Communities

Writing
Writing
Codidact Meta
Codidact Meta
The Great Outdoors
The Great Outdoors
Photography & Video
Photography & Video
Scientific Speculation
Scientific Speculation
Cooking
Cooking
Electrical Engineering
Electrical Engineering
Judaism
Judaism
Languages & Linguistics
Languages & Linguistics
Software Development
Software Development
Mathematics
Mathematics
Christianity
Christianity
Code Golf
Code Golf
Music
Music
Physics
Physics

Dashboard
Notifications
Mark all as read
Q&A

Welcome to Codidact Meta!

Codidact Meta is the 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.

Why is there a rep system in Codidact?

+5
−3

I had visited Codidact quite some time ago when it was still being built. At that time, there weren't many talks about having the reputation system built on this site.

Partly, The reason why Slack communities and Discord communities are so easy-going and helpful has something in common between them both - the lack of any actual rep points.

You don't need a number to show your expertise - your arguments should do that. Treating everyone on an even playing field produces a much more productive debate than any other measure.

I am going to be brutally honest here - I was initially interested because Codidact seemed something new, but now it's another StackExchange in the making.

The rep system is completely useless and negatively affects the flow of debate:

  1. Your arguments should be your support in a constructive debate, not reputation

  2. Trust Levels seem to be a better way (established by upvoted answers and the like) but showing a title rather than a flashy number.

  3. Making it a rep game would lead to lower quality answers and questions as the primary aim would be points, not for spreading knowledge.

  4. People who want to answer questions (and are knowledgable) really need no 'fake internet points' as an incentive - having a trust system would work pretty well giving them extra privileges, while not signifying that they are all-knowing.

Simply put, there is no amount of reasons or arguments that can offset an actual real-life example - StackOverflow has already become what it was always destined for, and now is the last chance for Codidact.

Either you have a smaller range of numbers (1-10) to denote their moderation powers, or you take trust levels. That would be the closest simulation to Slack and Discord while working far better than both by having a formal framework.

Please don't spell death for this forum!

Why does this post require moderator attention?
You might want to add some details to your flag.
Why should this post be closed?

0 comments

5 answers

+10
−1

We didn't set out to have a reputation stat, but because we started by adapting code that had it, we started out that way. When we brought up removing it, a few people objected strongly. (See, for example, this discussion.)

We've "nerfed" reputation; unlike on SO, reputation doesn't affect your abilities. It's just a number. Abilities are instead governed by specific, related activity; for example, you earn the privilege of editing directly by having enough of your suggested edits accepted.

We've also made the "usercard", the stats that accompany your name and avatar on posts and in the users list, configurable. A community that feels strongly that rep is important can show it; a community that wants to downplay it can remove rep from that block of stats. I now realize that might be a good thing to try here on Meta; I'll discuss that with the team.

We don't yet have a way to show you which of your posts have gotten recent voting. We've kept the rep number in the header (when you're logged in) so that, as a poor substitute, you might at least notice when that number changes. We do intend to do better there; it's on our list, but we're a small team and we haven't done that yet.

Some say that the reputation number is a measure of expertise. Maybe an average rep per post would be, but the total alone doesn't tell you whether the person has a few good posts, a bunch of mediocre ones, or a single hit. On The Workplace over on SE, we often saw a snarky answer on a hot network question skyrocket for the entertainment value. That someone has a couple thousand rep from that single answer doesn't actually indicate any expertise -- but, if people are evaluating posts based on author rep, that person's next answer would look stronger than it should. This is why, in the "usercard", we show the number of posts (by broad type). One of my early designs of that blob of stats showed more of a breakdown, something like "41 answers (39 positive)", but it was cluttered and we decided we needed to look for another approach.

Why does this post require moderator attention?
You might want to add some details to your flag.

0 comments

+4
−3

Actually Codidact should support rep better than it does now. Most of the dislike of rep seems to come from a misunderstanding of its purpose. It's not to show who is "better" somehow, but a measure of how much someone has contributed useful content to the site as judged by all the other users.

Your arguments should be your support in a constructive debate, not reputation

Right. Rep has nothing to do with validity of arguments. However, it might give a sense of trustworthiness when other direct metrics aren't available. In that sense it works like reputation in the real world.

Let's say you like reading horror stories, and there are a bunch of book in front of you to pick one from next. You've never heard of any of the authors, then you notice one by Steven King. You've read and liked his works before. You pick that one because his reputation suggests that you are more likely to enjoy that book than a randomly chosen one.

Trust Levels seem to be a better way (established by upvoted answers and the like) but showing a title rather than a flashy number.

But "established by upvoted answers and the like" is this reputation you don't like. Even if you don't want to show it, the system still needs to calculate and track it in this scheme. The only difference is that you want to show one result of that rep score instead of the score itself, but haven't explained how that is supposed to help.

Making it a repo game would lead to lower quality answers and questions as the primary aim would be points, not for spreading knowledge.

Wrong. Gamification means people will want to do those things that they get rewards for. If you've chosen the incentives correctly, then they'll be doing exactly what you want them to do, which is mostly providing high quality answers. Remember, the quality of these answers will be judged by everyone else, so they have to be good to get those points you don't like. Giving out free internet points with some recognition seems like a pretty easy way to incentivize people to contribute good content to a site.

People who want to answer questions (and are knowledgable) really need no 'fake internet points' as an incentive

This is hopelessly idealistic and naive at best. People do things for a reason. There is no such thing as pure altruism. Some people may feel internally rewarded by helping others, but there is always some reward or reason. Few people are going to spend the significant free time you need from them to provide good content with nothing in return. Giving them a little recognition for all the volunteer help is doesn't cost you anything.

What you are basically advocating is that volunteers should come here to contribute their time and expertise for free, and not even be thanked or recognized for doing so.

having a trust system would work pretty well giving them extra privileges

Privileges are different from public recognition. Privileges might be an incentive for some, but isn't going to be as powerful as public recognition for many.

I know from personal experience on EE.SE (I was #1 with about 280 k rep when I left there in 2018), privileges weren't much of a driver. I found some privileges useful, but didn't use others much at all. When I got a new privilege it was more of a surprise, and wasn't what I was aiming for. My thought was more "that's cool, maybe that will be useful some day".

Rep, on the other hand, made things competitive. We couldn't have Andy or Spehro with higher rep than me. I mean, geesh, that would tear up the fabric of space-time and end the universe as we know it. Can't let that happen.

while not signifying that they are all-knowing.

Right, and rep doesn't do that. It's a measure of appreciated and useful contributions.

Why does this post require moderator attention?
You might want to add some details to your flag.

4 comments

I have a bunch of rep on Writing from technical-writing questions. People shouldn't assume I have special standing if I answer a question about poetry. In fact, if I thought people would judge all my answers based on my rep, that might push me to not answer other questions (that I can also answer, but maybe not as well), for fear of misleading. We've wanted to show some form of tag-based background for a while; just haven't worked out how yet. (Show my expertise on TW Qs, not poetry Qs.) Monica Cellio‭ 22 days ago

@Monica: If you screw up on a question somewhat outside your domain, others will downvote it fast enough. You can also start with "This isn't my area of expertise, but ...". People seem to be more forgiving of errors that way too. Olin Lathrop‭ 22 days ago

One point that's missing here is a way to measure domain expertise. Rep was always a poor way to measure it, but it's better than nothing. At best rep might reflect your overall broad knowledge of the topics that the community is about. What worked best at Someplace Else was badges. You got bronze, silver and gold badges depending on the number of votes of your posts below a certain tag. These are much harder to gain than rep, and it was actually the best way to measure domain knowledge. Lundin‭ 21 days ago

And domain knowledge matters for more than "bragging rights" - quite often you need extensive domain knowledge to moderate certain posts, to judge if they are on-topic, duplicates etc. Codidact will get there too, where specialized knowledge is needed to moderate certain topics. Lundin‭ 21 days ago

+0
−0

What is repo/reputation?

Reputation is just number. Any user can know how active you are in Codidact by seeing your repo.

Why reputation is needed?

As @Monica Cellio answered is a next-generation academic platform for physicists and astronomers, in#answer-281378) repo doesn't affect your ability. While it does in SO. Actually I was very active user in SO and Unix SE. Sometimes I ask question right there not every time I got answers. Sometimes lot of user answers "simultaneously" we get in trouble that which answer I should try(It happens only when user is completely new to that technology or knowledge). So, obviously I will choose a professional's answer. By seeing the reputation we can know who is professional. That's why reputation is needed.


Look reputation isn't useless. Sometimes lot of user asks poor quality question. No one wants to review on a poor quality post. And, I personally like the reputation system of Codidact rather than SE (SE has the feature for answer not for question) cause, when we get downvote our reputation decreases but, if we remove that post than we get our reputations again while it doesn't have in SE. While user asks a poor quality question/answer we mark that as negative. There's another function I like there's differences of downvotes and upvotes. That's why we can know how helpful the post is by seeing the votes. You may say that your question was about reputation not upvotes and downvotes. But, your reputation decreases or increases depending on quality(upvotes and downvotes) of post that's why I added information of it.

Not only Codidact or SE has the reputation system most of Q&A sites have the features. I am adding some Q&A sites.

  1. Kali Forums
  2. Linux questions
  3. PhysicsOverflow
  4. Physicsforums
  5. Manjaro Forums

There's lot more. I didn't add Quora and, Reddit cause, people uses them as Social Media (I was a Quora user also).

Why does this post require moderator attention?
You might want to add some details to your flag.

0 comments

+1
−1

Part of this question is about making Codidact more Like Slack and Discord, which are not even question and answer sites. The two of those have their own share of problems, and the model that was used for them simply does not fit the requirements of a Q&A site. Social media and chat apps do not encourage quality or contribution toward the site. It encourages sitting at the same screen all day, waiting for some form of feedback from the site. There are many, many things which distinguish the two from each other.

Reputation is a system which(partly) goes against that kind of problem. You will be discouraged for spamming and posting bad content. You will be downvoted for posting poorly supported answers. Although it isn't perfect, it is at the very least, designed for a Q&A site.

Why does this post require moderator attention?
You might want to add some details to your flag.

0 comments

+1
−2

I agree with the notion that reputation numbers cause more harm than good, at least in general and at least in the long run.

I was not convinced by any argument I have read so far to why its "sorely needed" here in Codidact.

People ask questions and give answers about fields of knowledge that they have want to acquire knowledge in / be grasped as understanding in, respectively; so if the main motivator for such operations was points, then Wikipedia articles and community websites (say, MediaWiki support desk, Drupal forums, etc.), weren't so rich as they are with data given by volunteers.

Why does this post require moderator attention?
You might want to add some details to your flag.

0 comments

Sign up to answer this question »