What information should we show with "usercards"?

This is worth looking at. Since at launch all votes will be new/recent, we don’t need to work out how we’ll age things for MVP. But we should think about it when sites are a year or two old.

(We could in principle import post scores along with the posts themselves, but we have no way of matching up votes with people. On Writing we imported posts and logged the imported scores in the history, but started everything at a score of 0 on the new site. I think that’s a good approach; let the votes on this site reflect the decisions of the users on this site – and also eliminate any concerns about double-voting.)

1 Like

One thing that might be worth considering is to consider the imported votes as tie-breaker for the sorting order. That is, if two posts end up with equal score, the imported post that had the higher score on the original site is sorted earlier (but I would sort non-imported posts before imported posts).

However we also have to consider that the move of writing from QPixel to Codidact will also, technically, be an import. In this case, it might be worth importing the votes, too.

Moves from QPixel to Codidact should preserve votes. Communities that start building with us early shouldn’t be penalized. That’s different from importing a decade’s worth of votes on SE from people who might not even come here.

The tie-breaker idea is interesting. Don’t know how often it’ll come up, but worth keeping in mind.

3 Likes

Discourse doesn’t have/show such a thing (and I do not miss it, good riddance IMO).
I don’t mind the data being discoverable – i.e a couple of clicks away on a user profile or a site statistics page – I’m not keen to rub people’s noses in it.
I might opt out (at least of displaying this information on “my” card) if the software allowed me to.

Proposal: Everyone can see Question and Answer count; TL2+ can see score.

That doesn’t really solve the main issues. Showing the sheer volume of questions or answers doesn’t mean much without their quality. A bot could look good according to those stats. The total received vote score is much more meaningful to judge someone by, and also provides some of the rewards to those who spend a lot of time providing good content.

1 Like

So if we go this direction, I’d at least suggest that the default, publicly visible number should weight newer posts more heavily than older ones, or even, for simplicity, exclude posts older than a year or two.

Sorry, but I don’t agree on this. Although on SE there are highly voted posts because they were posted when standard were low, this doesn’t mean that every old post has lost significance. This is not a decision an algorithm should take. Its something that should be judged by people.

For example, on EE.SE there are quite old posts that have become canonical just because they are considered the best answer even today.

Moreover, imagine a high rep user that started his “career” on codidact writing lots of good answers, then after a couple of years has no more time to contribute as frequently. His user card will present him as a mediocre contributor to new users, whereas he might be the resident guru of the site!

This is a great way to piss off core contributors (I would certainly be!). We are trying to build a more community-focused site than SE. I want to be judged by my peers, not by an algorithm! If my contributions have grown obsolete, I’m ok there is a mechanism for flagging my answers as “only historically relevant” (and then remove their score from whatever metric one uses to show a user contributions), but if they are still the best answer after two or more years, it means they were great answers and they do deserve all the score they have gained!

4 Likes