IDEAS

Help shape the future of CurseForge!

Balanced Reward System

At the moment the reward system biases people towards a) Creating a lot of small utility mods b) Creating a lot of library mods to get more downloads in order to get a bigger share of the reward pool. Although I know it is difficult to find a good solution, the current system disincourages creating bigger mods especially on the long term (since those are more difficult to port to newer minecraft versions too)

  • Team Overwolf
  • Jul 1 2020
  • Aug 5, 2020

    Admin response

    Balancing and tinkering with the system is tricky and we're treating this very cautiously - Our first step is to migrate CurseForge smoothly from Twitch, including the reward system as it stands. After that we'll look into different option and how we can rework the system and address questions such as the distribution between separate mods, mod packs, content quality, popularity etc.
    We have a lot to learn and we'll introduce any change if required gradually and adequately.

  • Attach files
  • ROM Void commented
    30 Nov, 2020 04:39pm

    Being a owner of a Library mod and working with them most of the time. While they do offer a lot of help and make creating / maintaining mods easier for the developer, since they most of the time don't actually add anything playable for players. I always thought these should be excluded from accruing points. And points should be awarded to the actual content adding mods.

  • Henry Loenwind commented
    24 Nov, 2020 05:39pm

    For modpacks it's relatively easy, having more mods the same author gets rewards for could have diminishing returns. 1 mod = 1 share, 2 mods = 1.9 shares, 3 mods = 2.7 shares, 4 mods = 3.3 shares, 5 mods = 3.8 shares, 6 mods = 4.2 shares, 7 mods = 4.5 shares, 8 mods = 4.7 shares, 9 mods = 4.8 shares, each additional mod +0.1 share.

    There's still some potential for gaming by teams, e.g. having 3 people and 3 mods could give either 3 shares or 2.7 shares in total (each mod with one author vs. all mods with 3 authors) but I think that should be low enough it can be ignored.

    For direct downloads it's not quite as easy, tracking which mods a user downloads e.g. in one day could work. Alternatively the number of mods an author gets rewards for on a day could be used. The issue here is that the number of downloads for those mods are not the same, so applying the factor gets a bit trickier---there's the potential that the total reward amount goes down when one more mod is downloaded.

  • Arco Dielhof commented
    13 Oct, 2020 10:47pm

    As some one else said:
    "Rewards being scaled by mod size is a bad idea. if implemented a lot of mods would suddenly take on code bloat."

    I think it is fine with what it is. If we get more based on how big our mod is, I will start writing code that isn't efficient and small, I'll just write code that take 50 lines for doing what can be done in 5.

    Terrible idea that would be.

    I also think that frequency of updates, combined with amount of downloads and popularity would be a better alternative!

  • Kree Zuhl commented
    19 Aug, 2020 10:23am

    rewards being scaled by mod size is a bad idea. if implemented a lot of mods would suddenly take on code bloat.

    A possible solution might be a system that factors:

    1. how often the content creator logs on to the hosting site

    2. how often the content is updated

    3. how often the content is downloaded

    4. how often the content's host pages are viewed

    with heavier weights being applied to 3 and 2 respectively. Granted 3 and 2 are certainly weight by 1 already.

    A 5th possible factor could be added too, a rating. players would be encouraged to rate content. Granted this could be abused.

  • Androsa commented
    11 Jul, 2020 01:55am

    Perhaps points could be scaled by the size of a file? Generally speaking, mods with more content end up having large file sizes, though this could be abused by authors by adding a large unused file to artificially bloat size.
    Alternatively, have it so that certain mod types (for example, library mods) may have lower or restricted points earned, as they can be used as reward system abuse. Another approach would be making mods have an opt-out option so they don't earn points. Perhaps a more in-depth review system to determine whether a mod should earn more points based on content?

  • +5