Welcome to Codidact Meta!
Codidact Meta is the "town hall" (meta-discussion site) for the Codidact community network and the Codidact software. Whether you have bug reports or feature requests, support questions or rule discussions that touch the whole network – this is the site for you.
Post History
Just a thought that occurred to me: One problem when seeing the Wilson score when not knowing it is that it's not clear how to interpret it. I think that could be helped by simply displaying 2*sco...
#2: Post edited
by
celtschk
·
2025-11-15T06:38:30Z (10 months ago)
Added a list of advantages
- Just a thought that occurred to me:
- One problem when seeing the Wilson score when not knowing it is that it's not clear how to interpret it. I think that could be helped by simply displaying 2*score-1 instead. This is a linear and strictly increasing transformation, so it would still fit with the ordering according to Wilson score, but its sign tells immediately whether there were more upvotes or more downvotes, just like with the difference.
To avoid fractional numbers, but still get across clearly that it is not an absolute number, it could be displayed as percentage, so the display would be between −100% (very bad) and +100% (very good). That also would immediately make clear the limits, as when seeing a percent sign, everyone knows that 100% is typically the maximum you can get.
- Just a thought that occurred to me:
- One problem when seeing the Wilson score when not knowing it is that it's not clear how to interpret it. I think that could be helped by simply displaying 2*score-1 instead. This is a linear and strictly increasing transformation, so it would still fit with the ordering according to Wilson score, but its sign tells immediately whether there were more upvotes or more downvotes, just like with the difference.
- To avoid fractional numbers, but still get across clearly that it is not an absolute number, it could be displayed as percentage, so the display would be between −100% (very bad) and +100% (very good). That also would immediately make clear the limits, as when seeing a percent sign, everyone knows that 100% is typically the maximum you can get.
- **Edit:**
- Let me add the advantages I see of this scaled Wilson score:
- * Posts start out with a score of 0. That is a natural starting point.
- * The sign of the score immediately tells you whether the post is seen positively or negatively by the community. Indeed, the sign is exactly the same as for the vote difference on Stack Exchange.
- * The absolute value of the score also has an independent meaning: It tells you how strongly the community feels about the quality of the post.
- * It also fits perfectly with the existing graphical display: The middle from where the bars start is zero, the direction of the bar is the sign, and the length of the bar is the absolute value.
- * And, as mentioned before, the order is exactly the same as the Wilson score before scaling.
#1: Initial revision
by
celtschk
·
2024-06-08T09:46:33Z (over 2 years ago)
Just a thought that occurred to me: One problem when seeing the Wilson score when not knowing it is that it's not clear how to interpret it. I think that could be helped by simply displaying 2*score-1 instead. This is a linear and strictly increasing transformation, so it would still fit with the ordering according to Wilson score, but its sign tells immediately whether there were more upvotes or more downvotes, just like with the difference. To avoid fractional numbers, but still get across clearly that it is not an absolute number, it could be displayed as percentage, so the display would be between −100% (very bad) and +100% (very good). That also would immediately make clear the limits, as when seeing a percent sign, everyone knows that 100% is typically the maximum you can get.