One could notice that the number of submitted and reviewed Utopian contributions grew during the last weeks. Unfortunately, due to delays in upvoting contributions, the weekly statistics from Top of Utopian were less accurate. I introduce a few of minor enhancements in this post, which I believe can increase the accuracy of the presented numbers and readability of the table.
Note: Although the main account @utopian-io publishes the Top of Utopian posts, which are related to the blog news of the Utopian.io project, this post focuses on a side project called utopian.rocks maintained by @amosbastian. The data can be obtained from there.
Potential Issues
Due to delayed votes that Utopian casts for the contributors, the data presented in the table were not accurate. One could notice that not all columns corresponded to other related columns.
Number of Rewarded Contributions and Total Rewards
ToU statistics
For example, we can see that the last two rows say that Utopian rewarded contributions from those categories but the column with the amount of given (received) rewards remains zero.
If you read through the code, which anyone can inspect, you find out that the number of rewarded contributions relies on the assumption that adequately scored contribution is rewarded. This sole fact may not be wrong, however. The issue is the delay of the vote, which comes after the data are gathered, and that introduces the inaccuracy.
Definition of a Top Contributor
The table also includes the column with a top contributor in a selected category. Readers have little chance to understand how the top contributor is selected. Current implementation takes the most active contributor, which means that they published the highest number of contributions.
I honestly believe that the number of contributions doesn't make one a better contributor. As Utopian aims to have as good submissions as possible, the quality should take part in deciding whether a contributor is at the top.
If we don't want to introduce complex metrics, we can use either the score of contributions or the amount of rewards that contributions received.
Suggested Improvements
I present a few suggestions of a potential implementation to improve the issues mentioned in the previous section.
Introduce New Column in the Table
As I already mentioned, the number of rewarded contributions is calculated as the number of contributions that Utopian upvoted or will (may) be upvoted in the future. Solution to this problem is simple. We can count only those that received a vote from Utopian.
However, I think that we can add a new column "Rewardable" that says how many contributions are eligible for a potential vote.
With this addition, the following holds:
reviewed >= rewardable >= rewarded
Change the Definition of Top Contributor
From my point of view, we could change the definition of the top contributor in a category. There are a few solutions that I can think of.
- Top contributor is the one that has the highest average score of contributions in a category.
- Top contributor is the one that has the highest sum of scores of contributions in a category.
- Top contributor is the one that has the highest sum of weights of received Utopian votes for contributions in a category.
The first potential solution could be favourable for people who contribute less but receive a high score. Let's see a simple case with two contributors A and B. A has one contribution with a score of 70 and B has two contributions with a score of 50 and 80. In this case, A is the top contributor.
The third one seems to be working fine. Yet, we can't forget that Utopian has some bonuses for users who set beneficiaries for Utopian and bonuses for VIPO members. If the increased weight of the vote is not acceptable, then this solution may not be the best one.
Thus, the second suggestion looks most promising. Also, in this case, we can find an issue with finding the top contributor. Iamutopian category does not follow the same evaluation process as other categories, which means that the result may still be inaccurate. Besides, this suggestion looks to be working the best of the three proposed here.
Conclusion
This post focused on a table with weekly statistics included in the regular posts published by @utopian-io. I highlighted potential issues with the current implementation and proposed adjustments. I believe that introduced changes will better reflect the columns' names and purposes and will increase the readability of the presented numbers.
