Ask YC: (< (/ signal noise) too-low)
I've been a member of Hacker News (actually Startup News when I joined) for over a year now. I've learned a lot here and been exposed to a lot of interesting ideas. Recently, however, I've often seen the front page have at least 10% of the stories be either more Reddit-worthy or be submissions discussing how to combat Hacker News deteriorating into just-another-news-site.
With the value I'm getting from Hacker News being much lower than it was when I joined (and up to a couple months ago), I think it's time for me to go. There have been several posts I've seen on the front page where I've wanted to say, "Well, this is definitely it," but I think it's better this way, to just keep it to my own submission and leave.
Best of luck to everyone on their ventures and in your lives in general. If I ever get something going, I'll be sure to mention it here first.
Thank you for some great discussions.
88 comments
[ 3.1 ms ] story [ 103 ms ] threadbtw, your profile says to email you, but the email field in the profile is only visibile to admins. You have to list it again in your about me if you want it to be visible to the public.
See http://news.ycombinator.com/item?id=149052
Certain types of making fun of things make for valid comments. In this case, I was pointing out the striking similarity between what he said, and 1984.
I had already voted all three of the posts up, anyways, because I felt the highly-negative karma was a bit unfair.
And, yeah, I knew that asking why the downmodding would be downmodded. It's ok. It was worth it.
And I expected you had read the comment thread and just didn't remember my name, but it was still funny ;p
I've mentioned this earlier and requested the ability to dismiss items for view (archive), but pg didn't seem to think it was a good idea and/or worthwhile to implement. Every time time I see the same story on the front page 2 days in a row I get this sudden urge to add
127.0.0.1 news.ycombinator.com
to /etc/hosts. Sooner or later I'll go ahead and do it, and not return until HN gets archiving and keyboard shortcuts.
Perhaps a random selection of stories for the front page, with probability of selection weighted by their score? That would diversify the news, and give off-hours posts a fighting chance for views....
Paul, can you check this? The average points of news that reached the frontpage, the points of the max. scored piece of news, etcetera?
In addition, I think that the perceived amount of stay is longer. While the front page as a whole changes faster, there are certain articles that stay for quite a while. It's less noticible if it's an article you liked, but if it was something that you hold contempt for, you'll tend to notice it every time you login.
I'd be happy with a less drastic measure: Why store the previously read info in a cookie? Perhaps it could be stored with my account instead. This way the highlighting of previously-read articles would be consistent regardless of which PC I happened to be using.
The number of articles stored could be limited to a reasonable number (say 3x length of the # of items on the front page).
http://news.ycombinator.com/item?id=149789
Wow. Just wow.
The worst story on the frontpage by that standard right now is
http://news.ycombinator.com/item?id=149752
The third worst, ironically, is the one complaining about the problem:
http://news.ycombinator.com/item?id=149887
Feature requests might work, maybe not. In any case, it would also be necessary to fairly ruthlessly kill navel-gazing articles on the front page, or at least move them to navelgazing.news.yc.com...
Why is that ironic? No solutions are presented. No value is generated. An individual starting a company or hacking on a problem does not grow as a result of having read the story. It is self-referential, self-aggrandizing, and as you point out, among the worst entries in the derby.
It sounds as if vote-weighting has a good deal of merit, based solely on the three pieces of evidence presented.
- If we can get the list of the top 10-50 hostnames (xkcd, etc) posted on reddit, and deny publication of those hostnames on HN, the signal to noise ratio should go up (how much I don't know?)
- Obviously, some good posts would be missed, but it's alright to have a few false positives, since this is a website of aggregated news and not email
What do the folks here think?
In which case, how about NOT awarding karma for posting links for those hostnames, and award karma for the comments. what do you think?
http://news.ycombinator.com/item?id=150070 - highly-debatable claims presented as science
http://news.ycombinator.com/item?id=149708 - decent, but 30 points? Besides, the comments thread looks identical to something I could read on Reddit.
http://news.ycombinator.com/item?id=149467 - who cares?
http://news.ycombinator.com/item?id=149482 - already covered at least a dozen times. I would suggest beginning with a search engine...
http://news.ycombinator.com/item?id=149618 - ancient news
http://news.ycombinator.com/item?id=149196 - old news
http://news.ycombinator.com/item?id=149091 - one of the worst of hundreds of front page submissions regarding the mortgage crisis
I do understand that this is only my viewpoint and that others may have liked at least some of these submissions, but this is coming from someone who was never a member of a link aggregating site. I started reading Reddit towards the end of 2007, but that didn't last long, as I quickly tired of seeing the same headlines and C-C-C-COMBO-BREAKER!! posts, so I left.
Hacker News served its purpose for me for quite a while: giving me the ideas and motivation to create my own startup. I also like some of the interesting science and hacker-related articles. Lately, though, it just hasn't been what I've been looking for.
http://news.ycombinator.com/item?id=149482
All the rest, as far as I can tell by statistical measures, are like the stories that would have made the frontpage in the past. They also seem to me personally about the same.
Do folks not understand how teaching works? How coaching works? How leadership works? How a community works? This group isn't just a dusty library, filled with books that may be out of date and librarians who tell you to be quiet and RTFM.
People like having their questions answered by real humans. They crave contact. They crave personality. They need to know that someone is out there, listening. The voice of an actual human telling you what to do is often more real than your own voice, or the disembodied voice of an old book.
Yes, the questions are almost always the same. That's how teaching works: You're surrounded by novices, and once they stop being novices they leave and are replaced by another group of novices. The secret to being a good teacher is to take enough joy in helping others -- and, less charitably, in the sound of your own voice repeating itself -- to be able to do it again and again with only minor variations.
That said, I have seen similar Ask YC posts that do cover the exact same ground as old ones and annoy me a bit more than that one. I guess I was trying to use that one as an example of such a submission, although it wasn't really a good example.
I've been reading hacker news for months now and I personally haven't noticed a dramatic drop off in the quality of posts. The signal to noise ratio has gone up and there are a lot more "meta" posts recenty - posts to news.yc about news.yc.
Despite these things, I still find quality information on a consistent basis. HN is pretty timely...I'm as likely to read something first here as I am on reddit or digg or any other social news site. The posts tend to be more start up oriented and many of the technical posts are of a better quality than I'd find anywhere else. Despite the noise, there is still a lot of legitimate value on news.yc
I'm not trying to ask the question "well if you don't like hacker news, what site is better?". That would only start an argument based on preference. Instead I'm asking "Isn't some startup focused social news, even with a high signal to noise ratio, better than none at all?". It's possible that at some point, there will be so many worthless posts that sifting through them will be more trouble than it's worth. But at this point, it seems to me that there are still interesting posts and insightful comments everyday on news.yc and it's still a ways away from not being a valuable daily read.
I'm not saying the community isn't good for a lot of people, I'm just saying that it's changed enough that it's no longer what I came here for. I would apologize for making this submission, but it seems that enough people agreed with me that it did make it to the front page.
What I am worried about are these increasingly common AskYC posts reflecting on News.YC. This post, for example, is a somewhat self-centered broadcast of BrandonM's exit. I hope this doesn't become a common practice, because this is the kind of stuff I don't like seeing on the front page.
There's certainly a ton of awesome conversation still going on. But I wish the community would feel like off-white paint again: unnoticeable and non-criticizable. It's what the people in the community say that matters, not the color of the walls.
To keep the baby not the bathwater, what about putting up the submission-level down arrow, but only for Ask YC submissions?
Most of the objectionable AskYC posts could probably be shoe-horned into polls by editors - just with 'agree' or 'disagree' options.
Maybe every post could have additional [agree] [disagree] buttons, because a lot of poor quality submissions are equally opinionated.
If the down-arrows were karma-restricted to ensure adequate assimilation, though...
No offense dude, but this made me think of ESR.
Irony?
Thanks for the personal observations.
I find that I never make such a black-or-white decision: usually I have periods of increased or decreased usage which eventually peters completely out. So it's interesting to see someone who is either participating or not.
Don't let the commenters bug you -- one of the not-so-good traits here is that they sometimes seem to have a bit of a self-admiration society going on. It's not unusual that your post would get criticized. I found it right on-target for what I liked about news.yc -- some person making an observation about the market/program/philosophy-of-business and then asking others for comments. I remember some great AskYC and AskPG questions. Don't seem to see those a lot any more -- at least not as much as before. People who provide honest negative feedback are doing a great service for a community like this. PG owes you thanks for taking the time to rant a bit on the way out.
Perhaps more than anything else, this post was a note to myself to reinforce my decision. I remember hearing advice some time back to tell as many people as possible when you resolve to do something so that you will have that additional motivation of not wanting to seem like a failure. Lately I've been trying to prioritize the things in my life, and since I'm not getting as much value as I used to, it makes sense right now to leave.
I disagree, and here's why. PG is clearly benefiting from this bully pulpit as a means to recruit startup founders. Especially in the midst of Startup School (the periodic anti-September of YC) and a waning interest in risky ventures as the economy contracts, it serves as a barometer. Plus the type of crowd that sailed in from TechCrunch has been revealed (if that is the right word) as generally free of merit. I do not believe that judging news.YC as if it were yet another social news site (YASNS, if I have to use the phrase ever again) is appropriate, and hence the rant is perhaps less valuable than it would be on, say, reddit.
As BrandonM noted, it's still the best source for news that interests this particular subset of the public. However, as another poster noted, perhaps the interesting observations that previously lay unnoticed have been exhausted, and the hard work of turning over new stones, starting new companies, and learning new things must begin.
There's only so much you can learn in the library. Eventually you need to get in the laboratory or out in the field. I suspect that is as much to blame for the change in interest levels as anything else.
(JMHO)
I know that describes me, to some extent. For me, I think the original Startup News was liberating. It was a fresh voice I'd never heard before saying that I don't have to graduate school and go get a job in industry; I can actually do what I want and make a lot of money doing it. After a year of hearing that and getting some helpful tips along the way, perhaps it's just time for me to go do.
Hard to say... but lord knows I hope that you'll have a better idea as a result of perusing the better offerings of the site. I have found it useful to compare my performance against the theoretical optima that I see documented here, rather than just the work of other developers, in order to push myself. Hell, that's why I went back to grad school in the first place (and went into methodology, eventually succumbing to math envy, as it happens).
perhaps it's just time for me to go do.
If you look on the front cover of your copy of SICP, the little yin/yang has 'eval' and 'apply' halves. Another endlessly useful shorthand metaphor, IMHO... It's always tempting to go too far in one direction or the other, but fortune seems to favor those who balance the halves.
I sure hope I don't upset anyone, but here goes:
1. BrandonM, I recognize your name from encountering you many times in interesting threads. Sure hope you'll reconsider. We'd miss you.
2. BrandonM, you've been a member for 369 days and have posted an outside submission 8 times. AFAIC, your complaint carries no weight. If you don't like the content here, submit more better content.
3. If anyone sees a thread they don't like, don't vote up or click on it.
4. If you're tired of what I call "self-examination" threads, don't vote them up or click on them.
5. Personally, I find the threads discussing this site the most interesting. Here's why. Even though most of us are hackers, we're also users. (We'd better be, or we'd lose all perspective.) The one thing we all have in common is that we are users of THIS SITE. We NEED to keep talking about it. It provides valuable feedback to pg, and interestingly, it provides us much better feedback about each other. I learn more about how you guys think and work from your comments in these threads than any other type. Not sure why, just the way it is.
6. If you're concerned about the content here, DO SOMETHING ABOUT IT. I surf about a dozen other sites throughout the day, and whenever I see anything interesting, I submit it. I don't feel like I've gotten everything out of the story until I get feedback here. Many stories die quickly (I'm disappointed), but many others take off. If you find something cool somewhere else, don't keep it a secret; let us know about it.
7. Vote!
8. Visit the "new" page more often. Lots of good stuff falls out of the top 30 to a quick death. Maybe we need a more prominent link to the new page, a "split screen" default, or more than 30 new links.
9. Starting a software business takes a lot more than good hacking skills and raw intelligence. You need passion, perserverence, thick skin, and big balls. Don't like something (here or anywhere else)? Shut up and do your part to make it better. Whiners don't belong here.
Sure don't mean to offend anyone, especially you, BrandonM. At least you had the balls to say what others must be thinking but won't say. And you gave the rest of us a chance to get a few things off our chest. Thanks. Hope to continue to see your name around here.
I notice that your karma is over 4000.
Could it be that you're just one of the "chosen few" who are supposed to represent the community at large? If so, it would stand to reason that your karma would go through the roof and you would be finding the site more and more interesting.
I wonder if the reverse holds true: if by choosing the "holders of quality" it also has the effect of choosing certain personality types as being "anti-quality" In other words, when a site gravitates to being represented by a select few (which we all know happens) does it also tend to completely turn off people of opposite personalities and tastes?
This is probably not such a big deal with a low volume of articles and submissions, ie, everybody gets a chance to make a difference in the conversation. In a 10-article-per-day site, everybody's opinion can get something on the front page. But as the volume increases, individuals get lost in the system and those that don't exactly match the opinions of the select few feel more and more isolated.
Just wildly speculating here. I didn't mean to criticize you personally, just using your account and comment as an example. Hope that is okay.
I've never been "chosen" for anything in my life. That's why I'm a hacker. With few exceptions. everything I've ever had or accomplished has been a result of hard work and perserverence (good thing I love what I do).
"represent the community at large"
Ha, if only it were true! I often wonder if I'm the most contrary person here. I've never "audited" my comments here, but I'd be willing to bet many of them were "against the common grain".
A little background: I have absolutely NO formal computer education. I was a math major in a college without a computer science department. I have written tens of thousands of original programs in many technologies in over 80 companies. I don't fully understand social networking, or ad based revenue, but I do know how to build what users need. I don't blog, facebook, twitter, instant message, or text. I often worry that I don't fit in here, because I feel so different from the "typical" member, if there is such a thing.
My karma score is a direct result of participation, nothing else. Anyone can do it. I have no idea (nor do I care) if that score "entitles" me to anything. I love it here. I sit at my terminal from 12 to 16 hours 6 or 7 days per week. Except for the few precious hours with friends and family, THIS is my social life. Don't even have to get up to interact with other humans, alt-tab does it.
Yes, you are wildly speculating. But that is okay. How else would we know about things unless we talk about them?
1. Thanks, I appreciate the reassurance that I have been a decent member of the community.
2-7. The problem is that I don't use other link-aggregating or site-stumbling sites, and I don't have any RSS subscriptions. I'm just not that kind of Internet user. For me, Hacker News was my portal to all the things I'm interested in, but lately that has shifted too much for my tastes. If nearly 400 comments, over 400 votes on submissions, and countless reads through stories and entire comment threads are not enough for my complaints to carry weight, then I'm sorry. I don't believe that you and I are the same type of Hacker News user.
8. When I have some time to burn and I've perused the front page, I do visit the "New" page.
9. This wasn't intended to be a whining post, and if you would like to browse my old comments and point out where I have whined in the past, I will certainly consider my behavior. This submission was merely to attempt to gauge how some of the older users feel about the shift in purpose of Hacker News (what new user would vote it up?), and to see if there would be any point where others would consider leaving, as I plan to.
No, I'm not really offended, so don't worry about it. And that was the main purpose of this submission: for others to share how they felt about what exactly would be the point where this site has changed too much. Either some changes are going to be made, or this site will follow the trajectory that so many other great discussion-based communities have followed.
We may not be the same kind of internet user, but we do share the same fear. This is my 3rd "home" on the internet and by far the best. I hope I never have to look for my 4th.
(The whining remark wasn't directed at you. Just trying to make a general point about the fine line between constructive criticism and whining. I think the difference is in the action taken for improvement.)
I think people may have formed an internal model about the sort of things they think appropriate for this (or any such site) and get concerned when a submitted article doesn't conform to this. I feel it occasionally on programming.reddit.com, sometimes there are articles there that seem to be on there purely because it mentions the word "computer" or "internet".
But to my surprise when I visited archive.org and pulled up the front page from just over a year ago I see:
http://web.archive.org/web/20070221033032/http://news.ycombi...
There really isn't much change at all in the sort of articles being posted.
Out of curiosity, where will you go instead of HN? I've not encountered anything better than this place yet - the high comment quality especially.
I'm in the US, but I'm fed up with our politics and recession, and seeing how flawed the system is is just depressing. That's why I only visited the politics sub-Reddit for about a month before I quit going, and now I see some of the same discussions here in submissions and comment threads.
I also think you're right about the internal model. For me, though, my internal model of Hacker News isn't just about what I think is appropriate in a website, but what I think adds value to my life. When the site doesn't meet the model, it is inherently less valuable to me.
I will not go anywhere instead of Hacker News. I might pop in a couple times a month to see the "hacker sentiment" and to be referred to some interesting blog posts, but my overall goal is to waste less time on the computer and spend more time getting real things done.
I think that this is partly, because a lot of the truly great content has already been posted. Hacker and geek culture is really a small town once you get over the big learning curve. There are only so many PG essays, Greenspun essays, Spolsky articles, Jargon file content, The Cathedral and the Bazzar, In the Beginning Was the Command Line, PMarca blogs etc.. that can be posted. After the bulk of these got posted, the early adopters started having to work harder to find content that was as good. As they say, the low hanging fruit is easiest to pick.
I know that this has been true for myself, at least. I put a lot of hard work into finding great content and posting it on Hacker News. But, the well for me is running dry. It's harder to come across articles/essays/posts that are as good , because there are only so many great, game changing articles about hacking and startup life that have been written.
So, I've changed my focus from trying to post great articles, to trying to improve the quality of comments that I can post. And, I'm spending more time getting through school so I can get to work on my startup instead of just reading about them.
The only reason I mention it is that you can actually kind of measure this. If news.yc was perfectly homogeneous, then the new page and the front page would be exactly the same. So you can measure the percent homogeneity by dividing 25 by the number of new submissions since the oldest item on the front page. I would bet that there is some ideal number.
A question for everyone criticizing Brandon for being a "Drama Queen" (my choice of words, not yours). Which one of Paul's "levels of disagreement" applies?
Point two:
You can make the argument that his "suicide note" is contributing to the problem, but unless he has a flotilla of sock-puppets voting it up, his note is on the front page because the people here--on aggregate--vote things like this up.
Which means that he's right! If the community does not want to read this kind of thing and discuss this kind of thing, it would be voted into oblivion.
You can hardly blame him for 27 of the 28 points it currently has.
Karma systems have an awful track record, from my experience. It is not a solved problem.
I'm sure there are many reasons not to do this, but the idea still has merit, IMO.
As more "poor" submissions are made, the "mediocre" submissions begin to look better and the community standards are lowered.