Ask HN: How do you filter Hacker News?
I just glanced at my RSS Reader stats and was a bit horrified to see that I am sent over 100 items per day from Hacker News. Many of those items are duplicate or dead stories that still make it into the RSS feed. I would love a lightweight version of the Hacker News feed that only shows items above a certain threshold (say, 20 points).
I can't believe everyone here does this filtering manually. What do you do to separate the wheat from the chaff?
Comments
I have come to the conclusion that RSS is definitely not the optimal way to interact with HN.
If I have only a minute, I check the front page to see what the big stories are. If I have more than a minute, I check the "new page" and try to do my bit by upvoting anything interesting (maybe 3-4 stories) and answer any "Ask HNs" that are in danger of languishing. If I have even more time, I'll read the comments on the front page and, a la xkcd, make sure nobody is wrong on the internet :-)
In fact, I spend relatively little time actually reading original submissions, unless I am multitasking with something else.
I check a few times a day, as my schedule permits and try to up vote interesting submissions. If a large crowd only looked at stories after they were highly ranked, how would any stories ever get highly ranked?
Shameless self-plug, but I love this greasemonkey script I built and still use it across 3 machines. Pretty much you enter your tags/watchwords and it does a string match. It's very simple (I taught myself jQuery to write it) but I can't live without it.
http://userscripts.org/scripts/show/65211
I was having the same issue so I created a simple site that polls the site every 2 minutes and attributes points according to where in the top 30 a story was located and maintains the rankings for every day.
It is ridiculously simple, but I find I have a much better experience with HN content while using it and the top benefit is that I can afford to miss a day or two and still manage to not miss anything. Check it out! It may (or may not) prove to be useful for you too. Also, no ads or similar annoying stuff. :)
http://hnrecap.com
P.S. It is currently a bit messed up (stories are entered multiple times in the table) because of how (poorly!) I handled articles that make it across many days in the hour that it took me to code it. I will fix that ASAP and add support to view the recap for past dates and search the site.
When I'm busy, I usually skim through the front page and click on links with titles that are very relevant and interesting to me (things that are related to programming languages I use, etc). If I notice a story has been getting a lot of upvotes and I keep reading the title because it persists on the front page (for a day or two), then I will finally read the content. Throughout the day, I look through new submissions and vote up as necessary.
If I have time to check an rss reader, I don't mind the cost for opening HN, it's a fast loading page and the scoring metrics help filter great stories. So I'm a manual filterer
If I were to automate filtering here, I'd scrape the first page once a day
I skip the RSS reader for HN and use the Twitter feed. The rest of the time I'll check in on the main page occasionally.
http://twitter.com/newsycombinator
I've found the Twitter feed useful to get the most highly ranked articles, but every few days I'll take a quick 5 minute break from coding to look through some of the new posts.
I don't check HN at all when working on anything that I own myself.
I just log in and scan the posts when I take a coffee break. Quick and easy for me.
I don't. I just glimpse the list and click on whatever I like.