Ask YC: Do you care about Karma? | Hacker News Reader

Ask YC: Do you care about Karma?

11 pointskashif62 comments

Since the last influx of users, I have been thinking about social aggregation systems such as news.YC. And, to put it in a nutshell, I believe that for communities such as nYC which are, not mainstream, comparatively close knit and somewhat homogeneous, it is possible to improve the quality of the submissions and comments if we removed individual karma altogether but kept the voting system for submissions and comments intact.

What do you think?

View on HN

I care. Reddit was destroyed by trolls and joke accounts. They get downmodded quickly, but they keep coming back and spamming every article. Even though comments are the main point of Reddit, you get no credit for posting good ones and no penalty for posting spam.

Slashdot had the right idea when you were limited to one post a day (or whatever) when your karma got to -10. If you can't write content that other people enjoy, you need to leave or adjust your attitude. Karma is a way of making people feel bad enough to go away :)

The more people we can encourage to stay on /b/, the better... 'cause I really don't care to read that drivel, but I do care to read what people think about programming-related things.

I am sorry, I don't understand your point on how Karma relates to spammers going away and 'joke' accounts not being created, especially if they spammers kept coming back anyway.

My assumption is that reddit only really cares about keeping traffic up, while PG only cares about keeping quality up.

Having karma today allows PG to make changes tomorrow if he needed to take care of quality issues.

Okay, thanks. Much clearer :)

Did you just say /b/ instead of Hacker News? Those are like as extreme as you can get on the two ends of the spectrum.

I meant that we should convince the trolls that /b/ is where they should post, so they don't clog reddit and hacker news with their drivel.

yes but it's too easy to get bad karma on slashdot. I have bad karma on there (1-2 bad comments) and now none see my posts... it's too hard to redeem.

plus their karma system seems less easy to use.

I had a karma of ~350 until the software glitch at the beginning of the month. At first, I cared, but then I realized that it's not important in the grand scheme of things. However, I would like to get my #CBDDFF top bar back, and have the down arrows available again. Otherwise, it's no big deal :~).

Yeah, same reasoning with me. All I wanted was to change the top bar color. Then I started getting close to the leader board and I suddenly need to grind points:D Vote me down to save my soul from the karma!

- EDIT -

I want to add that my most recent submission that did get me a lot of karma was not posted for the sake of karma. I genuinely am curious and I appreciate all the responses.

down arrows are the plus.

Yes, I care, as a form of trust metric:

For the person: it is a simplest way to see if you fit or not in this community. For the community: quickly judge if the post or comment is spam or not.

I totally disagree. The best way to judge if a post is "spam" and if someone is a "spammer" is based on the content. I downvoted your post because it both is a poor form of reasoning and it is needlessly alienating to others.

I don't downvote often, and I don't have anything against you, but it is important that everyone, especially the lurkers, feel welcome here. My downvote is to show you don't speak for all of the community, since this response and my karma means your comment is self refuting.

keyword is "quickly".

Content based filtering is extremely expensive, just look at email. Reputation based filtering is quick and easy both for people and computers, but it has its own drawbacks, I agree.

By community I thought you meant the normal readers, not the moderators. For normal readers, it is easier to read a post.

How is it possible to know whether you fit into a community by looking at your Karma, especially if you are a newbie. If you aren't a newbie then you are already participating or observing the community anyway.

To illustrate another subtle point. Lets, take your karma and compare it with mine and compare mine with someone like nickb. Now what conclusions can I draw about whether I fit into this community.

Easy, you just have to take time since you've joined and # of posts and comments into account:

I can see that you joined a year ago but didn't gain much karma. That means that you're an old timer, but a lurker, who doesn't post or comment often.

nickb and other top 10 karma users define this community and they lead by example.

I joined just recently, because I somebody posted a story about my startup and I had to comment. But since that I've been posting stuff that people care about.

All this is visible by looking at the profile of the user, without knowing exact posting history. This is huge time saver for everybody involved.

So you suggest that all people who do not post often enough do not fit into the community. And that posting often is the metric to drive? And because, in our current example, you have a higher karma per day, you belong to this community as opposed to me? But if nickb was to read this comment then you and I might both be trivial?

Finally, as a supposed 'lurker', it might serve me better not to login and just read submissions instead of participating by voting. Clearly, voting and commenting are also measures of 'being in a community'. So as lurkers stop logging in they cannot vote and then voting drops making the submission rankings either volatile or static.

I do not understand your line of reasoning.

I agree, lack of karma shouldn't say anything negative about your inclusion in the community here. The info is free, so it's not like only major contributors belong.

In fact, if it were the case that only the high karma belong, there'd be nothing to contribute to and no high karma. People like kashif and the thousands of other lurkers are why this community exists in the first place, much more so than nickb or any of the other leaders.

Speaking for myself, I have a high karma because I'm a slacker. I consider it an inverse metric of my worth.

Simple: karma as a measure of your contributions to the community is not perfect, but it is much better than nothing at all. Do you know a better system?

That is not a response to the original chain of thought. Nonetheless, karma might be an unnecessary tool for the necessary effect.

Do you know a better system?

Yes, I am suggesting an alternate mechanism which uses only voting on submissions and comments while discarding individual karma.

The principal effect of karma here is not how it helps you judge your fit, but how it actively shapes the community/content. This effect was clearly visible on reddit as it was realized that the more sensational content was vastly more promoted in the reddit karma system.

So my question:

What is the value of high quality content ordering? Are the present ordering strategies on HN/reddit/[insert karma-based discovery site] good enough? Would the theoretically perfect ordering offer much more value than current systems?

Take a look at my startup, we're trying to push a different system for ranking stuff. I personally think there's a lot more value in it. But sometimes people prefer simple algorithms that are easier to understand, there's value in it too.

I like how you automatically assign a guest ID so visitors can try out the customization features right away, and think, "Hell, I'm halfway to joining -- might as well go the rest of the way now." The same thing could have been done invisibly with a cookie, but then the psychological hook wouldn't be there.

That's a very clever detail.

Hmmm..A better ordering system might improve things but I am not aware of any simple alternatives.

We're only talking about ditching karma, the running total for each user. We would still be able to upvote/downvote stories and comments, but those votes would only effect the individual story or comment, not the user who posted it.

No I don't. I just tell everyone how I feel and when I need to top up on Karma I just post a pg related article...>:)

he said pg everyone +1 for him.

You did too +1 for you!

awe come on... it's a joke.

It's not funny when it's true:)

no. that's why it IS funny, 'cause it's true.

It's sad too, though.

If pg says anything he gets at least 50 points for being pg. The rest is because pg is actually smart and doesn't usually say dumb unthoughtful things.

pg's median comment score is 4:

http://news.ycombinator.com/item?id=166110

More people read pg's comments, so he does get more upvotes than the same words would if coming from someone else. However, to say that you can mention pg in a comment to get more votes is neither true nor particularly funny.

he mentioned posting an article related to PG, not just comments

wouldn't mode or average be more useful?

For most questions I think average would be more useful. For the question of whether people upvote pg thoughtlessly or not, median is useful. If more than half of pg's comments end up with a score of 4 or less, then we can conclude that his comments that end up with 50 or more points are a select subset.

Submission scores might be a statistically different distribution then comment scores. Also while median is better than mean, it may or may not be better than mode - difficult to tell without looking at the shape of the distribution.

I think he would get more if our usernames stood out more. but I can accept that I am being grossly inaccurate and exaggerating.

Median score is not what's being discussed. It's quality vs score. Those individual comments that are of completely average quality and yet have phenomenal scores are the most telling.

Also very telling: score velocity over time. For example, if `joe` makes a good point with factual grounding and is getting lots of points. Then `bob` replies in disagreement and bob provides almost nothing to back up his position. Then suddenly the points reverse (joe goes negative and bob gets lots of points) just because bob said it it must be true and everyone blindly agreed. There are ways to measure this impulse response that would provide more concrete numerical measurements.

Median score... there are so many effects that would cause this to not relate to comment quality that that is pretty useless. For example, just commenting on more active topics can give you a much better median score.

I don't care about Karma.

Mostly.

One day I noticed that I had about 180 karma and decided to try for 250 so I could change the colour of the Navbar. After clearing 250 (and changing the color to a nice soothing green vs the default orange), I find I don't really care about Karma.

A soldier will fight long and hard for a bit of colored ribbon. - Napoleon

Perhaps, these customization freebies only become available at various karma levels because the intent is to get a user to increase karma.If we aren't chasing karma then this stuff can just become available as features.

Wait till you get close to the leader board. Then there will be a challenge again. It's insidious.

I think that people spend a LOT of time obsessing over karma and rankings/ratings.

This includes not only obsession with gaining karma, but obsession with how to better measure, display or distribute/earn karma points.

What I would personally prefer to see is someone present a valid, (semi)tested proposal for a better system. There have been a lot of threads withe what I personally consider to be vague half-suggestions for "improvement" that don't really articulate how the proposal will make news:YC cumulatively "better".

When I'm reading the site or submitting comments or linking I don't really find myself giving a lot of merit to the overall rankings. I'm just as likely to add a comment to a story with a ranking of 100 as I am to a story ranked 1, provided the story itself interests me.

Perhaps your argument would be more persuasive if you applied it to some submissions or comments showing how it would affect those items for the overall betterment of the site.

Alas, I do not have the necessary resources.

User-wise, I'm more interested in a user's average karma per post than their total karma.

Post-wise, I have some gripes with the way karma works, in that your high-value comment can easily go ignored because it starts off at the bottom of a well-populated thread. Which, for karma-gamers, means it makes more sense to reply to a comment with high karma to increase the visibility of your comment rather than starting a new comment thread. And it also means that your average karma per post doesn't necessarily reflect your average value per post.

I'd love if social news were more like an RPG where you could "level up" to get more features, customizations and other things I can't think of. Last I knew, users with 250 karma were allowed to change the color of the nav bar at the top.

I would personally play an RPG instead - if I was looking for such excitement. From nYC I want the best submissions. :)

But, the best submissions would still exist. People would still vote for the best things. There would just be more rewards for those who submitted them.

Yes, that is indeed the current hypothesis. I am suggesting an alternate one might improve the quality of the submissions even more. And, to test this new hypothesis we must necessarily discard individual karma.

then we could have karma farming bots :P

And a possibility of acquisitions by gaming companies...

Interesting.

I'm only happy when my karma reaches a certain significant number. Like 314 or 411.

No, I don't like the karma system. It has too much potential to discourage dissent. I like the upvote/downvote system for comments and stories, but keeping track of karma turns it into a punishment/reward system.

No, I think the added impetus to ad value to a conversation that karma generates for some is very helpful in keeping the signal-to-noise ratio at a decent level. Karma certainly doesn't HURT in regards to this and it has the ancillary benefit of quantifying, to some extent the reputation of a given contributor, which, for new users is very useful when it comes to understanding who to listen to.

Okay, but I am not sure how credible such a quantification is. Perhaps an extremely 'credible' contributor posts quality stuff but very rarely. How likely is it that I miss his submission if I was to look at karma as a measure of credibility. My point is that karma equates more to participation than substance - it is possible to have both.

I like it. But I think the amount of 'karma' that it takes to get on the feed should be raised to 5. There is way too much feed traffic.

I would be lying if I said I didn't.

But I care more for a good conversation.

http://vortices.appjet.net has a direct vote up/down on users as opposed to cumulative karma.

All I care about is reading great stuff to build great products.

I care some, but not as much as I once did.

human beings will always care about a number next to their name.

Earning karma points is not important to me. I could honestly care less.

However, that being said, I am more a "consumer" of this site than a "poster." I am not trying to be heard, trusted, or accepted by anyone here.

I think karma is a one-dimensional useless metric, and the only reason it seems to have some value is due to the initial homogeneity of a social news site's user community.

I didn't know hacker news added silly features based on karma. How pathetic.

Reply on news.ycombinator.com