4 ms·

Show HN: Do "How to" or "Why to" articles make it to the front page more often?

16 points by ZanderEarth32 14y ago

gravitronic 14y ago
It would be much more interesting to see the numbers as a percentage of total "How" and "Why" submissions that made it to the front page. I was sad that yesterday I did a "Show HN" with a fun website I made for DJing youtube videos and it got a single upvote [http://news.ycombinator.com/item?id=3890935 http://news.ycombinator.com/item?id=3890935] - sometimes front page is just not in the cards I guess.
- ZanderEarth32 14y ago
  Yea, I realize I could have done something more interesting with the data but I was really trying to keep it simple and focus on getting the code to work. That was my main goal, to learn how to pull something small like this off. Thanks for checking it out though. As for your Show HN not making the front page, I am not sure what motivates HN to click through on a submission and then up vote it. Possibly, the time of submission has a lot to do with it. There is a definite advantage to have by timing things right, http://blog.kissmetrics.com/science-of-social-timing-1/ http://blog.kissmetrics.com/science-of-social-timing-1/
- mrspeaker 14y ago
  Maybe 'cause it's not as good as this one I made in 2009 ;) http://turntubelist.com/ http://turntubelist.com/ Nah, but seriously - timing is everything with HN submissions - and if you don't get a few votes in the first 10 minutes of submission, then it's down the tubes. I find that posting early in the morning NY time (I'm not from the US, so it's actually late afternoon for me) is the best. I don't know if there's any data to back that up though!
  - gravitronic 14y ago
    HAH, nice website :) Yeah I figured for a music thing, mid-afternoon eastern would be the best bet. But the other posters are correct, you only last about 10 minutes on the "new" page at peak times.
- EzGraphs 14y ago
  It is pretty clear that when there is more activity (specifically on the "new page"), a site is more likely to be upvoted. It's one of the reasons I built http://hn4d.com/ http://hn4d.com/ (but this currently is only tracking the main page). The challenge is - others know this, so the number of submissions at prime time tends to increase and your post can get knocked down the list pretty quickly.
carlsednaoui 14y ago
Perhaps this can be attributed to a greater amount of submissions with the word "how". As gravitronic mentioned, it would be interesting to see everything as percentages.
bashzor 14y ago
For these numbers to make any sense at all, you need to scrape the new posts. Now you can only tell what you measured: How many how/why stories were submitted. By scraping the new posts and count the ones that make it to the homepage, you can tell which HN users prefer, if HN users prefer how/why stories versus other stories, and how many how-stories versus how many why-stories are submitted (probably closer to how many are on the web, so you might get an impression of that too). The current numbers are meaningless data, not information. Oh and you might as well track 'show HN' and 'ask HN', or just create a wordcloud. THEN we have really useful info!
zalew 14y ago
I'd like some data on if there are more 'why' posts that actually deliver an explanation of a comprehensively analyzed technical/business problem, compared to that common derivative opinions ping-pong.
haldean 14y ago
This is a fantastic first project to get started with Python! If, in the future, you make another scraper (or anything else that requires retrieving data from the web) I'd recommend using the Requests [0] library. It's much cleaner and easier to use than urllib, and it makes it very difficult to do something wrong. [0] : http://docs.python-requests.org/en/latest/index.html http://docs.python-requests.org/en/latest/index.html

16 points by ZanderEarth32 14y ago

It would be much more interesting to see the numbers as a percentage of total "How" and "Why" submissions that made it to the front page. I was sad that yesterday I did a "Show HN" with a fun website I made for DJing youtube videos and it got a single upvote [http://news.ycombinator.com/item?id=3890935 http://news.ycombinator.com/item?id=3890935] - sometimes front page is just not in the cards I guess.

ZanderEarth32 14y ago
Yea, I realize I could have done something more interesting with the data but I was really trying to keep it simple and focus on getting the code to work. That was my main goal, to learn how to pull something small like this off. Thanks for checking it out though. As for your Show HN not making the front page, I am not sure what motivates HN to click through on a submission and then up vote it. Possibly, the time of submission has a lot to do with it. There is a definite advantage to have by timing things right, http://blog.kissmetrics.com/science-of-social-timing-1/ http://blog.kissmetrics.com/science-of-social-timing-1/
mrspeaker 14y ago
Maybe 'cause it's not as good as this one I made in 2009 ;) http://turntubelist.com/ http://turntubelist.com/ Nah, but seriously - timing is everything with HN submissions - and if you don't get a few votes in the first 10 minutes of submission, then it's down the tubes. I find that posting early in the morning NY time (I'm not from the US, so it's actually late afternoon for me) is the best. I don't know if there's any data to back that up though!
- gravitronic 14y ago
  HAH, nice website :) Yeah I figured for a music thing, mid-afternoon eastern would be the best bet. But the other posters are correct, you only last about 10 minutes on the "new" page at peak times.
EzGraphs 14y ago
It is pretty clear that when there is more activity (specifically on the "new page"), a site is more likely to be upvoted. It's one of the reasons I built http://hn4d.com/ http://hn4d.com/ (but this currently is only tracking the main page). The challenge is - others know this, so the number of submissions at prime time tends to increase and your post can get knocked down the list pretty quickly.

carlsednaoui 14y ago

Perhaps this can be attributed to a greater amount of submissions with the word "how". As gravitronic mentioned, it would be interesting to see everything as percentages.

bashzor 14y ago

For these numbers to make any sense at all, you need to scrape the new posts. Now you can only tell what you measured: How many how/why stories were submitted. By scraping the new posts and count the ones that make it to the homepage, you can tell which HN users prefer, if HN users prefer how/why stories versus other stories, and how many how-stories versus how many why-stories are submitted (probably closer to how many are on the web, so you might get an impression of that too). The current numbers are meaningless data, not information. Oh and you might as well track 'show HN' and 'ask HN', or just create a wordcloud. THEN we have really useful info!

zalew 14y ago

I'd like some data on if there are more 'why' posts that actually deliver an explanation of a comprehensively analyzed technical/business problem, compared to that common derivative opinions ping-pong.

haldean 14y ago

This is a fantastic first project to get started with Python! If, in the future, you make another scraper (or anything else that requires retrieving data from the web) I'd recommend using the Requests [0] library. It's much cleaner and easier to use than urllib, and it makes it very difficult to do something wrong. [0] : http://docs.python-requests.org/en/latest/index.html http://docs.python-requests.org/en/latest/index.html