"The worst thing to post or upvote is something that's intensely but shallowly interesting."
Frankly, I think this happens way too much. Any post about Zuckerberg, Assange, Arrington, and others probably gets way more upvotes than necessary. While they have definitely created interesting things, the product is probably more important than the person.
For example, the whole Wikileaks/Assange situation has many aspects that would be interesting to hackers: IT security failures, a paradigm shift in how effectively governments are able to keep things secret, calls for Internet censorship, small groups of people/hackers/script-kiddies DDoSing major credit cards in an example of asymmetric warfare, transparency vs secrecy, and (to some extent, but less so) the contents of the cables themselves. However, because these are being revealed over a course of weeks, HN instead gets flooded with dozens of articles about small facets of this event, rather than a larger and more in depth analysis.
Does this make sense to anyone else, or am I completely off target here?
Don't get me wrong, some of these articles are very good, but way too many people vote up celebrity instead of content.
Don't get me wrong, some of these comments are very good, but way too many people vote up celebrity instead of content.
I could care less about brand, whether from commenters or famous people in titles. Just give me good content. I don't come here to join some kind of fan club, either for authors or commenters.
And of course the reverse is true. We have folks purposely downvoting based on authors or commenters. Which is just as whacked.
Thanks for the explanation.
Yep, I'd much rather read a dozen guys that nobody knows that help me out in life and help me make terrific changes than become pen-pals and an expert on the top 17 people in the tech industry or on HN. This herd mentality stuff drives me nuts at times.
You know, I also feel that people should own their comments, but something as simple as hiding the name until the mouse hovers over the pseudonym would prevent a lot of abuse. It's not like we all have to become anonymous. I think the board just needs to take steps to prevent automatic up or down voting. There are lots of solutions to choose from.
The celebrity writers always have a certain personality. Whether it's John Gruber or Paul Graham or Dave Winer, they've established exactly what sorts of things they're gonna talk about and how they're gonna talk about it. When we see their names, we know what to expect. Whether it's a good post or a shit post (and everybody has some of each), we still feel like we got our click's worth and are satisfied with that.
If I had a choice to see a new DaringFireball essay or another essay, completely unknown, completely at random, I'll probably go with the DF one. It sells me a certain level of quality, a topic that I usually am interested in, a writing style that I very much enjoy. The unknown essay might be more valuable or relevant, but I'm not as instantly willing to give it a chance.
...unless, of course, the author makes an effort to grab me, by titling his essay provocatively or writing in an entertaining style. But then once he finds success with that method, he's not going to want to drop the shtick and risk becoming unknown again. He'll continue with it. At a certain point he'll become a celebrity poster. Then we're back to dealing with the same old shit as we did before.
Part of the problem is that HN displays us both POINT VALUE and USER NAME before it even gives us the essay. There is no way to read a comment without seeing the author and the point value before hand. So we're unavoidably biased before we see a word.
The site also needs to do away with the leaderboard once and for all, and, in a perfect world, make it hard to find out how much karma another user has.
Seeing my own karma has absolutely made my writing suit HN better (I'm a lot meaner elsewhere). It's a valuable signal. I wouldn't wine if I got docked, or if karma went log-scored or something like that, because it's absolutely true that high-karma users also get an automatic point boost, but I'd rather not lose the signal.
I don't scan the comments as much as others probably, but when I do, I prioritize by a few things: Those that include links, and when I have the time, those that are lengthy, since often those can be from people who had a lot to say and you can tell by the tone of the first paragraph if it's worth your time.
I love the egalitarian nature of the system that doesn't single out new users who might just as well have valid points to make.
They care about learning something new, being intellectually stimulated and sometimes just being entertained, but given the limited way that karma is exposed, I'd be surprised if the average user goes to any great length to find out the karma count of a comment poster.
I may have to make it, now that I think about it. It'd probably change my voting habits a bit. Especially if I also removed the usernames until after I voted... flatten the playing field, so to speak, while still identifying like-minded people.
See, when Joel writes a small piece of advice for B2B sales ( http://news.ycombinator.com/item?id=1987223 ), knowing the background he's coming from, that advice get a disproportionately higher weight both in my mind, and my upvoting-finger, than Random J Commenter.
And Yes, I am implying here, that Random J Commenter doesn't know shit about B2B sales. With his advice, the most action I can take is careful, small-scale controlled experiments, 2 times out of 3 blowing right to my head. Because Random J Commenter's infosource is not the real world, but rather techcrunch, or, more likely, other Random J Blogger.
The other aspect of this, is user-based browsing, which I tend to do a lot (following 10+change HNers, and looking where they comment -as opposed to the stories on frontpage / askHN). This also gives them point leverage -after all, this way I see only the things they care to comment about.
The single most important observation I take from the above two, is that neither of these are the kind of shallow things PG is remarking in his articles -both are meritocratic worth of value provided by these "celebrities".
It just happens, that business knowledge is distributed disproportionately.
Additionally, in my opinion, you are going to find more success in following tangential ideas from someone successful. You aren't going to duplicate what pg or Joel or any of those guys have done. The real value comes from intelligent people discussing tangents to some idea or theory. And if popularity is disproportionally rewarded to the thought behind the discussion, then the result will be less contribution to the site.
I think the problem is that the way they think and post is more in line with the majority of users on here. There isn't necessarily anything wrong with that. I think the problem is that we (like any other group of people) tend to focus more on "I agree with that comment" and less on "That comment is insightful and brings a fresh perspective on the issue whether I agree with it or not". I wish I could say that I'm innocent of this, but I'm just as guilty of supporting my own point of view as anyone else is.
While true, the much bigger problem is not that these posts get too many upvotes, but rather that they're the only posts that reliably get any upvotes at all. It's hard to justify even bothering to submit stories since for even the best posts it's a complete crapshoot as to whether or not they get any votes.
In practice, the people that designed the upvote-downvote system and presented it to us as Web 2.0's way of making the Internet perfectidealolympus forgot that they were dealing with human beings. They assumed they were dealing with computers or something for which there's a such thing as "optimal". As a human being, sometimes I feel like a selfless contributor of optimizing upvotes and downvotes, other times I feel like upvoting because hey I like this guy, other times I feel like downvoting because fuck the world and fuck you buddy, and still other times my Dadaist urges kick in and I vote entirely randomly to sow discord and paranoia among the otherwise content bourgeoise. Or at least that's how I imagine some people vote.
The thing about dealing with vague information like "links" and "comments" is that there is no agreed consensus on what's optimal. Back when I first joined Hacker News two or three years ago, there was a faction of users (myself included for a time) who disagreed with snark replies and witty one-liners. That's almost all gone now; we accept that the community feels the need sometimes to be clever rather than overwhelmingly in-depth. There have been people who complain about HN's start-up news/technical news/YCombinator-related news/breaking news/programming news/obituaries/spammy motivational blog posts that tell you you should be happier than you are/Paul Graham's thoughts on his local deli's chicken salad. None of them are "right" per se; they all simply have a different vision of what they want. If YCombinator wanted to evolve Hacker News, they could attempt to install solutions to this perception problem, and come up with clever alternatives that would help each of us see only the things we want to see. But I'm guessing they're happy with HN as it is, warts and all; I'm also betting that most of us are accepting of HN as it is, or at least we suspect any attempts to fix problems will only make them worse.
So in the end ironically the votes up and down don't matter. Downvoted content is sometimes great. Upvoted content is sometimes really terrible. The lucky thing is that Hacker News is meaningless enough that we won't be shedding tears over its imperfections. Or at least we shouldn't be.
I think that is resignation. Every community needs a culture, forms to pressure others into doing the 'right' things, ways to be able to incorporate others. Once you loosen up and make it free for all this might as well be talking to the lowest denominator.
When I said it's meaningless, I mean relatively rather than in some nihilistic sense. Obviously, things on this site matter. But if we have a front page on which I'm only interested in 5 or so posts at a time because I think the rest are silly, that's not too bad if other people are looking at those other posts. I'm not going to get too pissed off at the sensationalism and the dumb stuff. That's just the way this site is. Trying too hard to change that might ruin all the things that I like.
HN is not particularly Serious Business, you know? It's a fun site with some great commenters, but it's not like MetaFilter where the culture is so unique and good that I'd get upset if it started deteriorating. Or like the old threads I read on Slashdot or Usenet where I realize there was such a good thing once but now it's gone into decline. HN has always had these community guidelines that nobody follows, and its successes have been despite that. If that makes sense.
What can happen in practice: more and more people filtering the posts they comment on using any of the techniques described on http://news.ycombinator.com/item?id=1960762
If documents are clean, you can compute the repulsion from the bag-of-word dot products between documents, but web pages often have a lot of noise in the outer template, so getting good results with web pages means you need some kind of template drop-out.