MVP Discussion: Reputation

Codidact Platform Ideas
61

You’re speaking about your experiences. That’s fine. But there are also other people saying, that they think that reputation is detrimental to a site or community. That’s fine too. It’s not clear-cut.

Reputation is different from other features. People are attached to it. You are attached to it. It’s not like a review queue or a nice markdown editor. If we change these, users might be sad, might be annoyed for a while. But if we change reputation, it will lead to an outcry. Because people are attached to it.

I mean, technically, what we are debating here is “removing reputation from the not-yet-build site”. And you don’t want to lose it, although you have not had the chance to earn a single reputation point here, yet. This isn’t meant as an offense or as a way to say that you are not productive/constructive/etc… It’s just a fact, because there is no site yet. And still you are attached to the idea of that number.

There are many ideas to do reputation, such as for example:

  • simply a sum of all votes
  • a weighted sum of all votes (what SE does)
  • a sum of all votes on some types of posts

Maybe some of them is best, maybe a totally different one. We know that the method Stack Exchange uses works … in some ways. We want to build a software that does certain things better. That learns from these ten years of experience.

I am not saying, that we’ll or that we should never do reputation. Quite the opposite. I am in favor of reputation. But I am also in favor of doing stuff right.

When we offer something from the beginning, people will assume, that it will be a more permanent stuff. When we label something as experiment, people will have less loss-aversion, because they know, that it’s temporary.

Reputation isn’t the only motivation-system that works. Discourse (the forum software we’re using) hosts some great communities with a lot of motivated people. Maybe that’s enough. Maybe it isn’t. But we don’t know for sure, how our communities will work. We can only guess, that they’ll be similar to current SE communities.

I promise, that we’ll do this discussion again, when we have two or three communities running for two or three months using our software. Then we’ll reach an even broader amount of people, with even more experience. We will be able to test stuff out properly (A/B tests, tests with different communities, tests with live feedback). But I would oppose doing it NOW.

4 Likes
62

That’s certainly true for me. I have 16k rep now, and I don’t really care anymore. Well, it would be nice to get those two remaining privileges. I’m especially curious about site analytics. If I ever reach 25k, then I will not care at all anymore.

When I think of it, I don’t really care for the reputation. It’s more about “leveling up”, which corresponds to getting privileges and badges. Well, especially privileges since they allow you to do more stuff on the page.

But yeah, even badges seem pretty unimportant. I have 25 edits of 500 left to do to get the gold badge “copy editor”, but I don’t really care.

1 Like
63

Really? I visit Wikipedia a lot. I don’t even see the names of the people who edited the pages there, let alone their “scores”.

Q/A count + vote tallies is what we’ve been talking about most recently. It gives people the actual information, rather than an opaque number. On SE you have no idea where that rep came from – asking, answering, hitting the HNQ lottery once, bounties, lots of editing… instead of giving people a number, I prefer to show a summary of key participation. And that’s about questions and answers, which form the centerpiece of a site.

6 Likes
64

They are not presented to the reader, in the Wikipedian community of editors however there’s an implicit and explicit system of credit. There are Wikipedia contributor pages explicitly listing contributions. Wiki Contributors; There are also dozens of studies about Wikipedia contributor motivations and the cycle of credit by amount of edits is their rep.

That also means if we don’t define a rep value from system side people will pick one, be it amount of question or answers or posts or whatever. Not defining a rep system means defaulting to the (in my opinion) worst one, counting basic stuff.

4 Likes
65

Wikipedia is really a very bad example of a rep system.

  • “Rep” (edit count) is thoroughly concealed from casual browsing behind multiple layers of obfuscation.
  • Comparing yourself with other users systematically can only be done via tedious sifting through lists of top users, with almost no breakdown by any sort of category. Contrast the vivid, easy-to-read statistics in the SE leaderboards, broken down by time period, as well as the lists by tag, and even the various reviewer stats.
  • The page that lists the top users goes to some pains to explain how wildly edit count can vary between editors of similar dedication, and makes special note of the fact that numerous very experienced editors take deliberate measures that artificially conceal many of their edits from the count. This is a clear sign that they have other priorities than optimizing for, or even passively accepting high levels of, this sort of status symbol. How many users of any sort on SE deliberately split their accounts in such a way that their rep will be substantially diluted? (No, creating a bot account and making 7 edits with it so it can flag doesn’t count.)
  • Edit count goes up immediately when you do something, rather than when someone else recognizes you doing something well, and does not go down if someone recognizes you doing something poorly.

It’s true, indeed almost axiomatic, that people in a group will seek to establish their status in some fashion. But trying to reduce the entirety of human status-seeking behavior to “rep” is unhelpful. Rep is a highly specific type of status: numeric (in fact, using linearly progressing integers), consistent, routinely visible, modified chiefly by others’ approval. There are other types of status that share few or none of these traits. (Money, for example, which is usually only visible through complex mechanisms that are themselves status systems.)

3 Likes
66

My chief worry was just that:

  • It’s fuel for “I am better than you” – https://en.wikipedia.org/wiki/Māna
  • It’s therefore fuel for sectarianism too, “the high-rep users are a bunch of arrogant meanies” and “the low-rep users are a bunch of parvenu help vampires”

I know that people are and/or have been motivated by it though, can’t have helped but notice it at least – especially that self-selecting population of people who became regular SE users.

I saw a notice on a bus once – “It’s nice to be important, but it’s more important to be nice!”

Not that I should complain about SE’s successfully gameifying the thing, but I expect it may have side-effects too.

5 Likes
67

On the rep thing… in some communities reputation (especially) Stack Overflow rep is perceived to be real and tangible - I’ve seen it on more than one resume (low 2000s) along with the number of subscribers and views for a YouTube channel.

This seems to indicate that even though that many will consider reputation to be imaginary internet points and unicorn poo - some people consider it to be a differentiator for their resume (rather than that other resume).

When taken into this light, reputation and perception that the elites are keeping people down so that they resumes look better (I’ve never seem a resume with a SO score of more than 3000) it changes the perception of the gamification and incentives for fun to things that people believe affects their livelihood (getting a job or not). Closing a question? Down voting an answer? Question ban? These actions make it that much harder for someone who perceives stack overflow rep as a differentiator between their resume and another one to stand out.

the real answer: spend a few months earning a five digit Stack Overflow reputation, and you’ll be getting job offers in the $100K+ range without an interview. – Joel Spolsky♦ Nov 21 '10 at 15:24

That this idea exists for how reputation is valued takes away from gamification for fun and leads to much darker feelings, actions and reactions to moderation and curation.

I’m certainly not saying that there doesn’t need to be a way to identify the people who are invested in the community (A Group, things to design for #2) but rather that gamification rep is not perceived to have real world value.

1 Like
68

I don’t know why it’s so but it is my impression that one gets noticeably less rep now for posting than one did when SO started – and to the extent that rep gain is noticeable that contributes to my feeling that SO is less welcoming now than originally!

69

I don’t think so. I plan it to be made clear right from the beginning that various aspects of the site are experimental, including how exactly rep is calculated. One way or another, though, providing lots of high quality content is going to get you high rep, so I’m really not that worried about the details to get a site going. If we change the formula after a few months, some users will go up a little and some down a little. I don’t see a problem with that. In the end, those consistently providing lots of good content will have high rep.

There are two independent things conflated here. We are debating whether there will be mechanics of a reputation system at all. I think it’s important to have for certain types of sites because I think it will lead to a better site. This has nothing to do with “losing” rep somehow. You underestimate my arrogance. :slight_smile: I’m confident that I’ll end up with high rep no matter how it’s calculated, as long as it reasonably rewards lots of good contributions, as judged by others.

Fine, so let’s pick something to try as early as possible. That will give us feedback as early possible. Personally, I suggest sum of all votes on answers, but having something to start with is better than nothing at all.

But how will you know what’s “right” until you try it? By that logic, you’d never roll out rep because we won’t be sure at the beginning if it’s the formula we want to keep.

I think exactly the opposite, especially if it’s made clear. “Come to our new site. It’s work in progress. Some things will change. Be a guinea pig and help us build something awesome.” This works better as early as possible. The longer into the process, the more users will expect things to be stable.

And you’ll never know without trying. That said, I think I have a pretty good idea what will work for certain types of site like I discussed earlier in this thread.

I want to be part of those two or three communities, and provide (hopefully) valuable feedback in the early formative stages. That’s how I think I can contribute here. But, I need some kind of rep to spin up the EE site I’m envisioning. Let me have rep for “my” test EE site and disable it for other sites if you’re worried about it. That solves both of our problems. I can’t do a test EE site without rep, and you’ll get feedback on how it works before applying it to other sites.

1 Like
70

This thread probably should have been resolved and closed a while ago, but since we’ve had more I’ll do it now. See Monica’s post above:

That still reflects the state of this decision: we will not have a single reputation number. We will have other numbers. We use trust levels for privileges.

3 Likes
71

So, this discussion is about those other numbers. What are they, and how do we compute them?

72

That decision was made over 3 months ago in mid November, when there were a lot fewer people here, and those people represented a narrower range of site types. Since then, there have been significant numbers of people saying that rep is important. This is something that really needs to be reconsidered.

I understand that you don’t want to be implementing to a moving target. On the other hand, I believe everyone understands that some parts of the spec will be evolving. Please correct me if I’m wrong, but it seems the implementation details for a basic rep system are minimal. Your are going to keep track of votes on answers anyway. You may already be summing those up per users. Again, I’m not a web software designer, but summing those value (if not already) and them displaying the result next to answer authors names seems relatively simple in the scheme of things.

As @luap42 points out, there is some uncertainty in this area, so experimenting as early as possible is important.

1 Like
73

I’ve had a full read through of this and other related threads. There have been a few people suggesting that rep is important, tempered by numerous other conflicting points of view. I have seen no material change in arguments since the decision was made, so we’re sticking with it.

Simple is orthogonal to whether or not we’re doing it (for which, see above).

2 Likes
74

Then what are you going to do to keep professional experts engaged?

1 Like
75

It’s the majority opinion that we don’t need rep to do that. We’ll have other engagement numbers displayed in user cards/profiles.

2 Likes
76

OK, what are those? Keeping professional experts engaged is critically important to some types of sites. We need to address that.

77

I also think rep is important, and is a big part of why I participated on SO. However, as I understand it from your recap post, the conclusion was such a point system would exist, just not be visible to anyone but the individual user, and gaining curation powers is not tied to such a number? My value on rep is not contingent on everyone else being able to see it, nor is it contingent on it being the sole/a basis for curation power promotion. The issue with trying to get input now is that we (including myself) have missed all the context surrounding your discussion. If the decision has already been made I would hope we could steer discussion in the direction of perfecting the Codidact team’s solution.

Tangentially related, but when Rust decided on it’s current async implementation after months to years of internal discussion on the discourse, the Rust reddit (not officially affiliated with the community) swarmed the Rust internal discourse forum with an onslaught of misinformed arguments and lack of awareness of previous discussion causing a lot of strife between language leaders and newcomers to the community. There were literally hundreds of new posts a day on a single thread that spilled into a dozen new topics. The problem was virtually every issue new people talked about was brought up previously, and discussed to death often within the same day.

I think if the official stance by Codidact leadership is what you’ve quoted and what is on the Wiki, then the context of the discussion (concerns, positives, negatives, rebuttals) needs to be documented so that people can focus on only new issues or solving open questions, instead of running around the same issues that were already resolved. Maybe it would look like an RFC proposal, but regardless that information needs to be collected in one place and used as a reference to avoid runaway discussion. I could see this topic come up again and again as more people join the discourse and not see the reasoning for previous decisions made. With a ledger of previous discussion, you can lock their threads and point them to this ledger.

1 Like
78

I don’t think we need rep for that, and I don’t think we need gamification for that. As I see it, experts are often willing to participate because a community where people exchange needed/helpful information is mutually beneficial.

As ArtOfCode said, I really only see a few users who are opposed to not having rep. Most people just want a good community that they can participate in and benefit from. If we can do this without rep, we should try, because as already discussed, there are downsides to having and showing reputation.

5 Likes
79

That’s not good enough for some types of communities, as I analyzed earlier in this thread. Those seem to be where the core contributors are professionals in the topic.

I think the reason you are seeing some against rep, and some say it’s important, is because of the background of what sites they have experience with. Perhaps you haven’t been on a site where public recognition was important to keep professional experts engaged, but I have.

The obvious solution is to implement a basic rep system that can be enabled per site. That gives everyone what they want.

The few reasons I’ve seen against rep have been are minor compared to the important advantage to gamification on the right sites.

1 Like
80

It shouldn’t be. Deciding whether to include a feature should be a cost/benefit decision. If the cost is low, then it doesn’t take much benefit to justify it.