The initial thoughts on schema would be that we need to track users, posts, comments, and favorites.
This is in no particular order. I'm just adding things to this as I think of them.
- user registration
- user login and authentication
- user authorization
- *** the ability to post
- *** the ability to comment on posts
- *** the ability to up or down vote
- *** an algorithm to weight the posts
- The algorithm needs to know the view count as well.
- the ability to favorite posts
- a division between the various programs GA offers
- *** a search feature to search the text of posts (stemming)
- cheatsheet and helpful resource links
- a user profile with github link and photo
- save the searches to better improve the stopword list and to have a body of data for a neural net
- left-side navigation bar
- mobile friendly
- wooden floor background for the sidebar nav
- white main area
- red accents (probably hover overs)
- First search the DB via tsquery, returning the matches
- Then search the comments for the highest scores
- How do I prioritize the post score and the comment score?
- Answer for ^... I think I simply add the two scores together. That way a hot post with hot comments will supercede a hot post with poop comments. Further, it gives older posts with really hot comments to rank. Do I even keep track of comments score separately or do I simply add any value from the comments to the post score?
*** Every search needs to have a single | pipe on it to make it an OR search between words