Thanks to @antonchanning my idea for a better steem feed based on discovery is now documented on github!
In my last post I asked for any volunteers to help me by testing my algorithm that is currently being run on a spreadsheet with google apps script. @antonchanning volunteered and is now hoping to turn this idea into a webpage as part of the @blockpress project.
If anyone else would be interested in trying out my feed algorithm, let me know in the comments.
If you don't want to read my previous posts, here is a quick summary of how it works:
The program looks at every vote you have made and finds the top 100 people that have voted most similarly to you in the past.
Those 100 people become your personal curators of content. Whenever one of them votes on a post, it will show up in your feed.
As you vote/skip over posts in your feed your curators will receive scores based on whether you vote on the things they have or whether you skip past them.
The curators with the best scores will have a greater influence on the order of your feed.
Every day your worst curator is replaced with a new one.