It should be noted that during the aforementioned six month periods, both the number of new comments made monthly and the average score for a given comment decreased, although this post claims a 30% increase to Hacker News traffic due to changes in the comment system, which is interesting assertion of causality. (in fairness, the decrease in number of new comments could be caused by the increased moderation of bad comments.)
It would be helpful if this article in this submission clarified how to flag comments, as that action is unintuitive (you have to click the comment permalink first) [EDIT: looks like this was added to the submission]
Comment scores are visible to the Algolia API after a period of time. Otherwise, the API returns a "null" for the points field. (example: https://hn.algolia.com/api/v1/items/2921983)
Oh, interesting. I stopped indexing HN comments after the points data became unavailable (it was still available through the Octopart guys' api, until Algolia took over). It looks like comment scores start showing up around the 30 day mark.
dang I like the current design (it is no nonsense), but could one change be made to the location of the vote arrows. Could the down vote triangle be moved to after the users name (e.g. ^ dang v)? This would make it so much easier to avoid accidentally down voting on mobile devices.
A handful of browser add-ons exist that will update the visual design and generally make the site more pleasurable to use. Here's one for Chrome: http://gabrielecirulli.github.io/hn-special/
A third experiment didn't go so well: we briefly made the software kill comments that had been sufficiently downvoted. Many users objected, arguing that killing downvoted comments is too harsh a punishment for unpopular opinions, especially since downvoted comments get faded to begin with. We heard that and reversed the change.
Good to hear they decided to roll that back.
PG opened a pandora's box when he said that downvotes were acceptable when you merely disagreed with someone, instead of doing so for civility. There is no end of sadness and bullshit caused by that remark.
It'd be nice to see an official policy statement changing that position, but I don't think it's likely. In the meantime, we will simply have to stop assuming that downvoted comments are actually, you know, bad.
> In the meantime, we will simply have to stop assuming that downvoted comments are actually, you know, bad.
I don't think this is true. Given that downvoted comments become faded, I have a strong reason to believe that the intent is to hide them because they are actually bad
Yes, I think that was the actual intent of the system. However, bad != disagreeing with the majority. If the up/down vote functionality is to be used to express agreement or disagreement, which is a change from the original philosophy, then the system should change accordingly. angersock's point is that, as it stands now, a faded-out comment may simply be expressing a good but unpopular opinion.
You're using circular reasoning - that the comments are faded because they're bad, and the evidence they're bad is that they're faded.
But, people downvote comments for any number of reasons, and even doing so for mere disagreement is within the acceptable guidelines - and of course for everything, despite the lack of clear context, the effect remains the same.
I delete comments to keep the site clean and on-topic, and not too argumentative. It's not 'threatening to delete my comments over downvotes' it's 'agreeing to withdraw a comment if it's too controversial.' (i.e. downvotes, no replies.) Not every thought needs to be voiced. Also, your feedback may have been better directed to my email, I do list one.
An example of where I recently deleted a comment was in response to this one: https://news.ycombinator.com/item?id=8569545 where my opinion was actively solicited. However, I decided that the reply to it sent the thread in a wrong direction, and also could be read without my reply. I looked instead for the person's email who had asked, but he didn't list one so I ended up deleting my comment. In effect I was saying that on reflection I chose not to comment on the linked essay in that thread.
People respond to incentives; the whole point of HN karma is to incentivize people to increase it. So I don't think we can blame people for deleting downvoted comments (when they tend to lose more karma if they don't)
how do you read 'angersock'? To me the plain reading is "an online persona distinct from one's usual persona; and one which has an extremely angry tone."
I agree there is nothing wrong whatsoever with his actual comments.
well, I've seen a lot of great posts where I wished 'angersock' would change his name. I am sure they will get visibility under a more neutral name as well, and you can always put in your profile that you had posted under angersock.
Honest question: when you read a comment, do you read the comment first, or the username? Because - and I realise it's difficult to judge 100% accurately - I think I always read the comment first, and often don't even glance at the username. And the styling seems to encourage that too.
Would it really be as crazy as it sounds, in the interest of promoting discussion above all else, to just hide usernames altogether?
I don't think they'll get more visibility at all. A name is but a sequence of letters, angersock to me does not mean more or less than logicallee and I'll read both of you with equal interest. Of course other people might have a different set of criteria but the reputation of a person should be the sum of their output and interactions, not <= 10 alpha characters picked from a limited set. And what better story than someone who enters a forum under a moniker like that and decides to change his ways. Beats the alternative!
I read an anecdote about the Swedish language today that's just too good to check[1]:
Being fluent in both Swedish and English, I know there are certain concepts, such as the difference between belief in an opinion and belief in a fact, where the Swedish language makes clear distinctions (tycka and tro) and English does not. English speakers spend ridiculous amounts of time arguing about these things, and Swedes simply don’t need to. It’s not that English speakers can’t conceptualize the difference between opinion and fact, but doing so is way more difficult in English, because the word “belief” in English is quite fuzzy, whereas in Swedish, the language makes it simply impossible to confuse the two.
HN voting seems to suffer from an inability to make this distinction: I think it's terrific when comments are downvoted because they're factually inaccurate, no matter how well they're written. So much nonsense ends up at the top of some comment threads simply because it reinforces existing beliefs or loyalties. On the other hand, if you make a good case on one or another side of some live debate, I don't think it should matter whether I agree with you or not. The discussion is always enriched by quality arguments.
> PG opened a pandora's box when he said that downvotes were acceptable when you merely disagreed with someone, instead of doing so for civility. There is no end of sadness and bullshit caused by that remark.
Since that remark came to my attention, I've put less thought and research into my HN comments.
It only takes a few net down votes to make a comment sufficiently hard to read that it is effectively dead. For example, suppose there is an article about government surveillance, which the article opposes, and the article makes an assertion about the 4th Amendment that is wrong. If you post a comment explaining the error and how the 4th Amendment actually works, some people will misinterpret this as implying that you are in favor of government surveillance, and down vote you for that.
The people who correctly interpret the comment are not as likely to up vote as the people who misinterpret are likely to down vote, and so the down voters have an advantage.
If down voting simply moved the comment toward the bottom, rather than making it almost unreadable, it would be OK. It the combination of making down voted comments hard to read and saying that down votes for disagreement is OK that is the killer. The message that sends is that comments here are supposed to agree with the majority opinion.
I would be interested in a combined content and comments link. I am often more interested in the HN commentary than the article itself, so I am always opening the comments as well. This would mirror a feature in the Reddit Enhancement Suite that I enjoy.
Is there any guidance on what stories are appropriate to flag? Is it just spam, or is it looser than that?
I sometimes flag blatant political stories, because I don't think they belong on HN, so I guess if I suddenly can't flag anymore then that was the wrong answer.
If I understand the rules correctly, downvoting is used for not agreeing with a comment, and flags are used for inappropriate comments.
But is fading the comment really the correct behaviour. Isn't the position on the page deciding whether the community agrees with the comment?
My suggestion is: remove the fading; let downvoting only move the comments down, since the comment is appropriate yet the community doesn't agree with it; and let flagging remove inappropriate comments.
Moving comments down completely can break the threading - e.g. you could have a highly upvoted parent comment with a single child comment that was heavily downvoted. You can't move the child any further down the tree.
You CAN indicative that the comment has been downvoted - the fading accomplishes that objective.
That sounds like an incorrect fix. Since the comments are separated by trees, they should be positioned and compared to comments relative to that tree, not relative to the master tree. Being at the bottom of a sub-tree marks the comment automatically as the most downvoted in that tree.
Applying fading to all the comments just because a downvoted comment has a parent that happens to be at the top, seems wrong.
Downvoting and threaded view is hard to combine. What you describe is the trivial way. It has the downside that if a comment has few subcomments, downvoting them don't accomplish very much.
Agreed. If they're this adamant about keeping the fading functionality tied to downvoting, they should just do away with downvoting all together, because fading just seems to worsen the discussion.
I've lost count of how many times I've seen perfectly good insightful comments around here that get ignored because some people decided to downvote them for some arbitrary reason. In fact, it's so bad, that I'd say I've barely seen any good instances of fading for the past year; they just seems to encourage group-think almost every time.
I'm ok with the idea of downvoting meaning disagreement, since that's how voting systems always end up being used anyway (not to mention pg approves of that interpretation as well), but penalizing comments to become practically non-existent based on a few downvotes (last time I noticed, a score of -1 started the fading) is just not conducive to interesting discussion.
Comments that most deserve to be downvoted (trolling, incivility, lies, etc) are inevitably going to garner a lot of downvotes anyway. It's only the ones which oscillate between one or two downvotes and upvotes in what may be controversial threads which most fall prey to what I consider to be the more petty examples of downvoting. The first couple of downvotes are the least meaningful, I think.
So perhaps a margin of, say, five or six downvotes should be necessary before any visual effect takes place. Not only would this no longer bias readers against a comment before having read it, but it would guarantee that the faded comments more accurately reflect the standards of the community as a whole.
> downvoting is used for not agreeing with a comment
I hope downvoting has nothing to do with agreement; I thought it was for comments that aren't valuable to the conversation (e.g., not substantive, poorly reasoned, false, or poorly communicated).
I want to see many more comments that are valuable and challenge the community (and challenge my thinking too). The groupthink is already well known and if not, will certainly be posted by someone.
In practice, I do see downvoting used on comments that are valuable but challenging. It's disappointing.
> I hope downvoting has nothing to do with a agreement; I thought it was for comments that aren't valuable to the conversation (e.g., not substantive, poorly reasoned, false, or poorly communicated).
That's a weak distinction when there is a factual disagreement, since most people probably believe that factually incorrect statements are not valuable to the conversation.
pg has said many times that it's OK to downvote out of disagreement (particularly since it's OK to upvote due to agreement.)
I generally don't unless the disagreement has another value-subtracting component (lack of substance, poor reasoning/phrasing, factual errors, unnecessary incivility.) For disagreeable but worthwhile comments, I prefer responding instead. But I recognize the value in having differences of opinion over the use of downvotes -- comments of marginal quality (which some people disagree with, and others don't find valuable enough to upvote even if they agree) will end up dropping down the page.
> pg has said many times that it's OK to downvote out of disagreement (particularly since it's OK to upvote due to agreement.)
What value does this provide to the reader? Community groupthink is not strongly related to truth, IMHO, and provides little marginal value, because I already know what it is and it's easy to find. In fact, I wouldn't read HN if all I found was the groupthink -- I gain far more from new or challenging ideas.
EDIT: When Amy up/downvotes, it is for the benefit of Bob and other readers. Amy receives little benefit besides maybe a little emotional satisfaction. If the vote mainly was for Amy's benefit, HN wouldn't need her vote to impact anyone else's threads -- we all could have personalized, independent scores for each comment. What benefit do I, Bob, receive if Amy is just voting on what she agrees/disagrees with?
Downvoting because you disagree provides the exact same benefit to the reader that upvoting because you agree provides. "Downvotes aren't for disagreeing" is an absurd redditism.
> "Community groupthink is not strongly related to truth"
In strongly technical discussions, which are common on HN and which HN should be optimized for, "groupthink" and "correct" tend to correlate quite nicely. Whenever there is a concrete notion of "correct", it works well. It is only when "correct" becomes subjective (such as in political discussions) that shit gets messy.
> Community groupthink is not strongly related to truth, IMHO, and provides little marginal value, because I already know what it is and it's easy to find.
In other words, downvote and upvote according to what you think those actions mean.
Compared to Reddit, HN has a way, way smaller number of instances of "downvote to disagree". You can't prevent it entirely because sometimes it is the right behaviour (such as someone spouting crap on vaccines causing autism, for example...), and sometimes there's an opinion people feel too strongly not to downvote but not strongly enough to reply instead.
But the amount of times I've seen controversial opinions rise to the top just because they're thorough and elaborate has been really fantastic, and I make sure to upvote things I disagree on myself. On reddit, the top comment tends to be a one-liner pun -- and again, it feels wrong to me to upvote things I disagree with on Reddit, so I don't do it and it just makes the whole situation worse.
I think HN should make voting more structured. I'm not sure why my karma should suffer if I don't have a popular opinion. Such example is low-carb diets. I always get downvoted, because most people are hooked up on carbs. Why should my karma reflect that? A niche like hackers penalizing niches is not very karmic!
> pg has said many times that it's OK to downvote out of disagreement
That doesn't mean it's a good idea. I would urge everyone (including PG) to reconsider the acceptability of this practice -- it strikes me as being harmful.
To put it a different way: PG & friends have put together a quality discussion forum. One of the reasons it remains relatively high quality is necessarily that many users of the site do not use downvotes to express disagreement.
>> One of the reasons it remains relatively high quality is necessarily that many users of the site do not use downvotes to express disagreement.
I think the reason it's high quality is because it started as a place for the kind of people interested in startups. To me that means (statistically) well educated, ambitious, interested in STEM topics, etc...
Just like in the old days of internet news before AOL made it available to the masses. For example, sci.math had awesome discussions among leading mathematicians, and then the masses came looking for help with their homework and those guys left.
Like all good forums, if it does really well to the point that the general public takes interest, it will likely become crap.
Like all good forums, if it does really well to the point that the general public takes interest, it will likely become crap.
That isn't an inevitability, it's a consequence of poor forum design. At one time, if water started leaking into your boat, it sank. Then someone had the bright idea of pumping the water out and sinking was no longer inevitable.
One mechanism that might prevent the dilution you describe is to hide low-scoring comments from low-karma users, but show them to high-ranking users. Then the high-karma users can exert influence on the low-karma users by upvoting quality comments and downvoting poor ones. It becomes a self-organizing system with a bias towards quality, rather than the open loop 'majority rules' system currently in place.
Reddit/HN create one interaction with the comment(upvote/downvote) and then expect that they must use this button to make the post more prominent or less prominent, even though the most strong emotional reaction is agreeing/disagreeing.
Its like you invite people to a debate, where you say "please only boo when someone is saying something obscene and nasty or breaking the rules of the debate, not when they are saying something you disagree with".
Downvoting for mere disagreement has been blessed by the top level; it's an HN 'thing'. It would be less of a problem if it didn't also fade your comment.
I rarely see well considered dissenting opinions down-voted into the gray. That is something that I typically only see when somebody is being uncivil, committing logical fallacies, mis-characterizing the arguments of others, or is factually incorrect yet self-sure.
>"I hope downvoting has nothing to do with a agreement; I thought it was for comments that aren't valuable to the conversation (e.g., not substantive, poorly reasoned, false, or poorly communicated)."
I agree.
Thing is, the "when to downvote" debate is ancient at this point and no closer to consensus.
These days, I'm given to thinking that any voting system should assume the behavior of least resistance, that it's thoughtless, reflexive and therefore only potentially valuable after being sanitized / aggregated / weighted.
I think you don't understand the rules correctly. Or rather, I know a different version. Maybe that is because that question does seem to not really be answered in the rules, at least I found nothing.
I was under the impression that downvoting is for comments that you do more than just disagree with. If you disagree, but it is a valid interesting argument, I try to vote a comment up, in any case not to vote it down.
Flagging is for inappropriate comment, stuff that is harder than just a bad comment, like racial insults or spam.
"If I understand the rules correctly, downvoting is used for not agreeing with a comment, and flags are used for inappropriate comments."
In theory, no, downvoting is not supposed to be used to disagree. It's supposed to be used to move "stupid" comments (inarticulate, illogical, poorly argued, etc.) down towards the bottom of the screen. Flagging, on the other hand, is meant to call out wildly inappropriate, offensive, or completely off-topic comments.
A few hypotheticals:
1) I see a comment that is well articulated and generally inoffensive, but I disagree completely with its premise or its arguments --> I do nothing.
2) I see a comment that is irrational, illogical, poorly constructed, whiny, or off topic --> I downvote.
3) I see a comment that is extremely offensive, completely unrelated to the topic at hand, spammy/astroturfy, or clearly intended to instigate a flame war --> I flag.
That said, many people seem to use downvotes to disagree with a comment, and I'm concerned that lowering the downvote threshold will lead to even more of that behavior.
Maybe the problem is that downvoting and flagging share too close of a hypothetical use case. Downvoting is clearly easier to do, and it's doable in the main thread view. So that could be why people routinely use it to disagree with comments. I have flagged maybe a grand total of one comment in all my time on HN, and I generally regard flagging as an option of last resort. (It could also be that I arrive at various threads after the worst stuff has been flagged and removed, providing me an artificially sunny view of the threads).
That's a reasonable take on what downvoting is for, but the HN take is that it's better to downvote a disagreement than to write a noisy bad comment to express the same disagreement.
Interesting. At first I thought, wait a minute, isn't "don't downvote to disagree" in the HN Guidelines? Then I checked, and no, it isn't. The guidelines talk about downvoting ("downmodding") very briefly, and only to say that one shouldn't complain when it happens to him or her. There are no guidelines on how or why to downvote.
Maybe this is a problem, in and of itself? Seems different people have different takes on the use case for downvoting.
Gotcha. I'll take your word for it, because I regard you as about as much a lay authority on HN posting as HN has. And if pg has weighed in on this topic, well, he's quite the authority, too. :)
That being said, I still feel weird about downvoting to disagree. The way I see it, disagreement-downvoting creates noise in its own way. Less noise than unproductive or bad replies, certainly, but more noise than doing nothing.
"The simple way to metabolize this system is just not to get unhappy when you get down voted."
True, and this is also an official guideline. (A good one, too, IMO).
> is that it's better to downvote a disagreement than to write a noisy bad comment to express the same disagreement
No, quite the opposite. (See I wrote a comment explaining that I disagree, I didn't just click the arrow. Much more valuable to you and the reader.)
And that's why you are wrong, by saying why you disagree you might (rarely) change the posters mind, or more likely change a 3rd readers mind about the topic.
This isn't my argument, but you're missing its point. Think of disagreement downvotes as a pressure release valve for the downvoter. The idea is to prevent you from expressing your disagreement in an unproductive comment.
There is no particular purpose for HN downvotes other than to express dislike. It's not needed to move 'stupid' comments down, because only 'non-stupid' comments rise up. The comments which are never voted on have no intrinsic value, so they might as well be stupid anyway.
I wasn't even aware of flagging on comments until someone mentioned it, and to be honest it's so inconvenient i'd probably never do it. Downvote is an easy enough way for me to say "I don't like this" and it helps reinforce the group dynamic, which is what they want (this whole post is about manipulating group dynamics).
The only way to fix downvoting is to put an express purpose behind it, like a button that says "I disagree" next to a button that says "I dislike" next to a button that says "Inappropriate comment". But for some reason forum admins always prefer generic feedback which give rise to unexpected behavior. Heh, it probably keeps people upset, which reinforces user engagement.
Maybe I'm in the minority on this, but I don't approach voting from an agree/disagree axis. I approach it from a productive/unproductive axis. I upvote comments I find helpful, articulate, or informative. I downvote comments I find unhelpful, inarticulate, or uninformative. I leave alone comments that I find "meh," or that someone has already said better, or that neither add to, nor subtract from, the conversation. Basically: something either adds to the conversation (upvote), subtracts from the conversation (downvote), completely derails the conversation (flag), or is neutral.
Case in point: I didn't necessarily agree with your comment, but I found it articulate and interesting. It presented a perspective I hadn't considered. It made me think. I upvoted it.
Now, one could argue that agree/disagree and productive/unproductive are basically the same thing in practice. They both represent a desire to move the conversation in one direction or the other. (To your point about the group dynamics). But I don't know if I totally buy into that. I have no idea if this is true, but I'd hypothesize that voting patterns would be different under one explicit guideline or the other. I'd also think that downvoting-to-disagree runs the risk of fostering an echo chamber, which discourages dissenting or different viewpoints, particularly on hot-button topics.
>There is no particular purpose for HN downvotes other than to express dislike. It's not needed to move 'stupid' comments down, because only 'non-stupid' comments rise up.
The problem is that smart-but-unpopular comments may fail to rise up. We want to prevent groupthink by ensuring that the top comments don't exclude ones that are well written and productive while expressing a contrary viewpoint.
Well first of all, I disagree with your assertion that top comments with contrary viewpoints might prevent groupthink. Sometimes being contrary is inherent to groupthink. And being well-written and productive doesn't mean the comment is insightful, intelligent, or informative, which one assumes would be a preferred top comment. (personally I think promoting comments that are well-written, productive and contrary just gets you a class of people who can bicker in a very civilized manner, like two English gentlemen having a polite row)
Why do we vote in directions? It would appear that according to social media, we only have one orientation in voting (vertically) and can thus only go in two directions (up and down). But this makes no sense. For one thing, the 'direction' of a comment/conversation may go many different ways... conversations do not go in a straight line. And the idea of 'elevation' in a conversation is also confusing: if anything, a conversation would be more three-dimensional than two, moving around a changing landscape.
In addition, 'up' and 'down' gives an immediate subconscious reaction where 'up' is associated with goodness (and therefore correctness) and 'down' with badness (and therefore incorrectness). I say this moral implication is the default because it makes no sense for only one direction to be correct, unless you needed to know where you were supposed to go, and your default assumption was that you should never go down. (That would be a bad presumption for airplanes, for example, as they would never land until they ran out of fuel)
So what does it mean when HNers 'vote down'? Are they agreeing with something, or disagreeing? Are we assuming they're trying to impose a nuanced influence on a particular part of a person's commentary? If it was a generic part of their comment, was it their moral or ethical stance, or their writing style, or the correctness of their comment? We have no idea; all we know is they wanted that comment to go 'down'.
A simple vote down, regardless of its supposed intent, doesn't provide any specific benefit if it comes without context. The lack of a down vote would do the opposite, however: it would force people to only be able to provide positive feedback, or no feedback at all. How does this effect group dynamics? It actually works on individual behavioral conditioning: provide positive feedback, and you get positive behavioral conditioning, and negative behavior gets weeded out automatically (because the individual wants to be rewarded, which it knows it only gets for positive behavior). This is what practically all dog-trainers and child psychologists suggest as the best method to promote positive behavior. This would have two similar effects: people would be nicer, and they would provide 'better quality' comments.
So I think 'downvoting' is a loaded and useless way to regulate comments. I think that valid alternatives would be votes for specific feedback, or if you wanted to provide a way to simply "move thread position", a button that said so specifically. These would create specific, immediate changes in the group dynamics and the interactions of the users, unlike a generic downvote button. If you're trying to prevent groupthink, give the users tools to allow them to express themselves better.
>Sometimes being contrary is inherent to groupthink.
And lack of being contrary, is more conducive to groupthink.
>And being well-written and productive doesn't mean the comment is insightful, intelligent, or informative, which one assumes would be a preferred top comment.
Only if you're using a non-standard definition of productive. Could you spell out what you mean here? In common use, a productive discussion is one in which at least one side emerged with a better understanding of the issue. (If you can do that without being insightful, more power to you!)
This contrasts with a discussion in which arguers blur important distinctions, talk past each other, and so on.
It may be helpful (productive?) to give an example of (what I consider) a comment that is insightful, informative, and correct, but unproductive: https://news.ycombinator.com/item?id=7840204
In this case, it's because the poster simply reiterated the logic behind the point his critics already accepted, thereby doing nothing to convey understanding of why they should take his position.
So, I don't think you actually disagree that "productive = good" here, but have non-standard criteria for what counts as productive.
What's the difference between a super stupid comment and a comment which gets no votes at all? I'll bet you the answer to that is "I really didn't like it", which mostly means emotional value. So the purpose of the downvote there is to basically have an emotional reaction button. Take away the downvote and you take away the emotion from the voting. Then all you have is upvoting, which is based purely on merit. It sounds like people don't like the idea of forum meritocracy, though.
> "What's the difference between a super stupid comment and a comment which gets no votes at all?"
Feedback for the comment writer. Shows the difference between "nobody thought much of my comment either way" and "some people really disliked my comment".
I think it's been very effective at weeding out the completely pointless comments (memes, jokes, "this", "I agree", etc.) and is also somewhat effective at weeding out subtle incivilities (I've changed my behavior in response to downvotes before.)
Some comments degrade the quality of the conversation. We might disagree about which ones, but surely we can agree about that, right? The obvious candidates are purely asshole comments, trolling comments, intentionally racist comments, etc., and there are less obvious ones too. (Almost any joke where the author feels a need to reply, "guess no one around here has a sense of humor" after being downvoted probably fits.)
The discussion is better if those comments are lower on the page and, ideally, discouraged from being made at all. Downvoting does both.
> Then all you have is upvoting, which is based purely on merit. It sounds like people don't like the idea of forum meritocracy, though.
I don't understand what this means or what a "meritocracy" is in this setting. (The moderators are in charge; the highest ranked users will never form a ruling class or be in power. Most forums are not "meritocracies", "socialist exeriments", "democracies", "plutocracies", "oligarchies", "monopolies", or anything that implies something other than a loosely managed commenting system.)
HN has many readers that know an absolute shitload about different subjects, and I'd like them to share interesting information and anecdotes about interesting articles. I don't want to read stupid jokes and memes on HN, because there are other places that are better for that. So I like that there are ways to explicitly encourage certain comments and explicitly discourage others.
> Some comments degrade the quality of the conversation. We might disagree about which ones, but surely we can agree about that, right?
Yes.
> I don't understand what this means or what a "meritocracy" is in this setting.
Basically, thread sorting would be based (mostly) on merit.
> the highest ranked users will never form a ruling class or be in power
The highest ranking users have high social power. Their power is simply a tendency to steer the group without telling anyone what to do. Usually it's unintentional, and sometimes they have no idea they're doing.
Most forums are actually discrete social communities made up of ingroups, with their own cultural standards and hierarchies, and often a few custom heuristics. Some of them are really incredible from a behavioral standpoint, like an internet chat galapagos island. HN is not so special, but it does have a clearly defined social structure.
I agree with you that the stupid jokes, memes etc should stay off HN. That's why I think some kind of meta-voting is necessary: because a simple downvote is used for a lot of things right now, and not just "this-comment-is-out-of-bounds" like you're talking about. Barring meta-voting, removing the downvote keeps the troll talk but it still filters down to the bottom with the other useless comments.
As far as I can tell, there are no guidelines or FAQs about what downvoting should be "used for" on HN.
On Reddit, the Reddiquette [1] standard is that you should only downvote if a comment doesn't contribute to the conversation. I think is is an excellent standard. It's been suggested in the past that a similar standard should be present on HN. [2]
Considering that flagging is a privilege that can be lost if someone decides that you're abusing it, ideally it should be reserved for the egregious cases that are unquestionably bad. Downvoting on HN should be for comments that truly don't contribute to the conversation, though the comments in question may not rise to the level of flagging.
As an example: Say you have a comment expressing an opinion with no support, no references, and factual errors. Is it "inappropriate"? Arguable. Does it contribute to the conversation? Not in any positive way. I would suggest downvoting that comment would be the right answer, even if you agree with the sentiment, because it doesn't positively contribute to the conversation. But flagging wouldn't be appropriate considering the comment is on topic and not spam (which are given as criteria in the HN Guidelines).
If you encourage downvoting because you disagree with an opinion, then minority opinions get buried. And sometimes it's useful to see that an opinion isn't unanimous, and that there's another side to an argument, even if you disagree.
I think disagreeing with a comment is not enough to merit a downvote. I'd rather see downvoting as a signal to the author about the quality of their post - while flagging is a signal to the system itself in cases where the post is unacceptable and should be removed from the site.
When I disagree with a comment, I'd rather write a dissenting one or a reply stating my counter points. Another method of dealing with disagreements is to simply upvote other comments you agree with.
Using a downvote to express disagreement with somebody here is going to come off as highly inflammatory towards them because downvotes actually cause their comment to fade out, as you mentioned. In fact, there have been instances where I've upvoted comments that I didn't agree with because they were so well articulated.
I believe that pg himself disagrees with me on this, but I honestly don't think the site is designed correctly with his view of voting in mind. The punishment for downvoting shouldn't be so harsh (both moving the comment down and fading it out) if it's just to express disagreement. That just encourages extreme group-think.
By far the best way to express disagreement IMHO is to neither vote nor flag, but to post a response.
> By far the best way to express disagreement IMHO is to neither vote nor flag, but to post a response.
I sometimes wish downvoting comments that have no response was impossible period, but I have no idea how to deal with the edge cases. Still, it's frustrating to see responses (actual ones, not just a bunch of insults etc.) made in obviously good faith greyed out without anyone pointing out what is wrong with them. When I do know, I try to reply, but usually I don't feel quite qualified. But the person who downvoted should know, right? If they can't put it in words, can it really be that good a reason? Among people who mostly type very fast, mostly idly browsing this site instead of being in a hurry, the time it takes is not an excuse, either.
To me it says someone has enough time to care about something someone said being somehow wrong, but not enough time to tell them what it is; that somehow the fact that they said something wrong makes them unworthy of being told what was wrong about it. Oh, and that nobody else who doesn't "get it" doesn't deserve to learn, either.
> If I understand the rules correctly, downvoting is used for not agreeing with a comment ....
Certainly not.
Up & downvotes determine what is shown on the site, and what is prominently featured. Is your goal for this site for it to be full of things you agree with? I hope not. That's not my goal.
Now, there are varying opinions on the issue of downvoting for disagreement. Regardless, it is certainly never officially stated that this is the purpose of downvotes.
>>I only do it for stories which I know will only produce flamewars.
You shouldn't. For example, stories about big players (e.g. Apple, Google, Microsoft, Facebook) almost always produce flamewars, but the discussions can also be interesting and informative.
You should flag stuff only when it doesn't meet guidelines. For example, most news stories (e.g. "Man murders family, followed by car chase") are inappropriate for the site.
I used to do that, but I had my flagging privilege revoked for a long while. So don't do that, I guess.
It's restored now. I don't think I've flagged many, if any, stories.
I flag any comment I perceive as an attack on a person rather than the ideas of that person. I don't feel that's appropriate. And in particular it's not what I personally want HN to be.
If a comment makes a point in what I feel isn't a good faith manner, I'll merely downvote it.
If I just disagree but it's a fair point, I leave it alone.
In my opinion the only real annoyance with HN is when story titles are edited.
Maybe it will take an updated policy to fix the core issue, which is that while linkbait headlines are annoying, some original works are titled very badly and the poster may actually add useful information by customizing the title (often by highlighting what is most interesting/relevant about the linked article).
I think this aspect of HN has improved immensely in the past several months; dang in particular has done much to increase the transparency around title editing. I find myself less likely to knee-jerk reject a title change when the rationale is right there in the comments.
An earlier draft of this post that Sam sent me had "nearly everyone can help there". I'll ask him to put it back. But it's not like this is much of a barrier. Even this one comment moved you significantly (as of right now at least).
Just curious, but are you saying that because of the relative position of the parent's comment in this thread, or is there a threshold for karma that unlocks seeing the karma on comments other than your own?
I'm wondering if there's any official policy on linking to articles that are behind pay/register walls. I've flagged two such submissions, and only one got flagkilled.
There are a few reasons such articles usually do not get flag killed.
1. Most of the paywall news sites allow free viewing of a few articles per month and so many here do not hit the paywall.
2. It is often very easy to get around the paywall if you are over the monthly limit. Sometimes it is as simple as searching for the article by title in Google and clicking Google's link.
3. If you cannot get around the paywall, it is often the case that you can find coverage of the same event at another site that is not paywalled, which will give you enough information to meaningfully enjoy and participate in the HN discussion.
4. Even if you cannot find free coverage of the thing under discussion, it is often the case that the HN discussion is entertaining and enlightening. HN discussion often goes off on tangents inspired by the article, and those tangents can be appreciated without having read the article.
5. Many people here don't see the paywalls because they subscribe to the paywalled sites.
6. Most here do not have such big egos or senses of entitlement that they are bothered by others reading and discussing something that they have not been able to read.
I still read HN. There's a good article or two per day. That said, the moderation needs work. (Or, more accurately, to be put out of work.)
I've been under "rankban" for at least a year, if not longer. This means that my posts, even if they get a lot of upvotes, fall to or near the bottom. I don't really care, and while I could use sock puppets to get around that, I'm too old for that shit and I don't really care. Still, it's irritating.
I've used HN Search (https://hn.algolia.com/) to verify comment karma and validate my suspicion.
I'm also on "slowban", as Hacker News performs worse when I use it while logged in than when I do so in incognito, logged-out mode.
It's dishonest bullshit. My main reason for continuing to comment is (a) to spite the moderators and (b) because there's a certain joy in getting +10 karma even when your post falls to the bottom.
Even if this is true, I doubt his comments are undesirable enough to passive-aggressively try to get him to stop using the site by silently slowing it down (if his claim to be on slowban is true). This kind of behavior seems like it should be reserved for trolls and other highly noxious personages.
I obviously can't speak to whether MOC is experiencing it, but I can say that with my first HN account, before I understood the context and spirit of HN, I made a very Reddit-esque comment that got me banned. The first part of that was the so-called slowban, as the site was all but unusable when I was logged in.
This was years ago and I have no idea if the mods still use this tactic, but I imagine so in certain cases.
Which makes no sense because michaelochurch has a high comment upvote average and is one of the more prolific commentors. Even if the comments are off-topic, a high average suggests people like them.
They will absolutely put high-voted comments at the bottom of the page. This happened to me when i posted a post with a lot of constructive criticism about a product i wanted to pan out, but which had a lot of problems in countries other than the original country (to the point of threatening human lives). Others agreed, even the makers of the product were happy about the feedback, upvotes happened. Yet my comment is dead last on the page.
I mailed the moderators and one of them confirmed it had been marked "off-topic", and left it there, despite admitting it wasn't off-topic, because he did want to cut down on "negative" comments on new products.
I understand the stance, but disagree with it, and have taken it a fact of life that both the moderatorship, and the userbase at large, are unwilling to consider strong posts that aren't full of praise as worthwhile. I can even mildly understand that, since it takes more effort to verify strong criticism as being grounded, than it takes to verify strong praise. It's disappointing nevertheless.
That sounds like awful moderation - it isn't moderation's role to effectively do marketing for a product. Any good moderator would tell another to stay hands off as long as people are civil.
> because there's a certain joy in getting +10 karma even when your post falls to the bottom
It's possible that the ranking algorithm is actually helping your karma count. I personally only up/down vote comments when I feel they're in the "wrong position". If you made a comment which I found good and it was at the bottom, I would up vote it whereas I probably wouldn't bother if it was already at the top.
'I'm also on "slowban", as Hacker News performs worse when I use it while logged in than when I do so in incognito, logged-out mode.'
While it's my impression (from other comments, &c) that your claim is accurate, this alone isn't necessarily proof. I would generally expect a site to perform better for logged-out users, as 1) it gets to skip any tasks that are for logged-in users only, and 2) it should tend to have everything it needs to render that version hot in whatever caches might be involved...
Like I said, that's my impression. I'm just saying the differential alone doesn't show it. A tiny differential is expected. A large differential shows either a slowban or poor engineering.
It seems like early comments on any article generally have a huge advantage in garnering votes.
This appropriately rewards engagement, but also pushes down other worthy comments that come slightly later, often to effective near invisibility when replies to the top comment dominate the discussion.
I wonder if there is any mechanism that could help counter this.
I expect that there is a strong correlation between how soon a comment is posted and its points. Two thoughts from a relative noob (sorry if these ideas already have been discussed or tried):
* On threads with many comments, few users have time to read all or even most of them. Perhaps find a way to shorten the list, improve signal-to-noise, and surface newer comments by burying older comments that haven't received votes.
> This appropriately rewards engagement
* It rewards engagement by people who have the time to frequently check HN, read linked stories, and comment throughout the day. My guess is that the smarter someone is, the less availability they have for such things. I would create a system that rewards engagement by busy people.
> On threads with many comments, few users have time to read all or even most of them. Perhaps find a way to shorten the list, improve signal-to-noise, and surface newer comments by burying older comments that haven't received vote
So essentially the same kind of system that is used for the link page? I honestly don't see why link aggregators use one system for links (based on time posted and points) and another system for comments (based solely on points). A very engaging link and a very engaging comment both share the same qualities, whether it is informativeness or adding something interesting to the discussion.
That is a cool idea. It would have to be done in a really subtle way to avoid becoming garish, but I hope it gets tried out. Not sure about encouraging reloading the page though.. maybe some kind of AJAX would be better for protecting the server from overload.
A comment's location is affected both by its points and by its age. New comments tend to appear at or near the top in the first few minutes after they're posted, and then will drop down the page if they don't receive many upvotes.
It would be cool to see markdown support, but I totally understand that any addition to the formatting spec is risky, because any new change means older posts could be awkwardly rendered.
On that historical HN note, it would be great if someone made an app that dug up really awesome older stories from the HN archives. I'd use that a ton!
They could do something like a BOM. Any comment without the wchar in position [0] is rendered historically. Anything with it is rendered using markdown (or another renderer).
It adds no new columns, and better still if you pick a symbol which is nuked by TRIM() e.g. \x0B [U+000B] (vertical tab) then it might be ignored by existing infrastructure/trimmed.
> To prevent abuse, moderators review flagged stories and comments and revoke flagging privileges from users who flag inappropriately.
This seems like one of the cases where HN's moderation is both draconian and completely opaque. I can't flag comments, and I have no idea why. I certainly don't think of myself as an abusive user.
Maybe I accidentally clicked the "flag" link once. If that gets flagging privileges removed, HN should at least consider creating a way to undo an action!
Are you sure you are looking at the right place? To flag a comment you must click on "link", and only then the "flag" option appears. TFA mentions the threshold for flagging is only 30 karma.
This has been the source of much confusion and I'm not 100% sure on the reason why it was implemented.
There is a karma threshold (iirc, I know there is one for downvotes) to be able to flag and a link threshold that if you don't know where to find it, you won't use it. only the experienced HN people will 1. care enough to use it and 2. have the karma for(old members) it so it's a low pass filter of sorts against poor commenting.
More than once, I've unintentionally downvoted comments I meant to upvote, upvoted comments I meant to downvote, and/or clicked on a button I didn't mean to click (that includes clicking on 'reply' after writing something that doesn't really add any value and almost instantly regretting it).
So I agree, it would be great to have more/better undo options!
I've done this, too. If you e-mail the mods they'll reverse it pretty quickly. On the same hand, if you e-mail the mods all the time they might have more of an incentive to implement a self-undo.
Thanks for the tip. I'll consider doing that next time, even though I'm somewhat reluctant to try it, because I have a sense the mods already dedicate a lot of time to HN -- the last thing they want or need is a barrage of emails asking for essentially unimportant reversals.
I actually don't agree with this from a UX perspective. The downvote arrow is currently in a predictable position. If you move it after the timestamp, it will be in a variable position. Although, I can imagine how you could argue that the downvote arrow should not be easy to hit.
I guess in addition to raw usability perspective, how you view downvoting in general has an impact on the layout. If you want to place downvoting on equal footing with upvoting, placing them together does make some sense. If, however, you feel downvoting should be a more rare occurrence, it makes some sense to relegate downvoting to a worse position, especially in light of the seemingly frequent issues with accidental clicking.
I also find the single leading upvote to be a bit more consistent, as you are not able to downvote replies to your comments, only upvote. So in that regard, the UI would be a bit more uniform regardless of whether the comment was a reply or not.
Another straightforward possibility could be replacing the arrows with words, and making the time a link to the comment (removing "link" from trailing actions as gojomo suggested). This may result in better support for screen readers, in addition to mobile users, as well.
etcet 5 minutes ago | upvote | downvote | flag | parent
As other users have pointed out, to flag a comment you have to click on 'link' to go to the comment's item page [1]. I don't know why PG designed it this way but have always assumed it was a speed bump to reduce impulsive flagging, since flags are more powerful than downvotes.
We never remove flagging privileges because of just one flag. The concept of a mistake is all too familiar over here.
> We never remove flagging privileges because of just one flag. The concept of a mistake is intimately familiar to us, I assure you.
Is it possible to either allow people to upvote a comment they have previously downvoted (or vice versa), or at least to undo an accidental upvote/downvote?
Especially on mobile devices, the margin of error here is very low, and I know I have accidentally downvoted a comment that I meant to upvote.
(In addition, I know some people may upvote/downvote a long comment before they have finished reading the entire text. This probably isn't behavior that should be encouraged or catered to, which is why allowing people to change their mind on votes[0] may be overkill. But at least allowing people to revoke an incorrect upvote/downvote might mitigate this effect.)
Is there any way for me to tell if my flags are useful or stupid?
I tend to pre-emptively flag stories which I know will draw out the worst bits of HN. That's probably a bad thing, and I wouldn't be surprised if it caused my flags to be ignored.
I have only just become aware of flagging comments. Some of the threads tho' you would need to flag 100 and a lot of bad feeling is generated. Best nip it in the bud early I reckon.m
The threads that you are flagging are the ones that need to happen.
I can see why you would flag a whole thread, but I don't think that will ultimately get us anywhere. If you care about the accessibility of this site, then I implore you to upvote those threads^ and flag the bad comments within.
> Is there any way for me to tell if my flags are useful
Anyone worried about this is welcome to email hn@ycombinator.com and ask, and we'll be happy to take a look for you. But I feel like the kind of people who would worry about this probably tend to be the kind whose flags are useful.
I feel like the UI should be consistent dang, because I've been flagging entire threads where there's an inappropriate comment (not knowing that there was a comment flag there).
I think at the moment it's too easy to hit that flag button on the front page (as others have said, especially on mobile), so maybe it should be moved back too?
The consistency would be to remove the flag button from the front page, but clicking on a story and flagging it in the thread itself is equivalent to clicking on a comment to flag it in its page.
It's not great UI and it's made worse if, like me, you mostly read HN on a tablet so hitting the wrong link is easy to do. A better speed bump would be to have the flag link always available, but have it take you to a page that says "you are flagging the following item: <comment/link> by <author>. You should continue if <list of reasons> and not <list of bad reasons>. [OK, flag it!]"
This (a) draws attention to the exact thing you are flagging (the individual comment/link, not the thread) and (b) puts the confirm link somewhere else on the screen so you can't double-tap the same hotspot by accident. The advisory text is optional, but as others have pointed out, it's not really clear where the line between flagging and downvoting should be drawn.
There was one time when I accidentally clicked on 'flag'. I meant to click on the 'comments' link, but I was on my phone and the link was too small so I ended up clicking on 'flag'. I couldn't find a way to undo it. It seems unfair if my flagging rights get revoked because of that.
I'd love to see posts on the rationale between different design decisions, particularly with regards to things that change group dynamics. For instance, putting karma next to someone's name may produce leader-following social behavior, meaning your high-karma users would reinforce general behavior by encouraging other users to behave the same. But that would increase the already prevalent concern of high-karma users overly influencing discussion and getting upvoted based on karma alone. It would be interesting to see all these examples stacked up and then a post about why they went one way or another.
> First, I want to thank the community for all the work people have done over the past six months to downvote, flag, and comment on content that doesn’t fit the site guidelines
Sam & co, great job on experimenting and improving quality, it shows. I find this statement to be really confusing considering there is no guidelines[1] for downvoting and, in my experience, it is highly used to put opinions you don't personally agree with out of reach for the majority of users to read.
"To prevent abuse, moderators review flagged stories and comments and revoke flagging privileges from users who flag inappropriately."
When I first joined HN, I aggressively flagged stories and comments that didn't meet the guidelines. I thought that's what I was supposed to do. However, one day my comment flagging privileges disappeared with no explanation. If someone had contacted me and explained how I was misusing flags, I would have happily self-corrected.
I once had my flagging privileges disappear. I e-mailed the moderators, and they restored them.
It wasn't clear either to me or the moderator what had resulted in the loss of privilege. On the one hand, that made it easy to decide to reinstate the privilege; on the other hand that made it hard to learn what I'd done wrong.
Would you mind clarifying? The blog post (quoted by parent) seems to explicity say that flagging privs can be revoked, which directly contradicts what you are saying in the comments. Or was the quote from the blog post only meant to refer to flagging of articles?
My apologies—I wrote that in haste and I see now that it was unclear. What I meant was that you can't lose comment flagging privileges while story flagging privileges remain intact; the software just doesn't do that. Either all the flag links would disappear, or none would. So there must be another explanation for what sciurus described (as I understood it). Does that make sense?
Thanks, dang. Your response to sciurus makes more sense now. Though, his description was lacking enough in detail that I suppose he could have also lost story flagging descriptions.
But I am still confused about the apparent contradiction with what you've stated versus what was stated in the blog post, about losing flagging privs. When flagging privs are lost, are _all_ flagging privs lost? Or is minimaxr correct[0], that it works the other way, where you can lose story flagging privs without losing comment flagging privs?
I need to check that. Could you please email hn@ycombinator.com if you don't get a reply here soon? I don't intend to forget, but I have a lot of packets dropping at the minute.
Edit: Ok, I checked it. Story and comment flagging privileges are lost or retained as a set. You can't have, or lose, one without the other. It's not clear why minimaxir thought otherwise, since I'm pretty sure the code has always worked this way.
Can you mark comments with the score of 1 in some way?
I suspect that I'm not the only one who downvotes the comments he doesn't like or disagrees with, but who'd very rarely want to push a comment into gray. I mean, it's one thing to vote a comment down so that it won't float at the top and another is to punch its author in a face with a negative score.
That's a good idea. Like so many of these, it isn't obvious how to do it in a way that doesn't complicate HN's minimal UI. But it's worth noting that you'd get this for free if we implemented vote undo, which is something we're open to.
How does HN's moderation staff feel about users who are dead, completely oblivious to it, but who continue to post, sometimes for years after the fact without any kind of warning that they are wasting their time.
Sometimes I check the comment history of these dead posters; 80% of the time it appears like they are a genuine troll who was correctly moderated out of the community, however, a significant portion of the time, there doesn't appear to be anything particularly inappropriate in the comment history, but the user is still doomed to waste a considerable amount of time attempting to contribute to this board.
In the context of internet discussion boards, it seems a little harsh.
Off the top of my head, I feel like something akin to an automatic probationary resurrection period would be an interesting idea. Perhaps it occurs one year after hellbanishment, giving the community a chance to organically reevaluate the user's quality of contributions, and giving a second chance to those who have matured in a years time (or who was simply hellbanned in error).
I had this happen to me when I critiqued a YC backed company. Had a few thousand in 'meaningless internet karma points'.
You pretty soon realise after a few further comments anyway, so I doubt it's actually effective.
It's pretty mean spirited though - generally a "douchebag" move. It'd be like having a bad employee, but instead of firing him, or discussing his work, just don't bother paying him any more.
After being hellbanned I actually realised that commenting on internet forums is toxic, generally a waste of time, and not productive. So I quit. At the start, you sort of care about "karma". But you end up realising it's a measure of two things. 1. How much time you waste commenting on the internet, and 2. How much you can agree with the groupthink echo-chamber.
But for those that like spending time commenting, I would have thought following a subreddit model would work better here, with the growth. Spread the power out. If I get hellbanned from the "rust is the future!!!" subhackernews, big whoop.
It'd also help with filtering out all the non-interesting (self driving cars) stories.
But then perhaps it'd basically be reddit at that stage which would defeat the point...
Ever notice how people who make claims about why they got banned never provide links to the posts in question? That's because their claims are nearly always false. If users could look at the actual record, their perennial sob story of perfectly reasonable behavior struck down by bullying censors would evaporate. So they make new accounts and post linkless statements designed to be unanswerable.
We don't ban people because they "critiqued a YC backed company".
Edit: Some of the replies have made good points, and I realize that I overreacted. Sorry about that. Please shoot us an email at hn@ycombinator.com if you feel your account was banned unfairly. We're always happy to look into this—there's no question that we make mistakes; the most I can claim is that we're eager to, and do, correct them when they're brought to our attention.
That was before I started working on HN, so I don't have any inside details. But I'll give you my take on what I think happened, based on the time I did spend working with pg on the site. I need a few minutes though--there's another comment pending on my stack. Will come back and edit this.
(Edit: I haven't forgotten, but have to run out to an appointment, so this will need to wait for another hour or two. Sorry about that.)
Edit: back now. My gut feeling is that PG might not have had enough time to look into all the details. I say that because June 2012 was about the peak of when HN moderation was extremely time-constrained (I started a few months later) and the only option was to enforce the guidelines generically.
It's cool. I'm just curious about your take since, by necessity, you get to see far more comments than I do and get to see patterns that I may miss completely.
For the record, I don't think he deserved a hellbanning, but then again, this was a while ago. And the wild west hadn't probably settled yet.
> The guidelines (as you probably already knew) also say that if you have a question about moderation, send us an email instead of posting about it on the site.
So it sounds like got banned for refusing to follow the guidelines?
Please don't post on HN to ask or tell us something (e.g. to ask us questions about Y Combinator, or to ask or complain about moderation). If you want to say something to us, please send it to hn@ycombinator.com.
I think you could, when Paul Graham was actively moderating the site, get hellbanned for unproductively pissing Paul Graham off. I think it happened more than once.
Maybe you don't personally, maybe you don't as a group anymore.
As I suggested in my original post though, I have more useful things to do than look back at comments I posted years ago and compile evidence. I've moved on.
I'm sure it happens a lot more often than you realise, because in general when people get hellbanned, they come to the same realisations as me.
Not only did we never ban people for criticizing YC companies, the very first thing PG told me when I started moderating HN, and the thing that he emphasized most strongly after that, was never to do things that could be construed (or misconstrued) as censorship of anti-YC stories.
(I've edited out some irritation that leaked through in my original version of this comment.)
dang, I think HN has improved for the better, especially since some effort has been made to make things more transparent.
But it would be despicably dishonest for you guys to deny that routinely you guys do things here to protect YC companies (including manipulating voting points on comments/ submissions).
Even though Skeletor made that comment like ... 20 days after the HN submission, it somehow found its way to the top. Obviously, this was through manual action. Obviously, a non-YC company is not afforded such a privilege.
Idling in #startups of freenode (unofficial HN channel), I've heard too many of these stories. The stories of rankban upon some critical comment on a YC-funded company, a slowban because of a critical comment on some YC personality, etc. etc. There are countless examples.
When these things happen one after the other, you lose trust, we cannot believe you anymore. Please stop doing this. I think the only way to win trust back at this point is if you again expose voting points at all times.
I'm glad that you agree that HN has been getting better. We obviously have a way to go to win you over, but challenge accepted, we're absolutely willing to try.
It isn't hard to deny that general impression you're reporting, because it isn't true. But I don't see any way to refute such a sweeping claim convincingly. As far as I can tell it exists only on the level of rumor and is unanswerable. But I'm happy to reply about specific cases.
In the Drchrono case, we got an email from the founder asking to post a response after the commenting window had closed. I agreed, but not because this was a YC startup, but rather because we would do this for any startup in this situation, and have indeed done so for at least one non-YC startup I can remember. Before I agreed to the Drchrono founder's request, I told him I needed to make sure that we would do it for any startup in that situation, and I thought long and hard before concluding that was true. It was by far the most important factor in that decision.
Consider that the post was killed by moderators, then brought back alive, then drchrono individuals made a comment, and then that comment was shot up. There were a slew of submissions crying out "censorship" this day and they were all similarly killed. I can recall like 20 such instances of similar happenings involving YC-funded startups, and having a similar situation around them. So, honestly, dang, the plausibility of the chain of events you've listed from your side in this specific instance is tenuous at best, and unfortunately again, dishonest and deceitful at worst. But I'm going to suspend my tingly senses and give you the benefit of doubt at this moment, and not go on any further about this particular issue.
dang, I understand that you, in your position, have to be mindful of optics, you have to think of ways to say things that are best for YC. That's great, you should do that. All individuals of a company who have a public presence have to do that. The thing is, you must not focus entirely on optics -- you absolutely must in your heart have the right view. Not only because if you don't, some people will eventually find out what's up, but also because it's the right thing. Actions that favor YC companies (beyond a certain line) on a public forum such as HN are unethical. I think there have been enough things done at this point that the only way that trust can again be restored is by having more transparency -- for example, by showing comment scores at all times in some way.
This is an unsolvable problem. As good a job as you're doing, YC is the sponsor of HN and just because of that there will always be this perception. The only way to get rid of this issue would be to completely divorce YC and HN, something that may for a variety of reasons not be possible and for a whole pile of other reasons not be desirable.
You're doing an absolutely super job, probably far better than most or all of us here will ever realize simply because moderation when done properly is all but invisible so don't sweat it, this is a thing that is as far as I can see not solvable in the current set-up. Those lines were drawn long before you showed up and within those lines you're doing the best you can.
FWIW I too recall several instances of users that were banned imho unjustly as well as some threads where the pro-YC bias broke through but over the vast amount of content generated here those are very very few instances, not by far enough to claim systematic bias or to be used as evidence for some nefarious plot. More like genuine mistakes and things done in the heat of the moment. And on later reflection some of those were reverted.
(If you want I can probably dig them up for you but you're busy enough as it is.)
Skeletor's comment is a first-party report of the eventual resolution of the issue. It's the most substantive top-level comment on the page, so it belongs at the top.
We want to see such comments highly ranked regardless of YC affiliation. If you post such an important update and don't get organic upvotes because the story is old, plead your case to hn@ycombinator.com.
For what its worth, validity of the hellbanning aside, I do agree with the points he brought up re: emergent problems in scored boards. The karma system does tend to contribute heavily to the echo-chambery nature of many of these discussion forums; and while I certainly think HN is in a much better state than any others that come to mind, it's not immune. (Mentioning because to me, this point is far closer to home than the actual reason for the ban, yet often seems to go without effective discussion of possible remediation.)
Maybe sorting options would help? I'd prefer to sort responses by time (and maybe some javascript to mark new comments since last visit on the page, but I dream..), or even sorted randomly because why not. Maybe serve that from a static cache that only gets refreshed every 5 minutes or so? I would still use it, gladly.
I think account-level filters might alleviate a number of problems users tend to complain about here.
Just imagine how useful people might find it to simply block posts by keyword, or not show threads with a certain ratio of upvotes to downvotes, or be able to train their own Bayesian filter.
They wouldn't even have to change the UI (which they seem to not want to do) apart from what a particular user sees.
The filters would get very complex very fast though, I imagine. The symptoms of "echo-chamber-itis" are as varied as the humans that exibit them. Sometimes it's aggressive downvoting of an idea that doesn't align with the status quo despite validity for discussion, sometimes it's aggressive upvoting of an idea that aligns better, sometimes it's an aggressive amount of churn as various subgroups battle for "placing" the item, which results in the item being read as a net 0. (I'm being very hand wavy and referring to patterns I see broader than just HN; before anyone starts jumping up and down about how the HN algorithms avoid any specific problem I mention :P )
You're probably right - even though I think people do want their own bubbles, attempting to provide them might just result in more complaints when the filters inevitably fail.
Then again, there does appear to be a confligt in the way the HN community seems to be growing, with the site itself still geared towards providing a relatively low amount of content through a single channel. Stories overwhelming the site and reposts are a known problem, and probably almost no one bothers watching past the first page of /news so those top 30 slots might as well be the entire site. Although, many of the most obvious remedies to this would make the site look more like reddit, and we can't have that I guess.
* Ever notice how people who make claims about why they got banned never provide links to the posts in question? That's because their claims are nearly always false. If users could look at the actual record, their perennial sob story of perfectly reasonable behavior struck down by bullying censors would evaporate. *
When that kind of response comes from the site's moderator, I really don't see the level of toxicity improving any time soon.
It's likely statistically factual, but in context it's just another of the "mean, stupid things" that Paul Graham called you out as being here to address. And you appear to have done nothing to investigate whether the previous user's post was factually correct before slinging personal accusations.
If this gets me hellbanned too, so be it. Conversation and community on this site is a toxic mess that leaves people afraid to post anything. The main good thing is following users like patio11 and tptacek.
You may be right. I'll look at the comment later and see if I should have written it differently. There's no time to reflect just now.
I try hard not to let personal irritation leak through in my HN comments, but I do fail at it. The most I can claim is a willingness to correct mistakes.
It's wildly inappropriate to lob a blanket accusation of fraud against every single person who has ever complained about your moderation.
People have wildly divergent views as to what's appropriate, what's a little rude, and what's over the line. This means that even if you really truly believe that there was NEVER a mistake made during moderation, that some people will truly believe what they said.
Good moderation requires a ton of empathy and kindness.
I agree with this, but I think it's worth noting that the prompt here was someone making a specific claim. One doesn't need to believe they have made no mistakes to be certain they've never made a particular mistake, and there's significantly less room for differences in interpretation (though that's not to say there's none).
If users could look at the actual record, their perennial sob story of perfectly reasonable behavior struck down by bullying censors would evaporate. So they make new accounts and post statements designed to be unanswerable.
I interpreted that as dang attributing essentially every complaint to malice, and simultaneously dismissing all other explanations.
A misleading summary, since you're the one who introduced the idea of "malice" and "fraud". Dan's claim admits to HN users who believe they've been hellbanned for criticizing YC companies. Yours doesn't.
> Good moderation requires a ton of empathy and kindness.
Which is sadly lacking among many, if not most, moderators of online communities across the net. I'm not saying that's the case with dang; in fact, I wouldn't know. But it's a thankless job that is akin to working in a call center without pay. It takes a strong personality to keep one's head above the layer of filth floating atop the waters of discourse.
I do know, and Dan has an enormous amount of empathy and kindness. But it is a hard job, it takes a toll, and I think this thread demonstrates how much he's willing to re-visit what he's said. (Even though I'm quite sympathetic to what Thomas said upthread.)
It's probably a lesson to me that the one time I didn't hedge by saying "almost" or something like that, someone objects to my "blanket accusation". Actually, I originally wrote "almost never" (or something similar). But then I realized I couldn't actually remember a case where someone provided a specific link to back up his or her grand claim of why they were banned. So in a fit of impetuousness I lopped off the "almost". I did leave in "nearly always", though.
> Good moderation requires a ton of empathy and kindness.
I try, but don't always succeed. Thanks for the reminder. I appreciate it.
For what it's worth.. I don't even read that much, and I've not been here for that long, but I often was impressed by how much you actually do engage and do seem to care, a lot, to do right by everyone, in public. To say you lack empathy and kindness as a moderator in general would be just silly. I say this as someone who strongly dislikes hellbanning and even grey text (I still think slashdot nailed it with voluntary, customizable filtering), so you know I mean it :P
I appreciate your willingness to see both sides and think of a good-faith interpretation here. That's the Principle of Charity which HN can use a lot more of. However, the users in question are typically quite accomplished at making throwaway accounts for specific purposes. Several have done so in this very thread.
There's no way, barring some freak outlier, that we banned anyone for criticizing YC or a YC-funded startup. If someone really did feel that way, nothing would be easier to clear up.
The real issue, in the overwhelming majority of cases, is repeatedly flouting the HN guidelines.
I disagree! (with your pre-edit, reading "It's not that you expressed yourself poorly").
That's exactly what happened here. (And maybe Daniel can edit his comment.)
Daniel does a really good job, and is extremely responsive by email and on here. It is obvious where he wrote "Ever notice how people who make claims about why they got banned never provide links to the posts in question?" is borne of deep frustration. He would like to follow those links and improve the site, but can't. It's obvious that his comment is written out of frustration.
Let's be very clear: hellbanning is the worst and rudest thing that exists on any respectable Internet forum. Hellbanning literally wastes hours of the time of people who contribute great insight for free. The comments on this site are good and provided for free by people. Hellbanning turns this goodwill on its face, like a goodwill jar you can put bills into but which go into a furnace.
Daniel (and PG) knows very well that hellbanning is a nuclear weapon and the rudest thing that any Internet forum can possibly do, that is actually being done.
You have stories of people only learning they were hellbanned after literally taking the time to email someone a link to something thoughtful they had written. A lot of hellbanning has been (historically) in error.
It is one of the main reasons that I would never consciously leave a comment up if it reaches -3, even if I stand by it 100%, it's important, and the community happens to be wrong in its groupthink and I clearly have explained why. I would delete it instead.
Note that I have learned this behavior, and so have other contributors on this site.
It's one of the things that makes this site great.
So even though it is a nuclear option and the worst, rudest thing that any respectable forum does, in the sense that time is money literally stealing from users, and stealing donations at that and throwing them away, at the same time it is one of the things that allows this site to function as one of the best sites on the planet.
So you can bet that Daniel is extremely serious about following hellbanning claims and improving this process. It is difficult and he walks a very fine line.
He's doing a fantastic job at present in a very difficult undertaking. Kudos, Daniel, and keep up the good work. I can read your comment for what it is :)
"It is one of the main reasons that I would never consciously leave a comment up if it reaches -3, even if I stand by it 100%, it's important, and the community happens to be wrong in its groupthink and I clearly have explained why. I would delete it instead."
I am confused. Your preferred path is to avoid conflict such that you would rather delete than be disagreed with? If your opinion differs from groupthink, you would make it go away? I guess that is similar to not posting in the first place (because of groupthink you disagree with) , just retroactive.
Probably better than my not posting in the first place :)
His point, one that I strongly agree with, is that the threat of being hellbanned for comments that get downvoted is enough to stifle discussion on HN. Honestly, how often do you see passionate debate in HN comments?
My interpretation is at least a disagreement about whether the technique is effective. If your attempt at improving the discussion looks more or less the same as whining about downvotes, it's going to get interpreted as whining about downvotes.
edit: I guess the thread was getting cluttered and argumentative.
I'll go ahead and call Dan a friend, display my "not a shill for YC" bona fides, vouch for Dan as not a toxic person, and go on to criticize him for being so concerned about the appearance of toxicity that he's backing down on an issue he should stand his ground over. If he says HN moderation won't hellban someone for criticizing a YC company, take it to the bank.
I feel compelled to add: the accusation seems baseless. Over the several years I've been participating in these forums I cannot remember a single instance of anyone getting hellbanned or otherwise penalized just for criticizing a YC company.
I don't even think that answer was even that bad, especially when compared to the parent. Here was a wild and personal accusation with no support at all, not even anecdotal, and pretty toxic itself.
Replying courteously even to baseless and ranty accusations is probably a good idea, but I can't say I'd blame someone for not doing it.
I see the mod(s?) deal with this kind of conspiracy accusation almost on a daily basis, and I can definitely see how patience can wear thin when every nut does that. Dang implying that this accusation was baseless was probably rash, but that's the only thing I see even remotely out of line here.
Moderation is a pretty thankless job, and like sysadmins, people never appreciate you for getting things to work right when they work right. The definition of success is invisibility to users. But the second it goes even a bit wrong, it's a shitstorm. Here's yet another piece of empathy that one needs to consider.
But in the places I have been banned, it's not been my fault. The people who say "we don't ban people because of X" are, like most people, just telling themselves a pleasant lie.
There's that one guy, who while never doing anything outrageously obnoxious, rubs you the wrong way. And eventually you're going to find something borderline or even milder than that, and use your petty powers. This is human nature, I'd almost certainly do the same. Everyone would.
You don't tell people why they've been banned at all, or even when they've been banned. Is it any wonder that conspiracy theories would arise around such a purposely opaque practice?
My experience comes from a different (IRL) context, but the type of people you need to moderate or ban are also often the type of people who will move right up to the edge of any bright line you draw, and use it is a shield for further antisocial behavior. Sometimes ambiguity is its own reward, especially when the lines are well-understood by most already.
As just one example I'm the resident "soldier" for the government (as one HN user was so nice to call me), I often post comments that are in opposition to many of the tenets that are popular on HN, and yet I've not been banned, my karma is not negative, and those I debate with generally treat me and my arguments with respect. If it were just up to ideology I should have been hellbanned by now, and yet here I am.
1) you ban people who are rude and who you disagree with philosophically, while you do NOT ban the equally rude people who you agree with.
For evidence of the above, look at users like etherael (and his other names; not sure if he tors/vpns or if you can find them), who are raging assholes on a semi-routine basis, but who aren't banned because Libertarian BitCoin Lover matches your values. And let's face facts, you're less willing to ban people who agree with you, even if they're toxic assholes.
If you banned people who you agreed with for the same exact crimes as those you disagree with HN would be a better place.
As it stands, people who agree with you are allowed to be ruder and more toxic than people who disagree with you. This is used as a game by some of HN's worst users who brag on IRC about how it's fun to try to engage in flamewars where they don't get banned but the other individual does.
edit:
Not to mention other game that's played by a lot of folks, which is to be as big of a dick as is possible without actually using openly aggressive language. The goal there being to generate an emotional reaction while retaining some semblance of plausible deniability, because everybody knows that you won't ban them for "polite" taunting, even if it's toxic shit that can't go anywhere useful or interesting.
The people who hold the opposite ideology believe fervently that HN is biased the other way (liberal, politically correct, socialist, etc., are some of the terms they use). I realize it's a bit facile to say "both sides claim bias therefore we must be doing something right". But for what it's worth, no, we don't consider ideology when banning people, we consider incivility.
Which is why many toxic people have perfected the art of being an enormous asshole while being "civil". Because they know that you allow that to happen.
As to your claim that you aren't soft on Libertarians... LOL. I've actually A/B tested this, saying the same toxic shit but with the positions swapped. Upvoted in one case, banned in the other.
Part of the mechanic might be that you never even see the Libertarian version because it doesn't get flagged in the first place, despite being identical in profanity, intelligence, and toxicity to the version that does get flagged. but whatever the causes, there's a real and definite bias as to what positions are tolerated here.
---
Response to below: Etherael gives good conversation like:
"Congratulations, you are a retard."
"I have no hopes for you at all."
"Get off that lawn, you old crank."
"Why is this being downvoted? Are people really so naive as to not be aware of this?"
---
" Because when you present a point in this way with this kind of language and in this kind of behavior, you just lose everything."
You're right. I have the same amount of respect for dang as he has for the public at large. And that's my fault. I should be better, even if I really believe that HN is worse and less useful each year, and that dang is poorly suited to moderation.
He's probably a smart guy, but we're all just apes, and he's one who isn't the right fit for the job.
> No offense kid, but you're fucking delusional. And HN is worse because of your inability to self-reflect.
Even if you had a point, you just lost it. Because when you present a point in this way with this kind of language and in this kind of behavior, you just lose everything.
I'm a pretty skeptical person (see my skepticism on my last comment, for example!) who views most actions by most companies very skeptically and tries to see if there might be ulterior motives. And I've gotta say, you're wrong in this instance.
> As to your claim that you aren't soft on Libertarians.
They're not (they're not soft/hard based on ideological beliefs). For example, DanielBMarkham, perhaps the most outspoken libertarian on this site, is rankbanned. Now, I do have serious doubts about whether or not they're "soft" on folks saying negative things about YC companies/people.
Lastly, I've gotten to know etherael quite well -- he's got a sharp tongue, but he never quite struck me as an asshole. I do know that he's very talented at what he does, and almost always provides good, intelligent conversation about anything I bring up to him - and in that way, fits right at home here on HN. I guess though maybe you caught him in a bad time being especially rude? The best of us lose it sometimes. I hope etherael is more thoughtful in his future replies.
The worst I ever do is respond rudely to people who have already attacked me, and even there I try to avoid doing so. As for using other names that's just flatly false, as is any gloating about baiting people into flamewars and laughing when they get banned. I haven't even used irc in over a year.
Basically I have no idea what you're talking about.
Dan is a libertarian bitcoin lover? I've spent several hours in person talking to him. I'm a statist liberal Democrat who believes bitcoin is a ponzi scheme. He did not set off my spidey sense. I think you might be attributing generalized fears and frustrations onto specific people you don't know.
Being in the same ideological quadrant, I can say that I tend to get as many or more upvotes on political comments as on apolitical/tech-focused comments. And I tend to follow responses to my more contentious comments closely, and I have only very occasionally noticed even a single downvote on said comments.
I think there's a lot of outspoken libertarian/anti-statist types around here, and that's fine, but I don't think it's even the plurality among political stances of HN readers. I suspect being invested in politics to the point that you'll regularly engage in political discussions online is very strongly correlated with holding atypical political views (I include myself in that set).
>Ever notice how people who make claims about why they got banned never provide links to the posts in question?
>their claims are nearly always false.
>perennial sob story
Hmm, I wonder where these fall on Graham's disagreement pyramid...
Seriously, how hard is it to tell the guy "shoot me an email and I'll look into the ban" and then make your judgment after that? Maybe the guy's bullshitting us. Or maybe he really was banned unjustly. How can you possibly know?
Neither of your things in "quotes" are things that I said. The "shitty friend" was alexcap's own words. The phrase "think of, or care for, others" didn't show up in my comment at all.
I didn't call him a shitty friend. He wrote, "Now you’re probably thinking I’m a real shitty friend" after his explanation of the events — which I quoted, in quotes, very clearly a direct quote from his own words — and I wrote "Yes" to answer the question. 95% of my comment was advice for people on what to do in the situation that the friend is reaching out to them as that friend did, who got ignored and later died and that person didn't know for years.
If you write an essay about something you did that you think was bad, and post it to Medium, and post it to HN, and then say "You probably think I did something bad," is it reaaaally so controversial to comment with "Yes, I agree, I think you did something bad"?
Also, if you thought that was my last comment, maybe you don't see my actual last exchange? It was about my bootstrapping conference and upcoming web site. Perhaps it's truly hell banned as in not visible? It was this:
> "The phrase "think of, or care for, others" didn't show up in my comment"
It's a direct quote of a dead comment you wrote in a reply. Turn on "showdead" to see it [0].
The dead "think of" comment in the "shitty friend" thread appeared 405 days ago. Your comment about the bootstrapping conference was 406 days ago. (There's also a dead reply in the bootstrapping thread dated 405 days ago, which fits the timing of being hellbanned for the "think of" post.)
You call it the "shitty friend" thread is pretty inflammatory considering, again, that was the OP's own words (not mine). I turned on showdead and my dead comment that you're referring to sure looks reasoned and measured to me:
I never said anyone specific is self-centered. I said:
>How self-centered do you have to be to not even wish a friend with obvious health problems "good luck" or "feel better"?
That's a rhetorical question.
Then, in the dead comment, I wrote:
> Being a bad friend in this way doesn't make you a bad person…
> But it does make it seem kinda iffy to write a blog post about it, even include screenshots of the conversation, and not (apparently?) be socially aware enough to realize that the deceased man was nearly begging his friend to express some interest and concern. And so many of the commenters, from my perspective, were not picking up on anything the OP did not explicitly lay out in the essay itself, which is to say: his friend was telegraphing his problems in every possible way, and the OP ignored it. People seem to be reading it and thinking, "Oh, just one of those things." But it's only one of those things if you don't think of, or care for, others.
Again, I didn't personally attack alexcap. You are quoting it out of context.
Real inflammatory stuff. If this is what I got hellbanned for, it really makes me wonder about the actual, deliberate, specific, personal cruelty that goes by on a regular basis without banning.
Meanwhile the other dead comment… the one that's actually newest, not the one you claimed above was the newest… is about my bootstrapping conference and upcoming site:
> Oh, good question. I can see why you'd think that.
> Nope, BB will be more like HN, but specifically for bootstrappers and related topics only. IOW: public-facing, free to use. (Although I think we're going to do a MeFi-style $5 or $10 join fee, to encourage good citizenship.)
> It's not going to be a product. I'm not going to run it unilaterally, either. It's for the community.
That is the last thing I posted before being hellbanned.
I'm using "shitty friend" as a disambiguator -- whether they're your words or his, they act as a unique identifier. And I think your comments in that thread are the likely culprit. You seem to think what you wrote was reasoned and measured, but I can see why a moderator might read them and think "whoa, this is way too personal" and ban you. EDIT: remember, you know what you were thinking when you wrote it, but others might read the same words and take a different meaning. The words you used could have been interpreted as pretty inflammatory. (Again, I'm not defending the hellban as the right decision, just explaining that I see an obvious reason it may have happened. And yeah, I've seen people say worse who didn't happen to get hellbanned for it.)
Your other dead comment was probably posted after you were hellbanned (the point of a hellban is that you wouldn't have known it happened right away). That's why both comments are dead. If you'd been hellbanned for the last comment, only it would be dead.
No, being saucy is fine, and people much more disagreeable than you (cough) seem to thrive here. I can't see why you were banned, but I asked, because that's weird.
I think it's pretty complicated, actually. I've lurked here for years but only made an account within the last year. In that short time, I've said arguably worse things than you have (though not with ill intent; I can just be abrasive and opinionated) and as far as I can tell I've not been hellbanned yet.
It probably helps that I'm a "nobody"; I'm not a member of the startup culture, just an outside observer and occasional commentator. If it's true that politics are a factor in bans and heavy-handed moderation, then I can understand (though not agree with) you being hellbanned for what are fairly mild comments as judged by an outsider like me, as you are a member of the startup club.
And I hate to say it, but misogyny might even have played a role in it. I have zero evidence of that and it's not meant as an accusation, just that it's a remote possibility. It's a real problem in just about any internet based community, and I really doubt this one is immune to it.
Ohai, I've been hell-banned and don't know why. I guess as someone successfully running a software business _and_ being a well-known open source person I have to place in a forum for people who want to run successful internet businesses and are into tech.
So on the topic of unfair bans and the new system, here is a useful way for you to both improve transparency and figure out for yourself if the new system is working.
Post a list of comments which were [dead]ed under the new system but would not have been [dead]ed under the old system.
If the list contains a bunch of comments like "u r a gay homoz", you'll make a pretty convincing case (both to yourself and everyone else) that the new system is awesome. If the list contains a bunch of "I'm concerned about the security implications of ordinary users putting significant money into bitcoin..." then maybe the new system isn't so awesome.
[edit: I realize it's probably too late for this to be seen, perils of posting from IST.]
It's little different than pre-internet communities. I mean, we paint this adorable picture of small-town life from some previous golden era, in movies and in books, and even in the stories of your great-grandparents.
But people are assholes. And they found a few people they didn't like (for good reasons, for bad reasons, for none at all) and did the meatspace equivalent of hellbanishment.
Yeh, here, you're just wasting time on the internet, maybe you can stop caring about it. But people do this the world over, and there's no escaping it. This is what people do. They're mean fucks, and if you don't fit in, you're just left out in the cold.
"... After being hellbanned I actually realised that commenting on internet forums is toxic, generally a waste of time, and not productive. So I quit. ..."
Yes.
... But for those that like spending time commenting, I would have thought following a subreddit model would work better here, with the growth. Spread the power out. ..."
IF someone is a genuine troll, then hell-banning is a great way of dealing with it, because it maximizes the time wasted by the troll and minimizes the time used by the moderator.
However, if the person isn't actually out-and-out to mess with your site, it sucks. I've seen lots of communities go down the tubes because the moderators get busy with life, and then Something Happens On The Forum, and moderators passive-aggressively go "fine, we're just gonna act like this, and if you don't want us to do that, then you should make the community act better!" Basically announcing that they are going to put in minimal effort to moderation, and the community definitely notices.
Professional moderation, like HN uses, seems like the best way to go.
Indeed, it's a situation where the needs of the many (HN community as a whole) outweigh the needs of the few (banned non-troll commentators). The danger lies in becoming such an elite, closed group due to blind moderation and banning even the most innocent members over a perceived slight, that the moderators themselves end up the trolls of what is left of the community.
For example, there's a certain GNU/Linux distro that is maintained by a core group of devs who have become overtly hostile to any new users of their project, and actively seek to discourage "newbies" from seeking help and getting any benefit out of the project. One would think the toxic atmosphere would have killed the distro off long ago, yet it's maintaining popularity and even seeing an uptick in a certain niche community. I certainly don't understand how it thrives with such a rotten core; yes, it's overall a very well done distro, but even the best product normally can't survive that kind of cancer.
That's not to say that HN would ever end up like that; indeed, from what I've seen they are doing an excellent job overall with maintaining and moderating this community. I just hope it continues to stay that way or improve, instead of going down a dark path towards chaos.
My account segmond got hell banned, it's a miracle that this is not. HN will downvote any comment that they don't like down to hell. An opinion that is disagreeable with is flagged, it's like the opposite of facebook like button. Dislike.
You don't have to spam or be disruptive. I'm putting my account name on here so folks can look at it. I was new to HN, didn't know much, had no idea I was dead. :)
This account get's a decent amount of flag too when I make unpopular opinion, I have to stop myself from censoring myself because the moment I feel like I can't talk or participate, I will just leave the community. If I can't express myself around a bunch of "hackers" then what's the point?
We've got an idea about this that we want to try. It's one of the future experiments Sam alluded to at the end of his post.
(I'll add some details in a few minutes. Edit—well, quite a few minutes.)
The problem is that banning an account is an all-or-nothing system where inappropriate commenting is not necessarily an all-or-nothing phenomenon. The solution we want to try, which was suggested by several users, is to give the community the power to bring comments out of [dead] status. Letting fellow users make the call when a comment is (or isn't) ok seems like it might work, especially since the recent experiments with flagging seem to have helped. This is one of the things Sam was referring to when he said we plan to experiment more with community moderation.
Very much looking forward to that, it will have two good effects, it will take the edge of hellbanning and it will reduce the load on the moderators. Three good effects, and give those that have been hellbanned (if they are aware of it) a reason to try to be extra good in the future.
Four good effects, and it will make hellbanning more effective (because occasionally there will be answers to a comment). Five good... Wait, let me come in again.
If you mean: after a certain number of comments have been lifted by the community out of [dead] status, the account could also be lifted out of banned status, then I agree—we'd like to try that. Effectively, this would give the community control over banning and unbanning.
We wouldn't call these "dead" or "banned"; we'd probably introduce new statuses like "flagged" or "moderated" or something like that. The current form of banning would be reserved for obvious cases of spam, etc.
Apparently, this is the comment that got my old account banned: https://news.ycombinator.com/item?id=4342935. I'm not sure though, it's just the last one on the profile before they go dead.
I'm guessing that got caught in the spam filter for use of the words "jewelry" and some of the other keywords in the Amazon link. This is one of the unfortunate side effects of running a startup that happens to be in a field that generates a lot of (affiliate) spam.
I had an account get "hellbanned" a while ago (it was an account with less than 20 karma to begin with) and after some stupid comment on my part I got downvoted significantly. Didn't even realize it had happened until someone reached out to me via Twitter informing me that my account was "hellbanned".
Well it insults peoples feelings if they are being downvoted. It seems unrealistic to me to expect that they just accept the humiliation, even if it would be the rational thing to do. (I'm guilty of meta-commenting, too).
I agree with you, and so do the HN guidelines (as minimaxir pointed out). But it's clear that pleading with people not to do it isn't enough. We have sometimes toyed with implementing a lightweight guidelines reminder to indicate (privately, to the commenter) when a comment is breaking one of the guidelines. I'd still like to experiment with this someday.
Ideally, it would be a community thing and not a moderator thing. The spirit would be, let's all remind each other of the guidelines when we need to. The things that HN cares about, like civility and signal/noise ratio, don't come naturally to an internet forum. We all have to work at it (definitely including me).
Features like that are hard to get right, though, and until there's a good idea of how to do it that fits naturally into the existing site, it will remain prospective. Anyone with ideas here is welcome to send them to us at hn@ycombinator.com.
370 comments
[ 1.9 ms ] story [ 388 ms ] threadExcellent.
> Second, we've started indicating in the UI which comments/stories have been killed by user flags.
What does this look like in the UI? I haven't seen any examples of it yet, and I'd expect to see at least a few on a regular basis.
It should be noted that during the aforementioned six month periods, both the number of new comments made monthly and the average score for a given comment decreased, although this post claims a 30% increase to Hacker News traffic due to changes in the comment system, which is interesting assertion of causality. (in fairness, the decrease in number of new comments could be caused by the increased moderation of bad comments.)
It would be helpful if this article in this submission clarified how to flag comments, as that action is unintuitive (you have to click the comment permalink first) [EDIT: looks like this was added to the submission]
So that's how you flag! Thank you! This definitely should be more noticeable.
This is apparently unintended, as the official API has no points field. (example: https://hacker-news.firebaseio.com/v0/item/2921983.json?prin...)
I have been using a mobile-friendly version of Hacker News called HackerWeb on my phone and tablet: http://cheeaun.github.io/hackerweb/
A handful of browser add-ons exist that will update the visual design and generally make the site more pleasurable to use. Here's one for Chrome: http://gabrielecirulli.github.io/hn-special/
Good to hear they decided to roll that back.
PG opened a pandora's box when he said that downvotes were acceptable when you merely disagreed with someone, instead of doing so for civility. There is no end of sadness and bullshit caused by that remark.
It'd be nice to see an official policy statement changing that position, but I don't think it's likely. In the meantime, we will simply have to stop assuming that downvoted comments are actually, you know, bad.
I don't think this is true. Given that downvoted comments become faded, I have a strong reason to believe that the intent is to hide them because they are actually bad
But, people downvote comments for any number of reasons, and even doing so for mere disagreement is within the acceptable guidelines - and of course for everything, despite the lack of clear context, the effect remains the same.
So is threatening to delete your comments over down votes.
An example of where I recently deleted a comment was in response to this one: https://news.ycombinator.com/item?id=8569545 where my opinion was actively solicited. However, I decided that the reply to it sent the thread in a wrong direction, and also could be read without my reply. I looked instead for the person's email who had asked, but he didn't list one so I ended up deleting my comment. In effect I was saying that on reflection I chose not to comment on the linked essay in that thread.
And I have to say, you do delete comments with replies.
I agree there is nothing wrong whatsoever with his actual comments.
I agree there is nothing wrong whatsoever with his actual comments.
That's the nicest thing anybody's said about me all week. :)
Would it really be as crazy as it sounds, in the interest of promoting discussion above all else, to just hide usernames altogether?
Being fluent in both Swedish and English, I know there are certain concepts, such as the difference between belief in an opinion and belief in a fact, where the Swedish language makes clear distinctions (tycka and tro) and English does not. English speakers spend ridiculous amounts of time arguing about these things, and Swedes simply don’t need to. It’s not that English speakers can’t conceptualize the difference between opinion and fact, but doing so is way more difficult in English, because the word “belief” in English is quite fuzzy, whereas in Swedish, the language makes it simply impossible to confuse the two.
HN voting seems to suffer from an inability to make this distinction: I think it's terrific when comments are downvoted because they're factually inaccurate, no matter how well they're written. So much nonsense ends up at the top of some comment threads simply because it reinforces existing beliefs or loyalties. On the other hand, if you make a good case on one or another side of some live debate, I don't think it should matter whether I agree with you or not. The discussion is always enriched by quality arguments.
[1] http://ken.arneson.name/2014/11/10-things-i-believe-about-ba...
Since that remark came to my attention, I've put less thought and research into my HN comments.
It only takes a few net down votes to make a comment sufficiently hard to read that it is effectively dead. For example, suppose there is an article about government surveillance, which the article opposes, and the article makes an assertion about the 4th Amendment that is wrong. If you post a comment explaining the error and how the 4th Amendment actually works, some people will misinterpret this as implying that you are in favor of government surveillance, and down vote you for that.
The people who correctly interpret the comment are not as likely to up vote as the people who misinterpret are likely to down vote, and so the down voters have an advantage.
If down voting simply moved the comment toward the bottom, rather than making it almost unreadable, it would be OK. It the combination of making down voted comments hard to read and saying that down votes for disagreement is OK that is the killer. The message that sends is that comments here are supposed to agree with the majority opinion.
I sometimes flag blatant political stories, because I don't think they belong on HN, so I guess if I suddenly can't flag anymore then that was the wrong answer.
But is fading the comment really the correct behaviour. Isn't the position on the page deciding whether the community agrees with the comment?
My suggestion is: remove the fading; let downvoting only move the comments down, since the comment is appropriate yet the community doesn't agree with it; and let flagging remove inappropriate comments.
You CAN indicative that the comment has been downvoted - the fading accomplishes that objective.
Applying fading to all the comments just because a downvoted comment has a parent that happens to be at the top, seems wrong.
Another approach is simply to hide the offensive comment and weight the entire thread in relation to other threads when determining display order.
I've lost count of how many times I've seen perfectly good insightful comments around here that get ignored because some people decided to downvote them for some arbitrary reason. In fact, it's so bad, that I'd say I've barely seen any good instances of fading for the past year; they just seems to encourage group-think almost every time.
I'm ok with the idea of downvoting meaning disagreement, since that's how voting systems always end up being used anyway (not to mention pg approves of that interpretation as well), but penalizing comments to become practically non-existent based on a few downvotes (last time I noticed, a score of -1 started the fading) is just not conducive to interesting discussion.
So perhaps a margin of, say, five or six downvotes should be necessary before any visual effect takes place. Not only would this no longer bias readers against a comment before having read it, but it would guarantee that the faded comments more accurately reflect the standards of the community as a whole.
I hope downvoting has nothing to do with agreement; I thought it was for comments that aren't valuable to the conversation (e.g., not substantive, poorly reasoned, false, or poorly communicated).
I want to see many more comments that are valuable and challenge the community (and challenge my thinking too). The groupthink is already well known and if not, will certainly be posted by someone.
In practice, I do see downvoting used on comments that are valuable but challenging. It's disappointing.
EDIT: remove embarrassing typo
That's a weak distinction when there is a factual disagreement, since most people probably believe that factually incorrect statements are not valuable to the conversation.
I don't think this is true at all. I disagree that 1) this is a thing most people believe and 2) that incorrect statements are not valuable.
I would hope our disagreement would cause people to refine their thinking about what they post as well as what they downvote and flag.
[*] Most, referring to most HN readers
I generally don't unless the disagreement has another value-subtracting component (lack of substance, poor reasoning/phrasing, factual errors, unnecessary incivility.) For disagreeable but worthwhile comments, I prefer responding instead. But I recognize the value in having differences of opinion over the use of downvotes -- comments of marginal quality (which some people disagree with, and others don't find valuable enough to upvote even if they agree) will end up dropping down the page.
What value does this provide to the reader? Community groupthink is not strongly related to truth, IMHO, and provides little marginal value, because I already know what it is and it's easy to find. In fact, I wouldn't read HN if all I found was the groupthink -- I gain far more from new or challenging ideas.
EDIT: When Amy up/downvotes, it is for the benefit of Bob and other readers. Amy receives little benefit besides maybe a little emotional satisfaction. If the vote mainly was for Amy's benefit, HN wouldn't need her vote to impact anyone else's threads -- we all could have personalized, independent scores for each comment. What benefit do I, Bob, receive if Amy is just voting on what she agrees/disagrees with?
> "Community groupthink is not strongly related to truth"
In strongly technical discussions, which are common on HN and which HN should be optimized for, "groupthink" and "correct" tend to correlate quite nicely. Whenever there is a concrete notion of "correct", it works well. It is only when "correct" becomes subjective (such as in political discussions) that shit gets messy.
In other words, downvote and upvote according to what you think those actions mean.
But the amount of times I've seen controversial opinions rise to the top just because they're thorough and elaborate has been really fantastic, and I make sure to upvote things I disagree on myself. On reddit, the top comment tends to be a one-liner pun -- and again, it feels wrong to me to upvote things I disagree with on Reddit, so I don't do it and it just makes the whole situation worse.
That doesn't mean it's a good idea. I would urge everyone (including PG) to reconsider the acceptability of this practice -- it strikes me as being harmful.
To put it a different way: PG & friends have put together a quality discussion forum. One of the reasons it remains relatively high quality is necessarily that many users of the site do not use downvotes to express disagreement.
I think the reason it's high quality is because it started as a place for the kind of people interested in startups. To me that means (statistically) well educated, ambitious, interested in STEM topics, etc...
Just like in the old days of internet news before AOL made it available to the masses. For example, sci.math had awesome discussions among leading mathematicians, and then the masses came looking for help with their homework and those guys left.
Like all good forums, if it does really well to the point that the general public takes interest, it will likely become crap.
That isn't an inevitability, it's a consequence of poor forum design. At one time, if water started leaking into your boat, it sank. Then someone had the bright idea of pumping the water out and sinking was no longer inevitable.
One mechanism that might prevent the dilution you describe is to hide low-scoring comments from low-karma users, but show them to high-ranking users. Then the high-karma users can exert influence on the low-karma users by upvoting quality comments and downvoting poor ones. It becomes a self-organizing system with a bias towards quality, rather than the open loop 'majority rules' system currently in place.
Its like you invite people to a debate, where you say "please only boo when someone is saying something obscene and nasty or breaking the rules of the debate, not when they are saying something you disagree with".
I agree.
Thing is, the "when to downvote" debate is ancient at this point and no closer to consensus.
These days, I'm given to thinking that any voting system should assume the behavior of least resistance, that it's thoughtless, reflexive and therefore only potentially valuable after being sanitized / aggregated / weighted.
I was under the impression that downvoting is for comments that you do more than just disagree with. If you disagree, but it is a valid interesting argument, I try to vote a comment up, in any case not to vote it down.
Flagging is for inappropriate comment, stuff that is harder than just a bad comment, like racial insults or spam.
I'm sure that will be explained to us :)
In theory, no, downvoting is not supposed to be used to disagree. It's supposed to be used to move "stupid" comments (inarticulate, illogical, poorly argued, etc.) down towards the bottom of the screen. Flagging, on the other hand, is meant to call out wildly inappropriate, offensive, or completely off-topic comments.
A few hypotheticals:
1) I see a comment that is well articulated and generally inoffensive, but I disagree completely with its premise or its arguments --> I do nothing.
2) I see a comment that is irrational, illogical, poorly constructed, whiny, or off topic --> I downvote.
3) I see a comment that is extremely offensive, completely unrelated to the topic at hand, spammy/astroturfy, or clearly intended to instigate a flame war --> I flag.
That said, many people seem to use downvotes to disagree with a comment, and I'm concerned that lowering the downvote threshold will lead to even more of that behavior.
Maybe the problem is that downvoting and flagging share too close of a hypothetical use case. Downvoting is clearly easier to do, and it's doable in the main thread view. So that could be why people routinely use it to disagree with comments. I have flagged maybe a grand total of one comment in all my time on HN, and I generally regard flagging as an option of last resort. (It could also be that I arrive at various threads after the worst stuff has been flagged and removed, providing me an artificially sunny view of the threads).
Maybe this is a problem, in and of itself? Seems different people have different takes on the use case for downvoting.
I don't downvote disagreement, but that's because I'm a compulsive (and, from what I can tell, fast) commenter.
The simple way to metabolize this system is just not to get unhappy when you get downvoted.
That being said, I still feel weird about downvoting to disagree. The way I see it, disagreement-downvoting creates noise in its own way. Less noise than unproductive or bad replies, certainly, but more noise than doing nothing.
"The simple way to metabolize this system is just not to get unhappy when you get down voted."
True, and this is also an official guideline. (A good one, too, IMO).
No, quite the opposite. (See I wrote a comment explaining that I disagree, I didn't just click the arrow. Much more valuable to you and the reader.)
And that's why you are wrong, by saying why you disagree you might (rarely) change the posters mind, or more likely change a 3rd readers mind about the topic.
Clicking downvote is pretty useless.
I don't think it's the norm anyway, I think it's a minority that do that.
I wasn't even aware of flagging on comments until someone mentioned it, and to be honest it's so inconvenient i'd probably never do it. Downvote is an easy enough way for me to say "I don't like this" and it helps reinforce the group dynamic, which is what they want (this whole post is about manipulating group dynamics).
The only way to fix downvoting is to put an express purpose behind it, like a button that says "I disagree" next to a button that says "I dislike" next to a button that says "Inappropriate comment". But for some reason forum admins always prefer generic feedback which give rise to unexpected behavior. Heh, it probably keeps people upset, which reinforces user engagement.
Maybe I'm in the minority on this, but I don't approach voting from an agree/disagree axis. I approach it from a productive/unproductive axis. I upvote comments I find helpful, articulate, or informative. I downvote comments I find unhelpful, inarticulate, or uninformative. I leave alone comments that I find "meh," or that someone has already said better, or that neither add to, nor subtract from, the conversation. Basically: something either adds to the conversation (upvote), subtracts from the conversation (downvote), completely derails the conversation (flag), or is neutral.
Case in point: I didn't necessarily agree with your comment, but I found it articulate and interesting. It presented a perspective I hadn't considered. It made me think. I upvoted it.
Now, one could argue that agree/disagree and productive/unproductive are basically the same thing in practice. They both represent a desire to move the conversation in one direction or the other. (To your point about the group dynamics). But I don't know if I totally buy into that. I have no idea if this is true, but I'd hypothesize that voting patterns would be different under one explicit guideline or the other. I'd also think that downvoting-to-disagree runs the risk of fostering an echo chamber, which discourages dissenting or different viewpoints, particularly on hot-button topics.
The problem is that smart-but-unpopular comments may fail to rise up. We want to prevent groupthink by ensuring that the top comments don't exclude ones that are well written and productive while expressing a contrary viewpoint.
A simple "vote down for disagree" breaks that.
Why do we vote in directions? It would appear that according to social media, we only have one orientation in voting (vertically) and can thus only go in two directions (up and down). But this makes no sense. For one thing, the 'direction' of a comment/conversation may go many different ways... conversations do not go in a straight line. And the idea of 'elevation' in a conversation is also confusing: if anything, a conversation would be more three-dimensional than two, moving around a changing landscape.
In addition, 'up' and 'down' gives an immediate subconscious reaction where 'up' is associated with goodness (and therefore correctness) and 'down' with badness (and therefore incorrectness). I say this moral implication is the default because it makes no sense for only one direction to be correct, unless you needed to know where you were supposed to go, and your default assumption was that you should never go down. (That would be a bad presumption for airplanes, for example, as they would never land until they ran out of fuel)
So what does it mean when HNers 'vote down'? Are they agreeing with something, or disagreeing? Are we assuming they're trying to impose a nuanced influence on a particular part of a person's commentary? If it was a generic part of their comment, was it their moral or ethical stance, or their writing style, or the correctness of their comment? We have no idea; all we know is they wanted that comment to go 'down'.
A simple vote down, regardless of its supposed intent, doesn't provide any specific benefit if it comes without context. The lack of a down vote would do the opposite, however: it would force people to only be able to provide positive feedback, or no feedback at all. How does this effect group dynamics? It actually works on individual behavioral conditioning: provide positive feedback, and you get positive behavioral conditioning, and negative behavior gets weeded out automatically (because the individual wants to be rewarded, which it knows it only gets for positive behavior). This is what practically all dog-trainers and child psychologists suggest as the best method to promote positive behavior. This would have two similar effects: people would be nicer, and they would provide 'better quality' comments.
So I think 'downvoting' is a loaded and useless way to regulate comments. I think that valid alternatives would be votes for specific feedback, or if you wanted to provide a way to simply "move thread position", a button that said so specifically. These would create specific, immediate changes in the group dynamics and the interactions of the users, unlike a generic downvote button. If you're trying to prevent groupthink, give the users tools to allow them to express themselves better.
And lack of being contrary, is more conducive to groupthink.
>And being well-written and productive doesn't mean the comment is insightful, intelligent, or informative, which one assumes would be a preferred top comment.
Only if you're using a non-standard definition of productive. Could you spell out what you mean here? In common use, a productive discussion is one in which at least one side emerged with a better understanding of the issue. (If you can do that without being insightful, more power to you!)
This contrasts with a discussion in which arguers blur important distinctions, talk past each other, and so on.
It may be helpful (productive?) to give an example of (what I consider) a comment that is insightful, informative, and correct, but unproductive: https://news.ycombinator.com/item?id=7840204
In this case, it's because the poster simply reiterated the logic behind the point his critics already accepted, thereby doing nothing to convey understanding of why they should take his position.
So, I don't think you actually disagree that "productive = good" here, but have non-standard criteria for what counts as productive.
Feedback for the comment writer. Shows the difference between "nobody thought much of my comment either way" and "some people really disliked my comment".
I think it's been very effective at weeding out the completely pointless comments (memes, jokes, "this", "I agree", etc.) and is also somewhat effective at weeding out subtle incivilities (I've changed my behavior in response to downvotes before.)
The discussion is better if those comments are lower on the page and, ideally, discouraged from being made at all. Downvoting does both.
> Then all you have is upvoting, which is based purely on merit. It sounds like people don't like the idea of forum meritocracy, though.
I don't understand what this means or what a "meritocracy" is in this setting. (The moderators are in charge; the highest ranked users will never form a ruling class or be in power. Most forums are not "meritocracies", "socialist exeriments", "democracies", "plutocracies", "oligarchies", "monopolies", or anything that implies something other than a loosely managed commenting system.)
HN has many readers that know an absolute shitload about different subjects, and I'd like them to share interesting information and anecdotes about interesting articles. I don't want to read stupid jokes and memes on HN, because there are other places that are better for that. So I like that there are ways to explicitly encourage certain comments and explicitly discourage others.
Yes.
> I don't understand what this means or what a "meritocracy" is in this setting.
Basically, thread sorting would be based (mostly) on merit.
> the highest ranked users will never form a ruling class or be in power
The highest ranking users have high social power. Their power is simply a tendency to steer the group without telling anyone what to do. Usually it's unintentional, and sometimes they have no idea they're doing.
Most forums are actually discrete social communities made up of ingroups, with their own cultural standards and hierarchies, and often a few custom heuristics. Some of them are really incredible from a behavioral standpoint, like an internet chat galapagos island. HN is not so special, but it does have a clearly defined social structure.
I agree with you that the stupid jokes, memes etc should stay off HN. That's why I think some kind of meta-voting is necessary: because a simple downvote is used for a lot of things right now, and not just "this-comment-is-out-of-bounds" like you're talking about. Barring meta-voting, removing the downvote keeps the troll talk but it still filters down to the bottom with the other useless comments.
On Reddit, the Reddiquette [1] standard is that you should only downvote if a comment doesn't contribute to the conversation. I think is is an excellent standard. It's been suggested in the past that a similar standard should be present on HN. [2]
Considering that flagging is a privilege that can be lost if someone decides that you're abusing it, ideally it should be reserved for the egregious cases that are unquestionably bad. Downvoting on HN should be for comments that truly don't contribute to the conversation, though the comments in question may not rise to the level of flagging.
As an example: Say you have a comment expressing an opinion with no support, no references, and factual errors. Is it "inappropriate"? Arguable. Does it contribute to the conversation? Not in any positive way. I would suggest downvoting that comment would be the right answer, even if you agree with the sentiment, because it doesn't positively contribute to the conversation. But flagging wouldn't be appropriate considering the comment is on topic and not spam (which are given as criteria in the HN Guidelines).
If you encourage downvoting because you disagree with an opinion, then minority opinions get buried. And sometimes it's useful to see that an opinion isn't unanimous, and that there's another side to an argument, even if you disagree.
[1] http://www.reddit.com/wiki/reddiquette
[2] https://news.ycombinator.com/item?id=721314
When I disagree with a comment, I'd rather write a dissenting one or a reply stating my counter points. Another method of dealing with disagreements is to simply upvote other comments you agree with.
I believe that pg himself disagrees with me on this, but I honestly don't think the site is designed correctly with his view of voting in mind. The punishment for downvoting shouldn't be so harsh (both moving the comment down and fading it out) if it's just to express disagreement. That just encourages extreme group-think.
By far the best way to express disagreement IMHO is to neither vote nor flag, but to post a response.
I sometimes wish downvoting comments that have no response was impossible period, but I have no idea how to deal with the edge cases. Still, it's frustrating to see responses (actual ones, not just a bunch of insults etc.) made in obviously good faith greyed out without anyone pointing out what is wrong with them. When I do know, I try to reply, but usually I don't feel quite qualified. But the person who downvoted should know, right? If they can't put it in words, can it really be that good a reason? Among people who mostly type very fast, mostly idly browsing this site instead of being in a hurry, the time it takes is not an excuse, either.
To me it says someone has enough time to care about something someone said being somehow wrong, but not enough time to tell them what it is; that somehow the fact that they said something wrong makes them unworthy of being told what was wrong about it. Oh, and that nobody else who doesn't "get it" doesn't deserve to learn, either.
Certainly not.
Up & downvotes determine what is shown on the site, and what is prominently featured. Is your goal for this site for it to be full of things you agree with? I hope not. That's not my goal.
Now, there are varying opinions on the issue of downvoting for disagreement. Regardless, it is certainly never officially stated that this is the purpose of downvotes.
No, that is NOT correct.
Downvoting is a for useless comment. Flagging is for spam and abusive comments.
If you disagree with the comment, but the comment was well made and on topic you should UPVOTE it, not downvote it.
This is one of the main differences for HN vs Reddit, and it's one of the main reasons this site is so much better.
You shouldn't. For example, stories about big players (e.g. Apple, Google, Microsoft, Facebook) almost always produce flamewars, but the discussions can also be interesting and informative.
You should flag stuff only when it doesn't meet guidelines. For example, most news stories (e.g. "Man murders family, followed by car chase") are inappropriate for the site.
It's restored now. I don't think I've flagged many, if any, stories.
I flag any comment I perceive as an attack on a person rather than the ideas of that person. I don't feel that's appropriate. And in particular it's not what I personally want HN to be.
If a comment makes a point in what I feel isn't a good faith manner, I'll merely downvote it.
If I just disagree but it's a fair point, I leave it alone.
Maybe it will take an updated policy to fix the core issue, which is that while linkbait headlines are annoying, some original works are titled very badly and the poster may actually add useful information by customizing the title (often by highlighting what is most interesting/relevant about the linked article).
So I guess I am not able to help.
The problem is this comment was a total fluke really. I have _tried_ engaging in comments before with the results of only a vote or two usually.
Had my previous comment not been up-voted so much, it likely would have taken me another 10 comments to reach the threshold for just flagging.
1. Most of the paywall news sites allow free viewing of a few articles per month and so many here do not hit the paywall.
2. It is often very easy to get around the paywall if you are over the monthly limit. Sometimes it is as simple as searching for the article by title in Google and clicking Google's link.
3. If you cannot get around the paywall, it is often the case that you can find coverage of the same event at another site that is not paywalled, which will give you enough information to meaningfully enjoy and participate in the HN discussion.
4. Even if you cannot find free coverage of the thing under discussion, it is often the case that the HN discussion is entertaining and enlightening. HN discussion often goes off on tangents inspired by the article, and those tangents can be appreciated without having read the article.
5. Many people here don't see the paywalls because they subscribe to the paywalled sites.
6. Most here do not have such big egos or senses of entitlement that they are bothered by others reading and discussing something that they have not been able to read.
I've been under "rankban" for at least a year, if not longer. This means that my posts, even if they get a lot of upvotes, fall to or near the bottom. I don't really care, and while I could use sock puppets to get around that, I'm too old for that shit and I don't really care. Still, it's irritating.
I've used HN Search (https://hn.algolia.com/) to verify comment karma and validate my suspicion.
I'm also on "slowban", as Hacker News performs worse when I use it while logged in than when I do so in incognito, logged-out mode.
It's dishonest bullshit. My main reason for continuing to comment is (a) to spite the moderators and (b) because there's a certain joy in getting +10 karma even when your post falls to the bottom.
(dang has commented about marking his own comments off topic, so the support for that is in there, but I'm speculating about how it might be used.)
This was years ago and I have no idea if the mods still use this tactic, but I imagine so in certain cases.
I mailed the moderators and one of them confirmed it had been marked "off-topic", and left it there, despite admitting it wasn't off-topic, because he did want to cut down on "negative" comments on new products.
I understand the stance, but disagree with it, and have taken it a fact of life that both the moderatorship, and the userbase at large, are unwilling to consider strong posts that aren't full of praise as worthwhile. I can even mildly understand that, since it takes more effort to verify strong criticism as being grounded, than it takes to verify strong praise. It's disappointing nevertheless.
It's possible that the ranking algorithm is actually helping your karma count. I personally only up/down vote comments when I feel they're in the "wrong position". If you made a comment which I found good and it was at the bottom, I would up vote it whereas I probably wouldn't bother if it was already at the top.
While it's my impression (from other comments, &c) that your claim is accurate, this alone isn't necessarily proof. I would generally expect a site to perform better for logged-out users, as 1) it gets to skip any tasks that are for logged-in users only, and 2) it should tend to have everything it needs to render that version hot in whatever caches might be involved...
The first is certainly more likely for a host of reasons.
This appropriately rewards engagement, but also pushes down other worthy comments that come slightly later, often to effective near invisibility when replies to the top comment dominate the discussion.
I wonder if there is any mechanism that could help counter this.
However, I often see new (top level) comments on the top when entering a discussion here on HN, so it seems they at least get some exposure.
* On threads with many comments, few users have time to read all or even most of them. Perhaps find a way to shorten the list, improve signal-to-noise, and surface newer comments by burying older comments that haven't received votes.
> This appropriately rewards engagement
* It rewards engagement by people who have the time to frequently check HN, read linked stories, and comment throughout the day. My guess is that the smarter someone is, the less availability they have for such things. I would create a system that rewards engagement by busy people.
So essentially the same kind of system that is used for the link page? I honestly don't see why link aggregators use one system for links (based on time posted and points) and another system for comments (based solely on points). A very engaging link and a very engaging comment both share the same qualities, whether it is informativeness or adding something interesting to the discussion.
That way we can reload the page and see what's new.
Here's what is currently offered:
https://news.ycombinator.com/formatdoc
Basically italics and code blocks only. No lists or no bolding.
On that historical HN note, it would be great if someone made an app that dug up really awesome older stories from the HN archives. I'd use that a ton!
It adds no new columns, and better still if you pick a symbol which is nuked by TRIM() e.g. \x0B [U+000B] (vertical tab) then it might be ignored by existing infrastructure/trimmed.
This seems like one of the cases where HN's moderation is both draconian and completely opaque. I can't flag comments, and I have no idea why. I certainly don't think of myself as an abusive user.
Maybe I accidentally clicked the "flag" link once. If that gets flagging privileges removed, HN should at least consider creating a way to undo an action!
Make sure you're clicking "link" through to the comment's page -- the flag links don't appear when comments are shown in a thread.
This has been the source of much confusion and I'm not 100% sure on the reason why it was implemented.
So I agree, it would be great to have more/better undo options!
Something more like this..
note: they are supposed to be the same size.. oddly unicode down triangle is rending larger for me.I also find the single leading upvote to be a bit more consistent, as you are not able to downvote replies to your comments, only upvote. So in that regard, the UI would be a bit more uniform regardless of whether the comment was a reply or not.
Another straightforward possibility could be replacing the arrows with words, and making the time a link to the comment (removing "link" from trailing actions as gojomo suggested). This may result in better support for screen readers, in addition to mobile users, as well.
We never remove flagging privileges because of just one flag. The concept of a mistake is all too familiar over here.
1. https://hn.algolia.com/?q=author%3Adang+flag+link#!/comment/...
Is it possible to either allow people to upvote a comment they have previously downvoted (or vice versa), or at least to undo an accidental upvote/downvote?
Especially on mobile devices, the margin of error here is very low, and I know I have accidentally downvoted a comment that I meant to upvote.
(In addition, I know some people may upvote/downvote a long comment before they have finished reading the entire text. This probably isn't behavior that should be encouraged or catered to, which is why allowing people to change their mind on votes[0] may be overkill. But at least allowing people to revoke an incorrect upvote/downvote might mitigate this effect.)
[0] ie, the way Reddit does
I tend to pre-emptively flag stories which I know will draw out the worst bits of HN. That's probably a bad thing, and I wouldn't be surprised if it caused my flags to be ignored.
I can see why you would flag a whole thread, but I don't think that will ultimately get us anywhere. If you care about the accessibility of this site, then I implore you to upvote those threads^ and flag the bad comments within.
^presuming that they have some techy slant
Anyone worried about this is welcome to email hn@ycombinator.com and ask, and we'll be happy to take a look for you. But I feel like the kind of people who would worry about this probably tend to be the kind whose flags are useful.
I think at the moment it's too easy to hit that flag button on the front page (as others have said, especially on mobile), so maybe it should be moved back too?
This (a) draws attention to the exact thing you are flagging (the individual comment/link, not the thread) and (b) puts the confirm link somewhere else on the screen so you can't double-tap the same hotspot by accident. The advisory text is optional, but as others have pointed out, it's not really clear where the line between flagging and downvoting should be drawn.
the flag link is right next to the comments link.
it's very easy to accidentally flag something from the home page of hacker news, when you actually intended to read the comments.
You can also unflag stories.
Sam & co, great job on experimenting and improving quality, it shows. I find this statement to be really confusing considering there is no guidelines[1] for downvoting and, in my experience, it is highly used to put opinions you don't personally agree with out of reach for the majority of users to read.
[1] https://news.ycombinator.com/newsguidelines.html
When I first joined HN, I aggressively flagged stories and comments that didn't meet the guidelines. I thought that's what I was supposed to do. However, one day my comment flagging privileges disappeared with no explanation. If someone had contacted me and explained how I was misusing flags, I would have happily self-corrected.
It wasn't clear either to me or the moderator what had resulted in the loss of privilege. On the one hand, that made it easy to decide to reinstate the privilege; on the other hand that made it hard to learn what I'd done wrong.
Comment flagging privileges don't disappear (edit: alone). Please see https://news.ycombinator.com/item?id=8585939.
But I am still confused about the apparent contradiction with what you've stated versus what was stated in the blog post, about losing flagging privs. When flagging privs are lost, are _all_ flagging privs lost? Or is minimaxr correct[0], that it works the other way, where you can lose story flagging privs without losing comment flagging privs?
[0] https://news.ycombinator.com/item?id=8586144
Edit: Ok, I checked it. Story and comment flagging privileges are lost or retained as a set. You can't have, or lose, one without the other. It's not clear why minimaxir thought otherwise, since I'm pretty sure the code has always worked this way.
I suspect that I'm not the only one who downvotes the comments he doesn't like or disagrees with, but who'd very rarely want to push a comment into gray. I mean, it's one thing to vote a comment down so that it won't float at the top and another is to punch its author in a face with a negative score.
You are only supposed to downvote comments that are useless.
Sometimes I check the comment history of these dead posters; 80% of the time it appears like they are a genuine troll who was correctly moderated out of the community, however, a significant portion of the time, there doesn't appear to be anything particularly inappropriate in the comment history, but the user is still doomed to waste a considerable amount of time attempting to contribute to this board.
In the context of internet discussion boards, it seems a little harsh.
Off the top of my head, I feel like something akin to an automatic probationary resurrection period would be an interesting idea. Perhaps it occurs one year after hellbanishment, giving the community a chance to organically reevaluate the user's quality of contributions, and giving a second chance to those who have matured in a years time (or who was simply hellbanned in error).
It's pretty mean spirited though - generally a "douchebag" move. It'd be like having a bad employee, but instead of firing him, or discussing his work, just don't bother paying him any more.
After being hellbanned I actually realised that commenting on internet forums is toxic, generally a waste of time, and not productive. So I quit. At the start, you sort of care about "karma". But you end up realising it's a measure of two things. 1. How much time you waste commenting on the internet, and 2. How much you can agree with the groupthink echo-chamber.
But for those that like spending time commenting, I would have thought following a subreddit model would work better here, with the growth. Spread the power out. If I get hellbanned from the "rust is the future!!!" subhackernews, big whoop.
It'd also help with filtering out all the non-interesting (self driving cars) stories.
But then perhaps it'd basically be reddit at that stage which would defeat the point...
We don't ban people because they "critiqued a YC backed company".
Edit: Some of the replies have made good points, and I realize that I overreacted. Sorry about that. Please shoot us an email at hn@ycombinator.com if you feel your account was banned unfairly. We're always happy to look into this—there's no question that we make mistakes; the most I can claim is that we're eager to, and do, correct them when they're brought to our attention.
(Edit: I haven't forgotten, but have to run out to an appointment, so this will need to wait for another hour or two. Sorry about that.)
Edit: back now. My gut feeling is that PG might not have had enough time to look into all the details. I say that because June 2012 was about the peak of when HN moderation was extremely time-constrained (I started a few months later) and the only option was to enforce the guidelines generically.
For the record, I don't think he deserved a hellbanning, but then again, this was a while ago. And the wild west hadn't probably settled yet.
So it sounds like got banned for refusing to follow the guidelines?
https://news.ycombinator.com/newsguidelines.html
Please don't post on HN to ask or tell us something (e.g. to ask us questions about Y Combinator, or to ask or complain about moderation). If you want to say something to us, please send it to hn@ycombinator.com.
I'll admit it could be clearer but it has always been there since that page went up as far as I'm aware. Ctrl+F for hn@
If a moderator has a button "ban", and you piss them off enough, you'll get banned.
Give people buttons, and they'll use them. Probably more than they actually should.
As I suggested in my original post though, I have more useful things to do than look back at comments I posted years ago and compile evidence. I've moved on.
I'm sure it happens a lot more often than you realise, because in general when people get hellbanned, they come to the same realisations as me.
Not only did we never ban people for criticizing YC companies, the very first thing PG told me when I started moderating HN, and the thing that he emphasized most strongly after that, was never to do things that could be construed (or misconstrued) as censorship of anti-YC stories.
(I've edited out some irritation that leaked through in my original version of this comment.)
But it would be despicably dishonest for you guys to deny that routinely you guys do things here to protect YC companies (including manipulating voting points on comments/ submissions).
One example (on the top of my mind) is the drchrono post: https://news.ycombinator.com/item?id=7178004
Even though Skeletor made that comment like ... 20 days after the HN submission, it somehow found its way to the top. Obviously, this was through manual action. Obviously, a non-YC company is not afforded such a privilege.
Idling in #startups of freenode (unofficial HN channel), I've heard too many of these stories. The stories of rankban upon some critical comment on a YC-funded company, a slowban because of a critical comment on some YC personality, etc. etc. There are countless examples.
When these things happen one after the other, you lose trust, we cannot believe you anymore. Please stop doing this. I think the only way to win trust back at this point is if you again expose voting points at all times.
It isn't hard to deny that general impression you're reporting, because it isn't true. But I don't see any way to refute such a sweeping claim convincingly. As far as I can tell it exists only on the level of rumor and is unanswerable. But I'm happy to reply about specific cases.
In the Drchrono case, we got an email from the founder asking to post a response after the commenting window had closed. I agreed, but not because this was a YC startup, but rather because we would do this for any startup in this situation, and have indeed done so for at least one non-YC startup I can remember. Before I agreed to the Drchrono founder's request, I told him I needed to make sure that we would do it for any startup in that situation, and I thought long and hard before concluding that was true. It was by far the most important factor in that decision.
dang, I understand that you, in your position, have to be mindful of optics, you have to think of ways to say things that are best for YC. That's great, you should do that. All individuals of a company who have a public presence have to do that. The thing is, you must not focus entirely on optics -- you absolutely must in your heart have the right view. Not only because if you don't, some people will eventually find out what's up, but also because it's the right thing. Actions that favor YC companies (beyond a certain line) on a public forum such as HN are unethical. I think there have been enough things done at this point that the only way that trust can again be restored is by having more transparency -- for example, by showing comment scores at all times in some way.
You're doing an absolutely super job, probably far better than most or all of us here will ever realize simply because moderation when done properly is all but invisible so don't sweat it, this is a thing that is as far as I can see not solvable in the current set-up. Those lines were drawn long before you showed up and within those lines you're doing the best you can.
FWIW I too recall several instances of users that were banned imho unjustly as well as some threads where the pro-YC bias broke through but over the vast amount of content generated here those are very very few instances, not by far enough to claim systematic bias or to be used as evidence for some nefarious plot. More like genuine mistakes and things done in the heat of the moment. And on later reflection some of those were reverted.
(If you want I can probably dig them up for you but you're busy enough as it is.)
We want to see such comments highly ranked regardless of YC affiliation. If you post such an important update and don't get organic upvotes because the story is old, plead your case to hn@ycombinator.com.
Just imagine how useful people might find it to simply block posts by keyword, or not show threads with a certain ratio of upvotes to downvotes, or be able to train their own Bayesian filter.
They wouldn't even have to change the UI (which they seem to not want to do) apart from what a particular user sees.
Then again, there does appear to be a confligt in the way the HN community seems to be growing, with the site itself still geared towards providing a relatively low amount of content through a single channel. Stories overwhelming the site and reposts are a known problem, and probably almost no one bothers watching past the first page of /news so those top 30 slots might as well be the entire site. Although, many of the most obvious remedies to this would make the site look more like reddit, and we can't have that I guess.
When that kind of response comes from the site's moderator, I really don't see the level of toxicity improving any time soon.
It's likely statistically factual, but in context it's just another of the "mean, stupid things" that Paul Graham called you out as being here to address. And you appear to have done nothing to investigate whether the previous user's post was factually correct before slinging personal accusations.
If this gets me hellbanned too, so be it. Conversation and community on this site is a toxic mess that leaves people afraid to post anything. The main good thing is following users like patio11 and tptacek.
http://blog.ycombinator.com/meet-the-people-taking-over-hack...
I try hard not to let personal irritation leak through in my HN comments, but I do fail at it. The most I can claim is a willingness to correct mistakes.
People have wildly divergent views as to what's appropriate, what's a little rude, and what's over the line. This means that even if you really truly believe that there was NEVER a mistake made during moderation, that some people will truly believe what they said.
Good moderation requires a ton of empathy and kindness.
If users could look at the actual record, their perennial sob story of perfectly reasonable behavior struck down by bullying censors would evaporate. So they make new accounts and post statements designed to be unanswerable.
I interpreted that as dang attributing essentially every complaint to malice, and simultaneously dismissing all other explanations.
Which is sadly lacking among many, if not most, moderators of online communities across the net. I'm not saying that's the case with dang; in fact, I wouldn't know. But it's a thankless job that is akin to working in a call center without pay. It takes a strong personality to keep one's head above the layer of filth floating atop the waters of discourse.
> Good moderation requires a ton of empathy and kindness.
I try, but don't always succeed. Thanks for the reminder. I appreciate it.
There's no way, barring some freak outlier, that we banned anyone for criticizing YC or a YC-funded startup. If someone really did feel that way, nothing would be easier to clear up. The real issue, in the overwhelming majority of cases, is repeatedly flouting the HN guidelines.
That's exactly what happened here. (And maybe Daniel can edit his comment.)
Daniel does a really good job, and is extremely responsive by email and on here. It is obvious where he wrote "Ever notice how people who make claims about why they got banned never provide links to the posts in question?" is borne of deep frustration. He would like to follow those links and improve the site, but can't. It's obvious that his comment is written out of frustration.
Let's be very clear: hellbanning is the worst and rudest thing that exists on any respectable Internet forum. Hellbanning literally wastes hours of the time of people who contribute great insight for free. The comments on this site are good and provided for free by people. Hellbanning turns this goodwill on its face, like a goodwill jar you can put bills into but which go into a furnace.
Daniel (and PG) knows very well that hellbanning is a nuclear weapon and the rudest thing that any Internet forum can possibly do, that is actually being done.
You have stories of people only learning they were hellbanned after literally taking the time to email someone a link to something thoughtful they had written. A lot of hellbanning has been (historically) in error.
It is one of the main reasons that I would never consciously leave a comment up if it reaches -3, even if I stand by it 100%, it's important, and the community happens to be wrong in its groupthink and I clearly have explained why. I would delete it instead.
Note that I have learned this behavior, and so have other contributors on this site.
It's one of the things that makes this site great.
So even though it is a nuclear option and the worst, rudest thing that any respectable forum does, in the sense that time is money literally stealing from users, and stealing donations at that and throwing them away, at the same time it is one of the things that allows this site to function as one of the best sites on the planet.
So you can bet that Daniel is extremely serious about following hellbanning claims and improving this process. It is difficult and he walks a very fine line.
He's doing a fantastic job at present in a very difficult undertaking. Kudos, Daniel, and keep up the good work. I can read your comment for what it is :)
I am confused. Your preferred path is to avoid conflict such that you would rather delete than be disagreed with? If your opinion differs from groupthink, you would make it go away? I guess that is similar to not posting in the first place (because of groupthink you disagree with) , just retroactive.
Probably better than my not posting in the first place :)
edit: I guess the thread was getting cluttered and argumentative.
we are all, for good and for bad, unavoidably human.
Replying courteously even to baseless and ranty accusations is probably a good idea, but I can't say I'd blame someone for not doing it.
I see the mod(s?) deal with this kind of conspiracy accusation almost on a daily basis, and I can definitely see how patience can wear thin when every nut does that. Dang implying that this accusation was baseless was probably rash, but that's the only thing I see even remotely out of line here.
Moderation is a pretty thankless job, and like sysadmins, people never appreciate you for getting things to work right when they work right. The definition of success is invisibility to users. But the second it goes even a bit wrong, it's a shitstorm. Here's yet another piece of empathy that one needs to consider.
But in the places I have been banned, it's not been my fault. The people who say "we don't ban people because of X" are, like most people, just telling themselves a pleasant lie.
There's that one guy, who while never doing anything outrageously obnoxious, rubs you the wrong way. And eventually you're going to find something borderline or even milder than that, and use your petty powers. This is human nature, I'd almost certainly do the same. Everyone would.
As just one example I'm the resident "soldier" for the government (as one HN user was so nice to call me), I often post comments that are in opposition to many of the tenets that are popular on HN, and yet I've not been banned, my karma is not negative, and those I debate with generally treat me and my arguments with respect. If it were just up to ideology I should have been hellbanned by now, and yet here I am.
Go ahead and hellban me. If you strike me down, I shall become more powerful than you could possibly imagine.
1) you ban people who are rude and who you disagree with philosophically, while you do NOT ban the equally rude people who you agree with.
For evidence of the above, look at users like etherael (and his other names; not sure if he tors/vpns or if you can find them), who are raging assholes on a semi-routine basis, but who aren't banned because Libertarian BitCoin Lover matches your values. And let's face facts, you're less willing to ban people who agree with you, even if they're toxic assholes.
If you banned people who you agreed with for the same exact crimes as those you disagree with HN would be a better place.
As it stands, people who agree with you are allowed to be ruder and more toxic than people who disagree with you. This is used as a game by some of HN's worst users who brag on IRC about how it's fun to try to engage in flamewars where they don't get banned but the other individual does.
edit:
Not to mention other game that's played by a lot of folks, which is to be as big of a dick as is possible without actually using openly aggressive language. The goal there being to generate an emotional reaction while retaining some semblance of plausible deniability, because everybody knows that you won't ban them for "polite" taunting, even if it's toxic shit that can't go anywhere useful or interesting.
As to your claim that you aren't soft on Libertarians... LOL. I've actually A/B tested this, saying the same toxic shit but with the positions swapped. Upvoted in one case, banned in the other.
Part of the mechanic might be that you never even see the Libertarian version because it doesn't get flagged in the first place, despite being identical in profanity, intelligence, and toxicity to the version that does get flagged. but whatever the causes, there's a real and definite bias as to what positions are tolerated here.
---
Response to below: Etherael gives good conversation like:
"Congratulations, you are a retard."
"I have no hopes for you at all."
"Get off that lawn, you old crank."
"Why is this being downvoted? Are people really so naive as to not be aware of this?"
---
" Because when you present a point in this way with this kind of language and in this kind of behavior, you just lose everything."
You're right. I have the same amount of respect for dang as he has for the public at large. And that's my fault. I should be better, even if I really believe that HN is worse and less useful each year, and that dang is poorly suited to moderation.
He's probably a smart guy, but we're all just apes, and he's one who isn't the right fit for the job.
Even if you had a point, you just lost it. Because when you present a point in this way with this kind of language and in this kind of behavior, you just lose everything.
I'm a pretty skeptical person (see my skepticism on my last comment, for example!) who views most actions by most companies very skeptically and tries to see if there might be ulterior motives. And I've gotta say, you're wrong in this instance.
> As to your claim that you aren't soft on Libertarians.
They're not (they're not soft/hard based on ideological beliefs). For example, DanielBMarkham, perhaps the most outspoken libertarian on this site, is rankbanned. Now, I do have serious doubts about whether or not they're "soft" on folks saying negative things about YC companies/people.
Lastly, I've gotten to know etherael quite well -- he's got a sharp tongue, but he never quite struck me as an asshole. I do know that he's very talented at what he does, and almost always provides good, intelligent conversation about anything I bring up to him - and in that way, fits right at home here on HN. I guess though maybe you caught him in a bad time being especially rude? The best of us lose it sometimes. I hope etherael is more thoughtful in his future replies.
Basically I have no idea what you're talking about.
You're so full of shit it's fucking ridiculous. Just fuck off back into your whole, you professional troll.
I didn't deny that
> and even there I try to avoid doing so.
I just said that, like for example right now.
Thanks for confirming that you're an asshole who is happy to subject thousands of people to valueless bullshit because you require the last word.
If dang had balls, you'd be banned for that shit. But he lacks balls when it comes to idiots like you, Dan.
The fact that you aren't banned is 100% proof tat dang is a completely gutless and worthless moderator.
although maybe it is a ponzi scheme for the miners now that I think about it.
I think there's a lot of outspoken libertarian/anti-statist types around here, and that's fine, but I don't think it's even the plurality among political stances of HN readers. I suspect being invested in politics to the point that you'll regularly engage in political discussions online is very strongly correlated with holding atypical political views (I include myself in that set).
>their claims are nearly always false.
>perennial sob story
Hmm, I wonder where these fall on Graham's disagreement pyramid...
Seriously, how hard is it to tell the guy "shoot me an email and I'll look into the ban" and then make your judgment after that? Maybe the guy's bullshitting us. Or maybe he really was banned unjustly. How can you possibly know?
You're absolutely right. I'll add that to my comment.
Why not tell me why I was hellbanned then?
Here's my comment history. Such bullying, so horrible, wow: https://news.ycombinator.com/threads?id=ahoyhere
The only conclusion is that HN hellbans dissent. I'm about as far from a troll as you can get.
It's your platform; do what you want. But don't lie about it.
You called Alex (whose friend had died and he didn't find out for two years) a "shitty friend" and said he didn't "think of, or care for, others".
I don't know if that's worthy of a hellban. But I do know I wouldn't consider the offending comments mere "dissent".
I didn't call him a shitty friend. He wrote, "Now you’re probably thinking I’m a real shitty friend" after his explanation of the events — which I quoted, in quotes, very clearly a direct quote from his own words — and I wrote "Yes" to answer the question. 95% of my comment was advice for people on what to do in the situation that the friend is reaching out to them as that friend did, who got ignored and later died and that person didn't know for years.
If you write an essay about something you did that you think was bad, and post it to Medium, and post it to HN, and then say "You probably think I did something bad," is it reaaaally so controversial to comment with "Yes, I agree, I think you did something bad"?
Also, if you thought that was my last comment, maybe you don't see my actual last exchange? It was about my bootstrapping conference and upcoming web site. Perhaps it's truly hell banned as in not visible? It was this:
http://skitch-replacement.s3.amazonaws.com/ahoyhere_s_commen...
It's a direct quote of a dead comment you wrote in a reply. Turn on "showdead" to see it [0].
The dead "think of" comment in the "shitty friend" thread appeared 405 days ago. Your comment about the bootstrapping conference was 406 days ago. (There's also a dead reply in the bootstrapping thread dated 405 days ago, which fits the timing of being hellbanned for the "think of" post.)
[0 - edited my comment to add this footnote] in response to sage_joch's response to you in that thread. Link: https://news.ycombinator.com/item?id=6473532
I never said anyone specific is self-centered. I said:
>How self-centered do you have to be to not even wish a friend with obvious health problems "good luck" or "feel better"?
That's a rhetorical question.
Then, in the dead comment, I wrote:
> Being a bad friend in this way doesn't make you a bad person…
> But it does make it seem kinda iffy to write a blog post about it, even include screenshots of the conversation, and not (apparently?) be socially aware enough to realize that the deceased man was nearly begging his friend to express some interest and concern. And so many of the commenters, from my perspective, were not picking up on anything the OP did not explicitly lay out in the essay itself, which is to say: his friend was telegraphing his problems in every possible way, and the OP ignored it. People seem to be reading it and thinking, "Oh, just one of those things." But it's only one of those things if you don't think of, or care for, others.
Again, I didn't personally attack alexcap. You are quoting it out of context.
Real inflammatory stuff. If this is what I got hellbanned for, it really makes me wonder about the actual, deliberate, specific, personal cruelty that goes by on a regular basis without banning.
Meanwhile the other dead comment… the one that's actually newest, not the one you claimed above was the newest… is about my bootstrapping conference and upcoming site:
> Oh, good question. I can see why you'd think that. > Nope, BB will be more like HN, but specifically for bootstrappers and related topics only. IOW: public-facing, free to use. (Although I think we're going to do a MeFi-style $5 or $10 join fee, to encourage good citizenship.) > It's not going to be a product. I'm not going to run it unilaterally, either. It's for the community.
That is the last thing I posted before being hellbanned.
Your other dead comment was probably posted after you were hellbanned (the point of a hellban is that you wouldn't have known it happened right away). That's why both comments are dead. If you'd been hellbanned for the last comment, only it would be dead.
It probably helps that I'm a "nobody"; I'm not a member of the startup culture, just an outside observer and occasional commentator. If it's true that politics are a factor in bans and heavy-handed moderation, then I can understand (though not agree with) you being hellbanned for what are fairly mild comments as judged by an outsider like me, as you are a member of the startup club.
And I hate to say it, but misogyny might even have played a role in it. I have zero evidence of that and it's not meant as an accusation, just that it's a remote possibility. It's a real problem in just about any internet based community, and I really doubt this one is immune to it.
Post a list of comments which were [dead]ed under the new system but would not have been [dead]ed under the old system.
If the list contains a bunch of comments like "u r a gay homoz", you'll make a pretty convincing case (both to yourself and everyone else) that the new system is awesome. If the list contains a bunch of "I'm concerned about the security implications of ordinary users putting significant money into bitcoin..." then maybe the new system isn't so awesome.
[edit: I realize it's probably too late for this to be seen, perils of posting from IST.]
But people are assholes. And they found a few people they didn't like (for good reasons, for bad reasons, for none at all) and did the meatspace equivalent of hellbanishment.
Yeh, here, you're just wasting time on the internet, maybe you can stop caring about it. But people do this the world over, and there's no escaping it. This is what people do. They're mean fucks, and if you don't fit in, you're just left out in the cold.
Yes.
... But for those that like spending time commenting, I would have thought following a subreddit model would work better here, with the growth. Spread the power out. ..."
Agreed.
However, if the person isn't actually out-and-out to mess with your site, it sucks. I've seen lots of communities go down the tubes because the moderators get busy with life, and then Something Happens On The Forum, and moderators passive-aggressively go "fine, we're just gonna act like this, and if you don't want us to do that, then you should make the community act better!" Basically announcing that they are going to put in minimal effort to moderation, and the community definitely notices.
Professional moderation, like HN uses, seems like the best way to go.
For example, there's a certain GNU/Linux distro that is maintained by a core group of devs who have become overtly hostile to any new users of their project, and actively seek to discourage "newbies" from seeking help and getting any benefit out of the project. One would think the toxic atmosphere would have killed the distro off long ago, yet it's maintaining popularity and even seeing an uptick in a certain niche community. I certainly don't understand how it thrives with such a rotten core; yes, it's overall a very well done distro, but even the best product normally can't survive that kind of cancer.
That's not to say that HN would ever end up like that; indeed, from what I've seen they are doing an excellent job overall with maintaining and moderating this community. I just hope it continues to stay that way or improve, instead of going down a dark path towards chaos.
You don't have to spam or be disruptive. I'm putting my account name on here so folks can look at it. I was new to HN, didn't know much, had no idea I was dead. :)
This account get's a decent amount of flag too when I make unpopular opinion, I have to stop myself from censoring myself because the moment I feel like I can't talk or participate, I will just leave the community. If I can't express myself around a bunch of "hackers" then what's the point?
(I'll add some details in a few minutes. Edit—well, quite a few minutes.)
The problem is that banning an account is an all-or-nothing system where inappropriate commenting is not necessarily an all-or-nothing phenomenon. The solution we want to try, which was suggested by several users, is to give the community the power to bring comments out of [dead] status. Letting fellow users make the call when a comment is (or isn't) ok seems like it might work, especially since the recent experiments with flagging seem to have helped. This is one of the things Sam was referring to when he said we plan to experiment more with community moderation.
Four good effects, and it will make hellbanning more effective (because occasionally there will be answers to a comment). Five good... Wait, let me come in again.
Arguably, if a banned account is making reasonable comments worth responding to, that account shouldn't remain banned.
Maybe after a certain number of comments a ban should be lifted entirely. Unless that's how it's meant to work anyway?
We wouldn't call these "dead" or "banned"; we'd probably introduce new statuses like "flagged" or "moderated" or something like that. The current form of banning would be reserved for obvious cases of spam, etc.
I did like my old account.
Did you email hn to ask why you were banned?
https://news.ycombinator.com/item?id=4335811
I'm guessing that got caught in the spam filter for use of the words "jewelry" and some of the other keywords in the Amazon link. This is one of the unfortunate side effects of running a startup that happens to be in a field that generates a lot of (affiliate) spam.
Hellbanning people, and then asking them to email you later, is still a crappy way to act.
It's not a hard problem. Just raise the threshold for hellbanning. Use it sparingly.
[1] Coming Soon to Hacker News: Pending Comments
https://news.ycombinator.com/item?id=7445761
[2] Pending Comments Update
https://news.ycombinator.com/item?id=7484304
[3] Pending Comments
https://news.ycombinator.com/pending
I am sick and tired of seeing the scoring system injected into the comment thread:
"let the downvotes begin" "EDIT: you can downvote all you want, but ..." "EDIT: not sure why I'm being downvoted, but ..."
... and so on. Just comment and respond, don't add all manner of meta-commentary on voting and the score system into the discussion.
Thanks.
Ideally, it would be a community thing and not a moderator thing. The spirit would be, let's all remind each other of the guidelines when we need to. The things that HN cares about, like civility and signal/noise ratio, don't come naturally to an internet forum. We all have to work at it (definitely including me).
Features like that are hard to get right, though, and until there's a good idea of how to do it that fits naturally into the existing site, it will remain prospective. Anyone with ideas here is welcome to send them to us at hn@ycombinator.com.