SN release: trust algo v4, reward changes, pwa enhancements
3126 sats \ 5 comments \ @k00b 6 Jul 2023 meta

trust algo

tldr trust is more accurate, more evenly distributed, and the ramp to gaining trust is much more gradual rather than threshold based. The algorithm also generates a unique POV for every stacker, ie provides personalized feeds. Personalized feeds aren't live yet, but it's one of the next things we'll add (and will enable things like mutes, direct algorithm configuration by stackers, and "following").
The new algo is live and has been applied retroactively.

reward changes

We are running an experiment where rewards are only given to zappers. Zapping is very important yet it's the least naturally incentivized behavior on the site. My hypothesis is that incentivizing zapping will lead more folks to zap, signal will be better distilled, and because there's more zapping posters and commenters will make at least as many sats as they did before.
The zapping reward rules are exactly the same as before. You're rewarded in proportion to:
  1. how early you zapped the content
  2. how many sats you zapped the content
  3. how much trust you've earned by zapping over time
  4. how well the content you zapped ranked relative to other posts/comments
50% of zap rewards are given to top post zappers. 50% of zap rewards are given to top comment zappers. These are the same rules as always. The only difference is that all daily rewards are exclusively paid out to top zappers so zapping rewards are 3x bigger than they were before.
The title kind of says it all. We've been told rewards aren't obvious so this is a step in that direction.

pwa enhancements

  1. notifications close on click
  2. clicking on a reply notification takes you to the full comment thread
  3. iOS splash screen
  4. maskable icons for android
  5. enhanced app install screens on supporting browsers (mostly chromium based)
write
preview
related posts
148 sats \ 1 reply \ @OneOneSeven 7 Jul 2023
Reward change is awesome
reply
109 sats \ 0 replies \ @k00b OP 7 Jul 2023
I'm glad! I've been wanting to this for like a year but there were concerns that it wouldn't make sense to people.
reply
43 sats \ 0 replies \ @Bitman 7 Jul 2023
Genius! Should really incentivize people to be active in a more circular economy.
reply
10 sats \ 1 reply \ @mudbloodvonfrei 7 Jul 2023
Wouldn't this disincentivize zapping posts that might not be as popular with others, newbie posts, subs that don't get as much attention, etc? Before this change I could earn sats by posting content that people find interesting, and spent sats by zapping things that I find interesting. Now the two are tightly coupled, so I am likely to earn less for good posts unless I'm also a top zapper (of course if I post interesting content I still earn from the zaps people give me directly).
Anyway, it's fascinating to see all of this experimentation going on..
reply
0 sats \ 0 replies \ @k00b OP 7 Jul 2023
Wouldn't this disincentivize zapping posts that might not be as popular with others, newbie posts, subs that don't get as much attention, etc?
The presence of an incentive for one thing isn't a disincentive for another. It doesn't cost you extra to zap things that will be unpopular.
Before this change I could earn sats by posting content that people find interesting, and spent sats by zapping things that I find interesting.
I think you misunderstand the change. None of what you describe here is affected by this change. You still earn sats for anything people find interesting and can still zap what you find interesting.
Now the two are tightly coupled, so I am likely to earn less for good posts unless I'm also a top zapper (of course if I post interesting content I still earn from the zaps people give me directly).
Again I'm not sure you understand. There are two ways to earn on SN:
  1. directly from zaps by other users
  2. rewards (which are SN's revenue that it chooses to give back)
If you post good posts, (1) still totally and absolutely applies. The hypothesis of this change is that the quantity received in (1) will be higher and make up for the lack of (2) and then some.
reply