement of the algorithm and moderation we see here. I know the person who wrote the article did it from a place of skepticism, but it functions as a nice gold-star transparency report.
TheCoreh 3y ago
At a quick glance, I found several that don't match that criteria you mention, here are a few: Open Source Doesn't Require Providing Builds https://codeengineered.com/blog/2024/open-source-not-builds/ https://codeengineered.com/blog/2024/open-source-not-builds/ Sam Altman Says AI Using Too Much Energy Will Require Breakthrough Energy Source https://futurism.com/sam-altman-energy-breakthrough https://futurism.com/sam-altman-energy-breakthrough Avoid Async Rust at All Cost https://blog.hugpoint.tech/avoid_async_rust.html https://blog.hugpoint.tech/avoid_async_rust.html (Perhaps that last one could be renamed to be less hyperbolic, but the content was still an interesting opinion piece) I don't think this is being done by the mods, by the way. It's more likely some spam filter with false positives, report brigading, or an anti upvote ring mechanism.
- nottorp 3y ago
  YMMV. I don't want to see Altman's fearmongering and hyperbolic statements. At this point he's indistinguishable from a bitcoin advocate or a tv preacher.
  - threatofrain 3y ago
    It's dubious that HN mods think that way of Altman though.
    - nottorp 3y ago
      s/mods/users? everyone can flag stories.
      - wolverine876 3y ago
        It's not flags (or not sufficient to remove the story): https://news.ycombinator.com/item?id=39231055 https://news.ycombinator.com/item?id=39231055
        lolinder 3y ago
        Flags are sufficient. I just posted a comment on the comment you linked to: I have many times been the person who pressed "flag" on a story and then watched it immediately disappear. I think there's some threshold of flags to upvotes and possibly some other metrics that determines whether a story vanishes, but flags can absolutely tip the scales.
eureka-belief 3y ago
It really is impressive how HN has been such a quality community for so long. I can’t think of any of many other online communities that I have been using for 10+ years. So definitely much gratitude to the mods from me for the work they do.
sandworm101 3y ago
https://www.nasa.gov/solar-system/asteroids/nasa-system-predicts-impact-of-a-very-small-asteroid-over-germany/ https://www.nasa.gov/solar-system/asteroids/nasa-system-pred... Really? A NASA report, on the official .gov site? Maybe the comments were horrible but that seems right in the middle of what HN is interested in.
- dang 3y ago
  You're talking about this submission: NASA System Predicts Impact of a Small Asteroid over Germany - https://news.ycombinator.com/item?id=39126705 https://news.ycombinator.com/item?id=39126705 - Jan 2024 (18 comments) It was downweighted because it was a dupe (or quasidupe) of this: Scientists discover near-Earth asteroid hours before it exploded over Berlin - https://news.ycombinator.com/item?id=39103412 https://news.ycombinator.com/item?id=39103412 - Jan 2024 (46 comments) That's the system functioning as intended. We work hard to try to prevent repetition from taking over the site, because repetition is the enemy of curiosity (https://hn.algolia.com/?dateRange=all&page=0&prefix=false&sort=byDate&type=comment&query=curiosity%20repetition%20by:dang https://hn.algolia.com/?dateRange=all&page=0&prefix=false&so...). One thing I've learned today, after 11 hours of posting in this thread, is that it's easy to look at an article in isolation and say "Really? That got moderated?" - when in fact if you know the larger context there's nearly always a straightforward explanation. One can certainly argue that 86 points and 46 comments is too low a threshold to treat the repost as a dupe, but that's a different question, no?
  - defrost 3y ago
    > after 11 hours of posting in this thread The effort is appreciated by some if not most.
    - dang 3y ago
      It wasn't 11 hours straight by any means. But thanks! It's nice of you to say that. When I get tired I start to complain. It's a bad habit.
mrcwinn 3y ago
Garry Tan seems to benefit from this system as well. Nothing sensationalist about tracking his awful behavior.
- dang 3y ago
  We haven't touched those stories except reduce the penalties on them (user flags mostly) and moderate them less than we normally would (per https://hn.algolia.com/?dateRange=all&page=0&prefix=false&query=by%3Adang%20moderate%20%22less%2C%20not%20more%22&sort=byDate&type=comment https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...). I put one back on the front page last night despite this contradicting every principle HN stands for—every other principle, that is, than the first one, which is that we moderate HN less, not more, when YC or a YC startup is part of a story. I posted detailed explanations in those other threads: https://news.ycombinator.com/item?id=39224560 https://news.ycombinator.com/item?id=39224560 https://news.ycombinator.com/item?id=39210947 https://news.ycombinator.com/item?id=39210947 https://news.ycombinator.com/item?id=39172045 https://news.ycombinator.com/item?id=39172045 If you (or anyone) read those explanations and still have a question that I haven't answered there, I'd like to know what it is. These practices have been in place for many years and haven't changed.
  - bradly 3y ago
    @dang Thank you for the info. One questions I do have–I would guess posts critical of HN/YC are going to get a log of flags and have not the best discussion. This has a side affect of biasing the home page to not have posts critical of HN/YC. Do you see this as a problem?
    - jsnell 3y ago
      The home page already has a massive bias in favor for pretty much any kind of negativity, but doubly so for anything involving a tech company. It doesn't matter whether the stories are untrue, unverified, repetitive, etc, they'll still get voted up, and the comment threads will half full of low quality complaints repeated from past discussion, often only tangentially related to the submission. And it creates a very visible feedback loop, as users start to think that this is what HN is supposed to be. They're probably the biggest quality problem of HN.
hyperluz 3y ago
[flagged]
oramit 3y ago
Yeah this list seems to be pretty low quality stuff. There's a couple economic/political links that I think are interesting but I can totally see why they would be removed as off-topic or likely to produce a flamewar. It's pretty clear to me that any online forum needs good moderation to be healthy long term. HN has been good about this with a strong community providing upvotes/downvotes and a moderation team that seems pretty light handed but not afraid to say no when necessary. Please keep doing what you're doing.
tmaly 3y ago
Props to the mods for keeping the post quality high. However, I do see a few decent posts in this list that probably warrant a second chance.
- dang 3y ago
  Happy to look at specific links if you want to mention any!
kortilla 3y ago
The moderators are mainly the users. Flags are what kills a story quickly
- dang 3y ago
  You're right, but I'd like to add that we do turn off flags sometimes when we think a story (a) has a good (if not high) chance at a substantive discussion and (b) hasn't had much discussion previously. If anyone notices such a case, they're welcome to email us at hn@ycombinator.com and we'll take a look.
screye 3y ago
> mod*s* plural ?
reductum 3y ago
I witnessed a recent front page link silently get changed to point to a parody video, then silently changed back later, with the top comment that remarked on the change silently removed. That told me all I needed to know about the moderation of this site. Thankfully someone captured a screenshot: https://merveilles.town/@cancel/111834048502040552 https://merveilles.town/@cancel/111834048502040552
- mkl 3y ago
  Dang explained it was a mod error: https://news.ycombinator.com/item?id=39182625 https://news.ycombinator.com/item?id=39182625, which links to https://news.ycombinator.com/item?id=39170137 https://news.ycombinator.com/item?id=39170137 huppeldepup's comment in the screenshot is collapsed on p2 of the comments (after it was no longer relevant) https://news.ycombinator.com/item?id=39157010&p=2 https://news.ycombinator.com/item?id=39157010&p=2, and is also accessible as the parent of https://news.ycombinator.com/item?id=39170137 https://news.ycombinator.com/item?id=39170137.
  - paulcole 3y ago
    What did you think he was going to say? Explain that it was in fact a big HN conspiracy? I mean yeah it probably was just a mistake but what do you expect somebody to say?
    - mkl 3y ago
      I expected a matter-of-fact explanation of a simple error, and a comment thread that was collapsed - exactly what I found. reductum implied there wasn't an explanation and that the complaining comment thread had been deleted without comment, both of which were hard to believe, so I went looking.
    - dang 3y ago
      You just need to move your game theory to a different level. The expected value of lying about such things is super negative and the expected value of telling the truth is super positive. I'm sure there's a model in which lying some of the time but not too often has marginally higher expected value, but it's also going to have significantly higher risk and that's not worth it, plus you have to be disciplined enough to actually apply such a strategy. One slip and you're dead! I'm too lazy for that.
paulcole 3y ago
> I don't see a single story that I personally feel should not have been removed Another way to look at this is that the mods have the same biases you do. Depending on how you’re feeling on a given day, you could call that an echo chamber.

by a12k 3y ago

hubraumhugo 3y ago

dang has been doing a fantastic job for years now. How big is his team? What kind of tools are they using? Would love to read a writeup sometime, but I guess there are good reasons to keep this secret.

giraffe_lady 3y ago
https://www.newyorker.com/news/letter-from-silicon-valley/the-lonely-work-of-moderating-hacker-news https://www.newyorker.com/news/letter-from-silicon-valley/th...
- next_xibalba 3y ago
  Tldr the answer is “2”. Not sure how posting a link where this info is buried is helpful.
  - giraffe_lady 3y ago
    Then downvote it, that's what the button is for! The other commenter said they were interested in a write up with more information than that one question. But hey, can't please everyone.
  - hk__2 3y ago
    > Tldr the answer is “2”. Not sure how posting a link where this info is buried is helpful. The answer to OP’s first question, but there was a second one: > How big is his team? What kind of tools are they using?
- dang 3y ago
  sctb stopped working on HN in fall 2019, alas!

zug_zug 3y ago

Agree, and to be clear, that's why I upvoted this submission -- it really is an endorsement of the algorithm and moderation we see here. I know the person who wrote the article did it from a place of skepticism, but it functions as a nice gold-star transparency report.

TheCoreh 3y ago

At a quick glance, I found several that don't match that criteria you mention, here are a few: Open Source Doesn't Require Providing Builds https://codeengineered.com/blog/2024/open-source-not-builds/ https://codeengineered.com/blog/2024/open-source-not-builds/ Sam Altman Says AI Using Too Much Energy Will Require Breakthrough Energy Source https://futurism.com/sam-altman-energy-breakthrough https://futurism.com/sam-altman-energy-breakthrough Avoid Async Rust at All Cost https://blog.hugpoint.tech/avoid_async_rust.html https://blog.hugpoint.tech/avoid_async_rust.html (Perhaps that last one could be renamed to be less hyperbolic, but the content was still an interesting opinion piece) I don't think this is being done by the mods, by the way. It's more likely some spam filter with false positives, report brigading, or an anti upvote ring mechanism.

nottorp 3y ago
YMMV. I don't want to see Altman's fearmongering and hyperbolic statements. At this point he's indistinguishable from a bitcoin advocate or a tv preacher.
- threatofrain 3y ago
  It's dubious that HN mods think that way of Altman though.
  - nottorp 3y ago
    s/mods/users? everyone can flag stories.
    - wolverine876 3y ago
      It's not flags (or not sufficient to remove the story): https://news.ycombinator.com/item?id=39231055 https://news.ycombinator.com/item?id=39231055
      - lolinder 3y ago
        Flags are sufficient. I just posted a comment on the comment you linked to: I have many times been the person who pressed "flag" on a story and then watched it immediately disappear. I think there's some threshold of flags to upvotes and possibly some other metrics that determines whether a story vanishes, but flags can absolutely tip the scales.

eureka-belief 3y ago

It really is impressive how HN has been such a quality community for so long. I can’t think of any of many other online communities that I have been using for 10+ years. So definitely much gratitude to the mods from me for the work they do.

sandworm101 3y ago

https://www.nasa.gov/solar-system/asteroids/nasa-system-predicts-impact-of-a-very-small-asteroid-over-germany/ https://www.nasa.gov/solar-system/asteroids/nasa-system-pred... Really? A NASA report, on the official .gov site? Maybe the comments were horrible but that seems right in the middle of what HN is interested in.

dang 3y ago
You're talking about this submission: NASA System Predicts Impact of a Small Asteroid over Germany - https://news.ycombinator.com/item?id=39126705 https://news.ycombinator.com/item?id=39126705 - Jan 2024 (18 comments) It was downweighted because it was a dupe (or quasidupe) of this: Scientists discover near-Earth asteroid hours before it exploded over Berlin - https://news.ycombinator.com/item?id=39103412 https://news.ycombinator.com/item?id=39103412 - Jan 2024 (46 comments) That's the system functioning as intended. We work hard to try to prevent repetition from taking over the site, because repetition is the enemy of curiosity (https://hn.algolia.com/?dateRange=all&page=0&prefix=false&sort=byDate&type=comment&query=curiosity%20repetition%20by:dang https://hn.algolia.com/?dateRange=all&page=0&prefix=false&so...). One thing I've learned today, after 11 hours of posting in this thread, is that it's easy to look at an article in isolation and say "Really? That got moderated?" - when in fact if you know the larger context there's nearly always a straightforward explanation. One can certainly argue that 86 points and 46 comments is too low a threshold to treat the repost as a dupe, but that's a different question, no?
- defrost 3y ago
  > after 11 hours of posting in this thread The effort is appreciated by some if not most.
  - dang 3y ago
    It wasn't 11 hours straight by any means. But thanks! It's nice of you to say that. When I get tired I start to complain. It's a bad habit.

mrcwinn 3y ago

Garry Tan seems to benefit from this system as well. Nothing sensationalist about tracking his awful behavior.

dang 3y ago
We haven't touched those stories except reduce the penalties on them (user flags mostly) and moderate them less than we normally would (per https://hn.algolia.com/?dateRange=all&page=0&prefix=false&query=by%3Adang%20moderate%20%22less%2C%20not%20more%22&sort=byDate&type=comment https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...). I put one back on the front page last night despite this contradicting every principle HN stands for—every other principle, that is, than the first one, which is that we moderate HN less, not more, when YC or a YC startup is part of a story. I posted detailed explanations in those other threads: https://news.ycombinator.com/item?id=39224560 https://news.ycombinator.com/item?id=39224560 https://news.ycombinator.com/item?id=39210947 https://news.ycombinator.com/item?id=39210947 https://news.ycombinator.com/item?id=39172045 https://news.ycombinator.com/item?id=39172045 If you (or anyone) read those explanations and still have a question that I haven't answered there, I'd like to know what it is. These practices have been in place for many years and haven't changed.
- bradly 3y ago
  @dang Thank you for the info. One questions I do have–I would guess posts critical of HN/YC are going to get a log of flags and have not the best discussion. This has a side affect of biasing the home page to not have posts critical of HN/YC. Do you see this as a problem?
  - jsnell 3y ago
    The home page already has a massive bias in favor for pretty much any kind of negativity, but doubly so for anything involving a tech company. It doesn't matter whether the stories are untrue, unverified, repetitive, etc, they'll still get voted up, and the comment threads will half full of low quality complaints repeated from past discussion, often only tangentially related to the submission. And it creates a very visible feedback loop, as users start to think that this is what HN is supposed to be. They're probably the biggest quality problem of HN.

hyperluz 3y ago

[flagged]

oramit 3y ago

Yeah this list seems to be pretty low quality stuff. There's a couple economic/political links that I think are interesting but I can totally see why they would be removed as off-topic or likely to produce a flamewar. It's pretty clear to me that any online forum needs good moderation to be healthy long term. HN has been good about this with a strong community providing upvotes/downvotes and a moderation team that seems pretty light handed but not afraid to say no when necessary. Please keep doing what you're doing.

tmaly 3y ago

Props to the mods for keeping the post quality high. However, I do see a few decent posts in this list that probably warrant a second chance.

dang 3y ago
Happy to look at specific links if you want to mention any!

kortilla 3y ago

The moderators are mainly the users. Flags are what kills a story quickly

dang 3y ago
You're right, but I'd like to add that we do turn off flags sometimes when we think a story (a) has a good (if not high) chance at a substantive discussion and (b) hasn't had much discussion previously. If anyone notices such a case, they're welcome to email us at hn@ycombinator.com and we'll take a look.

screye 3y ago

> mod*s* plural ?

reductum 3y ago

I witnessed a recent front page link silently get changed to point to a parody video, then silently changed back later, with the top comment that remarked on the change silently removed. That told me all I needed to know about the moderation of this site. Thankfully someone captured a screenshot: https://merveilles.town/@cancel/111834048502040552 https://merveilles.town/@cancel/111834048502040552

mkl 3y ago
Dang explained it was a mod error: https://news.ycombinator.com/item?id=39182625 https://news.ycombinator.com/item?id=39182625, which links to https://news.ycombinator.com/item?id=39170137 https://news.ycombinator.com/item?id=39170137 huppeldepup's comment in the screenshot is collapsed on p2 of the comments (after it was no longer relevant) https://news.ycombinator.com/item?id=39157010&p=2 https://news.ycombinator.com/item?id=39157010&p=2, and is also accessible as the parent of https://news.ycombinator.com/item?id=39170137 https://news.ycombinator.com/item?id=39170137.
- paulcole 3y ago
  What did you think he was going to say? Explain that it was in fact a big HN conspiracy? I mean yeah it probably was just a mistake but what do you expect somebody to say?
  - mkl 3y ago
    I expected a matter-of-fact explanation of a simple error, and a comment thread that was collapsed - exactly what I found. reductum implied there wasn't an explanation and that the complaining comment thread had been deleted without comment, both of which were hard to believe, so I went looking.
  - dang 3y ago
    You just need to move your game theory to a different level. The expected value of lying about such things is super negative and the expected value of telling the truth is super positive. I'm sure there's a model in which lying some of the time but not too often has marginally higher expected value, but it's also going to have significantly higher risk and that's not worth it, plus you have to be disciplined enough to actually apply such a strategy. One slip and you're dead! I'm too lazy for that.

paulcole 3y ago

> I don't see a single story that I personally feel should not have been removed Another way to look at this is that the mods have the same biases you do. Depending on how you’re feeling on a given day, you could call that an echo chamber.