It would be funny if they cheated themselves to get this on the front page.
HN Cheaters: Catch me if you can
71–80 of 176 posts
Re: HN Cheaters: Catch me if you can
#72I've been on the front page a few times. I don't actually know anyone else on HN who would or could vote my submissions up (that I'm aware of). It seems to be a combination of content, timing and luck.
Hence there are some front page posts that have nothing to do with cheating. Just to add some balance here.
Re: HN Cheaters: Catch me if you can
#73Re: HN Cheaters: Catch me if you can
#74Earlier quoted context omitted.
Or your link could be about inequality, net neutrality, spying, self-improvement or a hot language/framework. HN has pretty predictable interests.
Lately everything deep learning seems to get upvoted, even rather mediocre stuff. I like deep learning and certainly think it is interesting but it is getting ridiculous.
I think the mediocre stuff is often necessary for people on the outside of relatively specialist areas - like I am - to have a way in. Not that I've been upvoting much deep learning stuff, but I'm sure I've upvoted similar.
I seem to recall that the same thing happened when HN had a fling with Bayesian statistics...last year? Two years ago?
I think it's a necessary consequence of having a large, broad technical audience that really enjoys investigating very narrow niches. Eventually even the people on the 'edges' of that topic's appeal want/need to find a way in.
Re: HN Cheaters: Catch me if you can
#75> This chart looks perfectly normal since the green line is smooth and the blue line is relatively stable. The first chart goes to over 1,000 votes. The other ones stay one or two orders of magnitude lower. Cheating or not, it's absolutely normal that a stochastic phenomenon looks smooth when there are many events, and not smooth when there are few of them.
Exactly this...
This comparison is statistically on very shaky ground at best.
Sidenote: HN page rank is not solely based on upvotes (for example the recurring recruitment thread). So analyzing it like it was will not provide reasonable results.
Re: HN Cheaters: Catch me if you can
#76What we need is collaborative filtering:
Posters would self-organise and voting gangs would talk to themselves. Gaming it would mean that voting gangs had to have convincing interests etc.
An old writeup on the idea: http://williamedwardscoder.tumblr.com/post/15581427232/self-...
Re: HN Cheaters: Catch me if you can
#77I have found my success/failure to be hit and miss with things I have submitted over the years... I just thought it was luck
Re: HN Cheaters: Catch me if you can
#78Good that someone raised this topic. 1. Many good articles & posts don't get upvoted. 2. There are pre-formed HN groups who game the system often. My Feedback 1. Some special users can be given special Front-Page-Post Button permission. Say, 3 Top Users have voted this Post for FP(Front Page) 2. Frontpage should not be the highlighting part of HN. At least it looks like that as of now. The other good parts of HN: new…
Additional feedback:
5. Use of better colors: For following keywords or topics & colors for users by active | top | influencers etc.
6. Verified accounts.
7. Use of on Demand Badges like: Trustworthy User Badge, Looking for(Badge), Looking for co-founder, Hiring, Remote jobs etc.
All these can be grouped under a Paid plan? I would definitely pay for value & additionals, even a good mobile app experience is missing.
8. There is lack of networking and peer discovery in HN.
They should add community managers & make this evolve. Take it to the next level.
9. Follow options: Follow this discussion, Alert-me's, Follow Posts by keywords.
10. Finally, there is lot of scope to Improve.
HN team & YC should Rise up to the expectations and deliver richer experiences. What are they doing?
11. Add a separate Top Link: Voice-UP your concern.
This is for: Alerting, finding loopholes in the system(gaming the system etc), User Site feedback, new interface ideas.
Creating a vibrant HN ecosystem is the way looking forward. Lets do it.
Re: HN Cheaters: Catch me if you can
#79Earlier quoted context omitted.
"I don't feel anything has changed over that time." Approaching 3000 days on HN. The things I notice most: * increase in users has increased the churn rate of new submissions. * the new page is dead because of the submission rate. good stories disappear off the stack quickly. * the up-click is not a good indicator of story quality. I now see stories with up to 80 votes sans one comment. I used to read the comments BE…
There are a few people who make very many submissions. Some people submit more than 2 stories per day, every day. And those are often duplicates-but-different source of other articles already on HN. Better enforcement of the deeply interesting guideline would help but stories that are intensely interesting get many upvotes so it's unlikely to change.
If there was a submission quota, either time-based or karma-based, the "new" page would clear up rapidly, especially if obvious spam was not displayed as well (though that may be a result of my "show dead" setting).
Re: HN Cheaters: Catch me if you can
#80So how do you fix it? I don't really see another way to rank large amounts of stories other than points. Sure, you can try to detect cheating, but you want a minimal amount of false positives, too. Edit: missing word
Few options: make system more complex, add randomization, increase moderation effort or make it less open. I am fan of adding randomization. It was previously proposed: show few random new items on a front page in random places, maybe with a little threshold. I was thinking about community page destined to fork. After community reaches certain threshold (size, activity) there is automatic fork and 2nd generation comm…
I think the easiest option would be more transparency. If HN had a system in by which its users could see how things get to the front page (e.g. with graphs and data), rather than simply having to trust a magic points and time algorithm, it would become immediately obvious to the community at large if/when someone were cheating and we'd be able to react accordingly, over time learning how to identify and punish these.