If you do the math, it's actually quite obvious that if your popularity contest for ideas inherently, by its structure, favors people who waste their own time, then your contest will produce winners which are actually losers. The most popular ideas will not be the best ideas, since the people who have the best ideas, and the ability to recognize them, also have better things to do and better places to be.
Regarding:
I want my blog banned from Hacker News. I'm not even kidding.
Suggestion: don't leave it up to News.YC to implement this. Set up a mod_rewrite, or even just a JS insert, which checks for a news.ycombinator.com referrer and bounces traffic to an error (or gross-out) page. Voila -- submissions will be rare, and those that are submitted won't be promoted.
Interesting that the very post in question is a pristine example of a flash of genius insight ("What's broken about the social news model and causes every social news site to turn into hogwash eventually") buried in the middle of a tedious rant.
To Giles: If you want to make the world a better place take a look at yourself and make a change! The value of this post could have been communicated in about 3 paragraphs, with no dubious pictures and ranting nonsense.
I actually liked the rambling, ranting quality of the post. "People are asking themselves how to save Hacker News, and they need to realize there is only one technology which can save Hacker News, and that is the time machine" <- priceless :)
I probably wouldn't have read the post if it were shorter and dryer, though it looks like it had the opposite effect on many.
Aside from the pseudo-zed-shaw rant, I think he had a very interesting and enlightening point about the karma-as-video-game problem, and the general less-productive-people-have-more-time-to-influence problem.
I would like to think we solve this here by being too boring for the "Hordes Of Moronic Imbeciles Expressing Their Opinions About You" (his term for it). However, being a jerk to people here and then being smug about it makes it hard not to consider one part of that group.
We need a word for "being a jerk to people here and then being smug about it" online. I think it's basically the same kind of problem as trolling. It doesn't help the signal to noise ratio of a site by generating another bit of rude noise, no matter how emotionally satisfying you think it is at the time.
He's right about delicious. After an hour browsing, I'm rather in awe of that site. It has a very good signal/noise ratio. Almost as high as dzone & much more readable.
What are we measuring, exactly? It's hard to talk about without defining what we're measuring.
A stab at defining some quasi-objective metrics, or, How I think about social news quality:
Are all the good stories getting submitted? (Sometimes the real question is: Are there any good stories on a given day, at all? HN usually nails the stuff I would choose to read in my blog reader. There is good volume here.)
How many bad stories are submitted? Does it matter?
Are the good stories being voted up? (I'd say 'usually' for HN. Though some gems are missed by the upvoters. Bowkett had his example. There was an awesome PDF earlier today that wasn't voted up. I blame a shoddy headline there.)
Are the bad stories being voted up? (I'd say sometimes for HN. There was that bad bout right after TechCrunch. And lots of Ask YC posts that bore me make it to the front page.)
How good is the discussion? (HN is the best I've seen for a site of this wide scope. People are harsh on lazy English, good at ignoring trolls, good at pointing out logical fallacies politely.)
How easy is it to choose good stuff to read? (On the front page, points and # of comments are great indicators. Lots of entries means there's plenty to scan from.)
One consideration is that I come back to HN at different times for different purposes. Sometimes I'm bored and looking for some short fluff to tide over a compile. Sometimes I'm bored but energetic and trawling for something interesting to learn. Sometimes I'm in the mood to check out what other people think about some issue, sometimes I'm not. Sometimes I want to bust 3 minutes on a quick tidbit, sometimes I want to bust 30 minutes on a paper or good Gladwell article. I can't judge HN for my mood swings, only for how easy it makes it for me to satisfy the varied cravings.
All in all, I like HN just fine, and I encourage Giles Bowkett and people like jrockway below to vote with their back buttons.
> People are harsh on lazy English, good at ignoring trolls
Wouldn't it be preferable to be good at ignoring lazy English? If the lazy English comes with trolling, it's ignored completely; if the lazy English comes with sincerity, everything's handled politely.
I think he has an interesting point about why these types of sites don't work very well. But I think he's wrong to say that there isn't any technological solution - you just have to do away with the idea that everyone is equal.
How about this as a solution:
* Everyone can write comments.
* Only moderators can vote on comments
* Only moderators can submit articles
* The number of moderators is fixed (at say... 1000?)
* If a moderator hasn't moderated in the last 3 months, they are removed from the moderator list, opening up a vacancy.
* The person that replaces the removed moderator is the commenter with the highest average point score for their comments (ie total points/number of posts), provided they have posted in the last month. It may be necessary to add a minimum total points score as well (say 100 points, and then it's the person with at least 100 points, and the highest average that becomes moderator).
This type of system should ensure that only things that interest people that themselves have been found to have interesting things to say will get voted up. There will still be trolls, but they won't be voted up.
Moderators? Points? This is a proposal completely ignoring what Giles is talking about. How would moderators be much different than what we have now? The people that spend the most time here would become the moderators, and we would lose a lot of the votes that "count".
Well, geeez, I don't know, maybe because you only become a moderator after having demonstrated your ability to generate original and interesting thoughts!
Apparently you struggle with basic maths, so let me help you out: my whole point was that you should reward interesting commenters, not prolific commenters. Indeed, considering it's actually quite difficult to be both prolific and consistently interesting, being less prolific should assist in increasing your average.
But hey, I understand why your nose has been put out of joint. Based on your post here, you wouldn't make the cut for moderator...
> my whole point was that you should reward interesting commenters, not prolific commenters.
In theory, great, but Giles' whole point was that in practice, this just wouldn't work. In the end, you're not rewarding interesting commenters, you're rewarding the prolific ones, because the prolific ones are the only ones who have enough time to spend on a site and be moderators. If anything, moderators should moderate content, NOT create it.
> But hey, I understand why your nose has been put out of joint. Based on your post here, you wouldn't make the cut for moderator...
And comments like this (and the first quotation) would make you eligible?
Really, though, it seems that a lot of people that could provide great content, but sparingly, wouldn't make the cut, either.
You seem to be under the impression that I agree with Giles. Let me refresh your memory of how I opened this little discussion:
"I think he has an interesting point about why these types of sites don't work very well. But I think he's wrong to say that there isn't any technological solution - you just have to do away with the idea that everyone is equal."
You see? I think he's wrong! But he is correct to point out that current systems favour prolific posters, not quality posters. I used that idea as a jumping off point to perhaps find a solution, rather than doing a Giles, throwing my hands up and saying it's all too hard, and can't be done.
You also seem to feel that by definition anyone that spends time on these sorts of discussion is exactly the kind of person that you don't want to see comments from. Bizarre. I mean, take PG for example. He has effectively adopted the strategy that I am putting forward - he takes a lot of time crafting his essays, but he doesn't make a lot of them. We all find these posts interesting enough that we hang out on his site, and in particular on HN.
My strategy was to try and capture that idea of quality over quantity. As far as I am aware, no one has attempted to moderate a site this way yet (although maybe I have just not visited the right site yet).
Anyway, do you have anything more constructive to say other than 'but, but, you disagree with Giles!!!!!'? Because, as I've already said, I started this converstaion by stating that I thought Giles is wrong - that I disagree with him is a given.
> You seem to be under the impression that I agree with Giles.
No, I merely stated that your "solution" took every "problem" that Giles observed, and magnified it: making all of those problems bigger.
> But he is correct to point out that current systems favour prolific posters, not quality posters. I used that idea as a jumping off point to perhaps find a solution...
Right, that was one of the problems he mentioned, and it's one of the problems I pointed out getting worse in your "solution." Your moderators would be those prolific posters: people who had time to be on the site regularly. The casual votes or submissions would be lost, and the system would continue to favor prolific posters. (OK, now I've brought this point up several times to you, you haven't addressed it, and you continue to act in a hostile manner.)
> You also seem to feel that...
I don't, but would like to note this fallacy of yours.
> I mean, take PG for example...
I read what he writes, and I'd recommend his recent essay, "How to disagree."
> As far as I am aware, no one has attempted to moderate a site this way yet (although maybe I have just not visited the right site yet).
Wikipedia is an example of this kind of moderated site with contributed content.
> Anyway, do you have anything more constructive to say other than 'but, but, you disagree with Giles!!!!!'?
Yes, I've mentioned it in every post, and you haven't addressed it, so I'll mention it twice in this one:
> But he is correct to point out that current systems favour prolific posters, not quality posters.
OK, so you _DO_ agree with him here.
> I used that idea as a jumping off point to perhaps find a solution...
And your "solution" goes completely against this. Moderators would not be, by definition, "quality" posters, but rather "prolific" posters. And if only moderators could submit content, we'd lose any casual "quality" submissions.
"Right, that was one of the problems he mentioned, and it's one of the problems I pointed out getting worse in your "solution." Your moderators would be those prolific posters: people who had time to be on the site regularly. The casual votes or submissions would be lost, and the system would continue to favor prolific posters."
You see, right there. You seem to think I was being nasty by suggesting your maths was subpar. And yet... Here, let me give you a nice simple, concrete example:
Poster A only posts once a month. A casual poster by most definitions. But his posts are always interesting and relevant. He averages 30 points for each of these posts, but in a three month period only gains 90 points. However, as he has only made 3 posts, he finishes with an average of 30.
Poster B is a prolific poster (by the way, maybe it's your understanding of the word 'prolific' which is the problem here - it means 'producing in large quantities'). She is on the site every day, posts messages at least 5 times a day, and gets about 2 points for each message. Over the three month period, she makes about 900 points. BUT, her average, due to the fact that she has made about 450 posts, is only 2.
So, under the system I propose, is it the prolific poster, or the quality poster that gets moderater status? Well, seeing as my system works on averages, it's obviously the quality poster. Which of course shows that when you baldly asserted :'Moderators would not be, by definition, "quality" posters, but rather "prolific" posters', you were just plain wrong.
I dont know, maybe your maths doesn't suck, but if that's the case, you have a serious problem with reading comprehension. Either way, you're just plain wrong in what you assert. Sorry if you don't like hearing that.
'And if only moderators could submit content, we'd lose any casual "quality" submissions.'
Ah hah! There! I knew you could do it! A valid point! This is indeed a downside to the system I proposed. The idea was to stop the forum being constantly flooded with techcrunch articles etc. But I'm willing to concede that this isn't perhaps critical. You probably could trash it, because in theory, the high quality moderators would vote down boring crap, and vote up interesting crap, providing an effective filter against rubbish submissions.
> You seem to think I was being nasty by suggesting your maths was subpar.
> maybe it's your understanding of the word 'prolific'
> maybe your maths doesn't suck, but if that's the case, you have a serious problem with reading comprehension
> you're just plain wrong
> Sorry if you don't like hearing that
> Ah hah! There! I knew you could do it! A valid point!
What's with the rampant condescension?
As far as that last "valid point", it's one that I stated awhile ago, and you've only addressed it now.
Look, I'm always willing to have an intelligent argument, but I'll defer a long discussion to when you're willing to make points without blatant fallacies. I'll offer this criticism:
Your latest example no longer talks about moderators. Ignoring this, if a poster averages 30 points per "post", this still falls back to points, and the idea of "popularity contests" and "vote when you agree" mentalities, which are _points_ Giles brought up in his post.
Sigh. Here's some helpful advice: when you're in a hole, stop digging!
> As far as that last "valid point", it's one that I stated awhile ago, and you've only addressed it now.
You did? I've just been back over your posts in this thread, and nowhere else do you mention moderators making (or not) submissions. But hey, I'll just add 'lacking honesty' to 'sucks at maths' and/or 'poor reading comprehension'.
> Your latest example no longer talks about moderators.
Oh? how about this: "So, under the system I propose, is it the prolific poster, or the quality poster that gets moderater status? Well, seeing as my system works on averages, it's obviously the quality poster."
This is the sort of thing that just makes you look silly. As I have previously said, either you have poor reading comprehension, or you're a liar - the evidence is there for all to see...
> Ignoring this, if a poster averages 30 points per "post", this still falls back to points, and the idea of "popularity contests" and "vote when you agree" mentalities, which are _points_ Giles brought up in his post.
Oh, here we go again; "but, but, you're disagreeing with Giles!". Yes, and what's your point exactly? I've already said that I disagree with Giles (twice before this post, but, hey, I'm happy to repeat myself if that'll help you). Apparently you're not terribly good at picking up on things that aren't explicitly stated (an even then!), so I'll explain.
Giles suggests that popularity contests (points systems) suck, because they are won by people that waste their time on blogs/news sites. I propose a system that keep points, but hopefully rewards quality rather than quantity of posting. Do you see what I have done there? I have attacked one of the assumptions of Giles' "Popularity contests suck" position - namely that popularity contests are always won by time-wasters, who generally speaking are not interesting people.
If you want to argue against my position, you need to demonstrate why my system won't correctly give voting rights to interesting people, or show why people that write interesting posts are not necessarily the best people to be deciding which other posts are interesting. Or maybe suggest a superior system that doesn't use points in any way.
What you can't do is simply repeat back to me Giles' position. I know what Giles' position is, he's been kind enough to write a big long entry on his blog to inform us all. Your repeated restating of his position, without adding anything new to the discussion, and completely failing to address any of the concerns I have raised is nothing other than good ol' fashioned trolling.
Look, what can I say. I suggest that you go back and read PG's essay on disagreeing again. Please note that he places a high value on actually addressing ideas presented in a post. Maybe you should try doing that rather than whinging about others being what you perceive to be condescending... When you start doing that, you'll be amazed by how quickly the condescending reponses disappear.
Even if the stories have stayed the same, comments like this make me think news.yc is declining. "Apparently you struggle with basic maths" - this kind of condescension is ridiculous and not what I come here for.
meh, perhaps you have a better way of describing someone incapable of figuring out that the way to get a high average is to post a low volume of high quality stuff, not to spam the list. Personally, I think that 'struggling with basic maths' is a succinct way of describing this.
With Slashdot, anybody can vote. Often, trollish behaviour gets rewarded with points, because, well, because trolls can vote too! Before being able to vote in the system I proposed, you have to have demonstrated to the current moderators (voters), that you are someone that on average posts interesting, thought-provoking stuff.
I don't know whether this would actually work, but it seems at least reasonable, and as far as I'm aware, it has not yet been tried
Systematic factors inherent to social news sites degrade their quality over time. Voting is broken because of the nature of the people who vote the most and thus have the most influence on the front page: They are people with too much time on their hands. That is, anyone with a worthwhile opinion is off doing other things more important than commenting or voting on a news site.
This isn't to say I agree, I'm just trying to summarize the relevant bits as best as I can.
Simplifying the key observation: free votes put more power in the hands of users who vote more. Users who vote more are, in his estimation, users who waste their own time. So social news sites based on free voting tend to promote articles that are a waste of time.
Restricting the content seems to solve that problem. As long as you don't allow the kind of articles these people like, they won't frequent your site.
I haven't noticed significant degradation in quality since we started News.YC. In fact, it's gotten better, because when we started it was called "Startup News" and there were a lot of crappy gung-ho articles about "entrepreneurship."
The other aspect is that he thinks there is only 1 type of source that's ok for information consumption, which is the type that spews out brilliance. All the other sources are worthless, including the people that never say anything meaningful and the people that say something brilliant once in a lifetime.
I don't think that's what he was saying. He was just complaining that he needs better filtering mechanisms, so he can find the 1 in 100 posts by e.g. Cory Doctorow that he would enjoy, without having to read the other 99.
allow me to summarize: giles bowkett is impolite on social sites such as this one, therefore they're all doomed. also, there are people on the internet that cause him trouble, like tim bray and cory doctorow.
whew, such a heady brew of straight-up TRVTH that guy is dishing out. i'm still reeling from the force of it.
These are really nice measures, but I think they still do not address the most important problem - people who have more time on their hands have disproportionate effect on the site.
I have some internal metrics to warn me if the top stories get bad, and there has been no degradation I can detect. Nor do I personally notice a difference.
There were a bunch of crappy submissions after we got that big influx of users from TechCrunch (both I and my software agreed about that), but the quality is now back up to where it was before.
The HN system works better than any other I have seen (it's kept me in line whenever I was tempted into adding noise rather than value).
I don't think degradation of HN over time is unavoidable, clearly a lot of smart people here put energy into resisting this force of nature.
One challenge is how to preserve the HN code of conduct. I've noticed some misunderstand the voting system as an agree/disagree button. I think the software should notice patterns that don't look right, such as both up- and downvotes on the same comment (popularity contest alert).
> 2. Because that kind of stuff is banned, the average digg/reddit reader, if he comes across News.YC, finds the content boring and leaves.
Yesterday we had both a "global warming" story, and a "you can't soak the rich" story. Those kinds of things are far more interesting and accessible to random non hackers, than say, this:
I voted up the Can't Soak the Rich story. It's a very interesting idea that tax revenues are a roughly constant fraction of the GDP. Not all stories about politically charged topics are merely political stories.
> Not all stories about politically charged topics are merely political stories.
I think between sources like the FT, The Economist, and various economics blogs, I run into a lot of stories that are interesting to me in that same way, that I would never consider submitting here, because they are pretty much guaranteed to degrade into the same old hashed and rehashed debates.
> 2. Because that kind of stuff is banned, the average digg/reddit reader, if he comes across News.YC, finds the content boring and leaves
I really think keeping it boring is key: not only keeping the stories boring to your average teenager, but keeping the site looking boring, like a book. :)
I think you missed the point, too. There are no mistakes for Hacker News to avoid -- the devolution of the site is inevitable. Sooner or later the trolls will take over, and we'll all move to some new site, who will promise to be better than Hacker News. This has all happened before, and it will all happen again.
oh, i read the article, all right. and i disagree with his primary point. i've been reading hacker news for over a year now, quality has remained high, trolls have gained no traction.
My startup is not making this mistake, but does anyone care? People seem to prefer trolls and fun social noise to high signal streams that demand your attention.
The majority of society demands to be entertained, not to entertain. Those that entertain are put on pedestals to be worshipped (either with praise or scorn).
So because Giles doesn't want to see his posts mentioned on HN, no one else should see them either, despite the chance that some of them might be worth reading?
Ingenious!
Oh, and he's a jerk on HN, but it's alright because it's intentional.
I'm not sure that deciding to behave like a jerk is any more noble than happening to..
If you do the math, it's actually quite obvious that if your popularity contest for ideas inherently, by its structure, favors people who waste their own time, then your contest will produce winners which are actually losers. The most popular ideas will not be the best ideas, since the people who have the best ideas, and the ability to recognize them, also have better things to do and better places to be.
Regarding:
I want my blog banned from Hacker News. I'm not even kidding.
Suggestion: don't leave it up to News.YC to implement this. Set up a mod_rewrite, or even just a JS insert, which checks for a news.ycombinator.com referrer and bounces traffic to an error (or gross-out) page. Voila -- submissions will be rare, and those that are submitted won't be promoted.