That said, I'd have to agree with you though, there is a lot of non-hacker stuff that makes the homepage and stays there.
What frustrates me more than that though is that there is plenty of hacker stuff that gets drowned out by non-hacker stuff.
Basically if a submission doesn't have 5 votes or so (on a normal day) before it scrolls off the 'new' page it will never make the front page.
That's not a bit loss for the submission itself (I use the 'new' page as though it were the homepage) but the discussion can be as interesting or more so than the submission, and that is a loss.
This is a good point. The chances of a "rate my startup" or a good discussion around a startup business idea making the main page are a lot less than they were a year ago. I've seen submissions roll off the new page within 30 minutes when the site is humming, drowned out by mass-market non-hacker stuff. That's simply too fast for HN'ers to spot and upvote good stuff -- and the number of upvotes required to have staying power on the main page has increased as well, compounding the problem.
ADD: The problem is, with a loose definition of hacker-worthy news and an increased audience, the field of potential material broadens immeasurably.
It's the same problem with humor: I love to make a joke or read something funny once a day or so. But if 10 thousand people each joked on HN once a day, the place would completely change character.
At some point, this paradigm simply doesn't scale. To me at least, looks like pg and others have been tweaking things under the hood for a long time. We might be reaching a natural limit.
These are your golden users -- great people to have.
I simply think it's a matter of just too many people posting too much stuff.
I've toyed with the idea of a site like HN except it refuses to work with any domain that has too much traffic or submitters that submit too many articles. This would force folks a) to actually evaluate the material instead of the popularity of the source, b) be constantly re-tuning as certain sites grew in popularity, and c) spotlight some truly original thinkers/issues
I liked HN when we almost exclusively talked about startups: pitches, VCs, iterating, marketing, sales, etc. But that got boring to everybody (including me) because after a while it seemed like we were hitting the same material over and over again.
These are not easy problems to solve.
EDIT: You know what I'd like to see? Anonymous startup failure stories. I would pay to read a few insightful startup failure stories each day. The way it is, we get a good one about once a month here.
Yes, but I'd rather not visit those sites and only read the best articles (the ones that get linked from here). HN as an aggregator works very well in that sense. I think the problem (if there is any) is just the number of users, nothing more.
So long as you don't find too many things inappropriate, apparently. I've had my flagging rights revoked. The first time I made a case to get them restored. The second time I simply decided to stop being as involved.
Just two examples:
facebook in latin ?
Why Does Facebook Want to Suck the Fun Out of Unfriending?
--
The interesting thing is that over the holidays the quality actually increased from 'normal' weekdays.
One could then link to their faves list in their profile so that others can see these links. Like-minded individuals could then see what links their counterparts found interesting.
I suppose this wouldn't have to be a specialized site. Any HN member could build a page in static HTML or whatever programming-language listing their favorite links.
http://jimlawless.wordpress.com/2009/12/27/preserving-my-fav...
And then someone comes in and points to the guidelines and how it's a newbie mistake to feel like the whole board is going to hell.
Then edwin519 comes and tells us, flag and submit more what you find interesting and vote up things that you want.
So far, this community seems to be self-correcting, as long as we're self aware that we're not learning from the front page like we use to. We can all get our yuks and giggles from elsewhere.
I am experimenting with something that may help. Here is a URL to the RSS feed of Hacker News, filtered to contain only hacker stories:
http://www.lemmatica.com/emit_topic?key=aglsZW1tYXRpY2FyDQsS...
There is an automated text classifier behind it and I'm continuously training it for hacker stories. The classifier and my definition of what are hacker stories are, of course, imperfect, so there will be errors (I'm training the filter with some delay, usually once per day). If you want to filter your own feeds like this please ask me for an account.
"Stories on HN don't have to be about hacking, because good hackers aren't only interested in hacking, but they do have to be deeply interesting.
"What does "deeply interesting" mean? It means stuff that teaches you about the world. A story about a robbery, for example, would probably not be deeply interesting. But if this robbery was a sign of some bigger, underlying trend, then perhaps it could be."
But who am I to complain? I don't contribute that much either.
You could argue that just about any story was deeply interesting or a sign of some bigger underlying trend.
I think the rule of thumb should probably be 1-3 stories about none-hacker/startup related post per week...
And the deep story should be a hacker's perspective on a regular story.. basicly nothing that would ever make it to reddit's front page.
Interesting discussion on the direction of hacker news, I see value in both.
Edit: I see that sli posted the same thing farther down. My bad for not reading the whole thread first >.<
It's a reminder to me that I ought to vote more. Thanks for posting this.
As I understand the algorithm, a vote on a story that has been posted for an hour and is on the way down is not the same as a vote for a story that's only five minutes old and on the way up.
Somebody correct me if I'm wrong in this
As a rule I now refresh the page after voting, if the vote does not 'stick' I don't bother voting for 24 hours or so.
Since there is no clear indicator of which votes are 'real' and which ones are 'dummies' and I refuse to be treated like a six year old pressing buttons that have no effect other than 'for me' (pacifier votes ?) there may come a point where I will simply completely stop voting.
Edit: hey there downmodder, whoever you are. PG is on the record about this, I don't know where you get your information but it doesn't get much better than that.
Votes are definitely not always counted even though you get the illusion that they are.
You are assuming that your votes are actually counted, this is definitely not always the case.
I'm not seeing this. Some of what's on the site is "lazy loaded," as I recall being officially said, but to the best of my knowledge and belief a vote is a vote, and votes get counted. (There are two HN voters in my household, who visit different threads most of the time, and neither of us feel our votes aren't counted.) What would be definitive evidence of votes not being counted (as contrasted with some other possible explanations for what appears to be that condition)?
Belief doesn't enter in to it:
http://news.ycombinator.com/item?id=871202
It persists to this date, you just have to work a bit to see it in action.
These days there is too much spam to make the "new" page work.
EDIT: wow, the new page also needs |flag| on the listing page. so much spam.
But the rate of spam has gone up so as something's tagged spam, another untagged spam gets posted.
I've asked for that many times, but since it seems that both my flags and votes are being ignored again I won't bother any longer.
So I've given up flagging more than once a week and voting a more than a couple of times a day. Sadly all the older HN members I talked too say the same thing.
I wonder if the site algorithm is actually favoring newer / less active members?
It already happens, I'm quite sure. Every time I visit HN, I first visit the main page, and then visit the new page. I mostly only flag egregious spam. I see some of that disappear as I flag it.
I also like using noobcomments view
http://news.ycombinator.com/noobcomments
every once in a while to double-check for comments that plainly deserve flagging. It's too bad that noobstories view
http://news.ycombinator.com/noobstories
doesn't also allow convenient flagging.
After edit: pg in the child comment below corrected me to point out that the noobstories view now does have flag links, and I can see them now as I follow the link.
I ask because is there really a tangible difference between a poster with say 300 karma points and another with 500?
Also, since (typically) older posters have more karma points isnt the whole system skewed in favor of those posters who have been here the longest (you can make the same argument for those who post vs those who submit).
Basically what do HN users derive from having huge karma point tallies?
(edit: aside from the obvious that is)
Something to aspire too I guess.
I wish there were more search/filter functionality. It would be interesting to filter submissions by posters who meet certain criteria (points, trending info, etc.)
I typically use that to avoid the newbie fluff that's been increasing.
And to me a visible points system is part of the problem. If people 'try' to get more points then they'll post lots of stuff that might just get in. Then you end up like Digg.
I've thought sometimes that it might be interesting to weight votes by the logarithm of the voter's karma, so that the interest / approval of a highly-ranked member would be enough to promote a story from new to the main page.
What bothers me not the points,karma,content or improper comments; Its the number of fresh news that increased a lot.I started to use RSS reader and what i observe that there are so much news in HN(79) then Slashdot(13) even delicious(48)?! The more i spend time in HN the more i miss good links because given more choice to the people, they tend to put aside the quality and take all.
A bit thinking about the number of fresh news i think several problems lately addressed here(other topics) closely linked to this. More news leads to more comments and more comments creates more fuss(noise) Quality of a discussion greatly reduce as more participants, lengthy comments(self reference here) and deep reply branches.Like your favorite TV discussion program where after couple of hours everyone has a word and shouting each.The quality of comments may be,may be, increase if (high karma)&(register date) take precedence. Think about the anchorman of your favorite tv discussion program: does he start with people in the backseats or starts with some wise figure?
1) the proportion of interesting (to me) links was high; 2) the comments revealed a community of clear-thinking, logical, rational, articulate individuals.
That hasn't changed significantly since last year, though I admit that I get the sense that the community has grown lately.
I don't have a big problem with HN's current state. If it has deviated from its original goals or gone through some other metamorphosis, that doesn't bother me; I still like what it is today. In fact, if it were extremely focused on entrepreneurship and startups and such, I probably wouldn't be interested. What I get out of HN is great tech/geek links and insightful, intelligent commentary.
If the percentage of links that interested me eventually fell below some threshold, and if the comments started getting inane, irrational, nonsensical or just plain stupid, I'd probably just leave HN and find another community, or start my own.
Shame, really.
Now, I'm not necessarily saying that I agree with this point of view. But the point of this kind of social news site is that the readers get to choose what is and isn't relevant to them.
That said, I personally would like to see more hacker-oriented type stuff. My advice: submit more hacker-related content and vote it up when someone else submits it. Remember that it usually just takes one vote to move things to the front page if you catch it early enough. So your one vote can make all the difference.
http://news.ycombinator.com/item?id=363
and if participants aimed their comments and suggestions there at subthreads that have already brought up the same issue, it might be more clear which features are most in demand among regular participants.
And that post, just like all other posts was solely for HN, so why should I bother writing this stuff up and restoring an old press page, if I don't even get a single comment or upvote out of it on here?
It's like the twitter/facebook promotion we just finished, I was going to write a post with all the numbers, what worked, what didn't, what prizes are the best bang for your buck, the goals behind it etc etc. But now I'm not so sure.
Why? Because it'll take like 3-4 hours to put up a decent post with all the data, pics etc. So why should I waste my time, if there is a 90% chance the post won't even get a single upvote?
// Hackerocity on a scale from 0 to 10
// for each entry on the front page of hacker news
// Hackerocity determined by edw519
//
Hackerocity = []
Hackerocity[01] = 1 ; // Ira Glass on storytelling
Hackerocity[02] = 9 ; // Lisp OS: what has been lost: Kent Pitman
Hackerocity[03] = 0 ; // They Killed My Lawyer
Hackerocity[04] = 7 ; // MIT Open Courseware - Free Lectures
Hackerocity[05] = 6 ; // Facebook marketing: one of the better campaigns I've seen
Hackerocity[06] = 10 ; // JMatch: Iterable Pattern Matching
Hackerocity[07] = 10 ; // Regular-expression derivatives reexamined
Hackerocity[08] = 1 ; // Sorry, Shoppers, but Why Can’t Amazon Collect More Tax?
Hackerocity[09] = 3 ; // Schneier on Security: Separating Explosives from the Detonator
Hackerocity[10] = 9 ; // jQuery Plugins
Hackerocity[11] = 0 ; // Has Hackers News Become Less About Hacking?
Hackerocity[12] = 8 ; // If You’re Nervous About Quitting Your Boring Job, You’re Sane
Hackerocity[13] = 10 ; // Hello HN, take a look at my new (beta) app BonMp3?
Hackerocity[14] = 7 ; // Why the web economy will continue growing rapidly
Hackerocity[15] = 8 ; // Ask HN: How can I move to USA?
Hackerocity[16] = 7 ; // Ample SDK - Open Source GUI Framework
Hackerocity[17] = 9 ; // Palindromes (Clojure vs. Common Lisp)
Hackerocity[18] = 3 ; // Incredible paper sculptures - many cut from one sheet of A4
Hackerocity[19] = 9 ; // If You’re Nervous About Quitting Your Boring Job, Don’t Do It
Hackerocity[20] = 5 ; // Why Does Facebook Want to Suck the Fun Out of Unfriending?
Hackerocity[21] = 10 ; // Machine Translates Thoughts into Speech in Real Time
Hackerocity[22] = 0 ; // Ask HN: If there was a bury/upmod brigade on HN, how would we know?
Hackerocity[23] = 8 ; // How to commit brand suicide
Hackerocity[24] = 10 ; // The Pmarca Guide to Startups: product/market fit
Hackerocity[25] = 0 ; // [dead] Ajax on Rails
Hackerocity[26] = 5 ; // One year of Redis
Hackerocity[27] = 2 ; // New wheel for your bicycle: The Copenhagen wheel
Hackerocity[28] = 0 ; // In 2010, Demand For US Fixed Income Has To Increase Elevenfold... Or Else
Hackerocity[29] = 2 ; // Phil Greenspun debunks Malcolm Gladwell on airline safety
Hackerocity[30] = 10 ; // Startup Killer: the Cost of Customer Acquisition
//
TotalHackerocity = 0;
for(i=1;i<31;i++){TotalHackerocity += Hackerocity[i]};
//
// Yep, TotalHackerocity aint what it used to be :-)
//That would be a tough act to follow without mturk assistance.
My resolution was to search for another community. Joel Community is also good, but there is few members. Still better, the day it gets a lot; quality will drop.
I'd be curious to know, however, what those ratios look like over time.