10

My Question:

  1. How is the UX of the best, or any, web reputation systems capturing and representing offline influence within a community?
  2. What are the signals that are quantified or illuminated in the examples?

Background:
I'm working on a UX that will include on a reputation system for our user base and am taking notes from popular systems such as:

  • Stack Exchange, Yahoo Answers, Ebay, Amazon, Yelp
  • Klout, LinkedIn
  • Dribbble, Behance
  • Xbox Live Gamerscore, Apple Game Center, Halo Rankings, League of Legends

Most of the reputation system examples focus on system activity to build up rank and do not capture real-life, offline influence. For example, my 134 score & 4 badges on UX-StackExchange is great at showing I haven't done much on StackExchange (by comparison to those using it), but not so great at capturing what I get paid for daily (and all the other people not using it).

Examples like these will not be sufficient for the reputation scoring I need in our UX. Quora approaches this problem differently, leveraging real identity and offline reputation to lend credibility to the answer. There's no score, just a headline that leaves the user to interpret the credibility. The fact is that all new users of our system should not start with a reputation score of zero (if numerical score is even the right display).

I'm looking for any UX examples (good or bad) of systems or signals that are capable of capturing offline influence. One such UX example is what LinkedIn has started doing recently with Skills & Expertise - your connections can upvote skills you've listed, or even suggest you to add new skills to your profile. I don't need a whole site breakdown, just the leads on where to look, and thoughts on why you're suggesting it.

Improve this question
9
  • 1
    Is there an actual question with a single best answer here, or are you just asking us to do your homework for you and give you our opinions?
    – Raye Keslensky
    Dec 19, 2012 at 1:00
  • 1
    ouch. well i know it was a long question but "I don't need a whole site breakdown, just the leads on where to look, and thoughts on why you're suggesting it." I don't expect a best, but do you know of ANY that do this?
    – Shash
    Dec 19, 2012 at 2:04
  • 6
    @RachelKeslensky I think this question has merit - it's clear from the background that OP has researched the topic and is coming to us for more depth.
    – Rahul
    Dec 19, 2012 at 2:52
  • 1
    For a broad perspective, perhaps look at the book "Building Web Reputation Systems" by Farmer and Glass. The principal author seems to have been a pioneer in the field. It isn't dedicated to UX though, but may be worthwhile in terms of designing the overall mechanics of a reputation system.
    – CJF
    Dec 19, 2012 at 8:26
  • 1
    I added some perspective from the academic community as an answer below. I doesn't address the UX, but may provide some insight on how the "offline" academic world handles reputation.
    – CJF
    Dec 19, 2012 at 8:45

3 Answers 3

Reset to default
1

I can't comment on the UX side of your question, but in terms of using numbers to represent reputation/influence I can offer some personal experience. In academic circles there are different measures of "influence". One of these measures is the "H-index" (copy/paste from Wikipedia):

The h-index is an index that attempts to measure both the productivity and impact of the published work of a scientist or scholar. The index is based on the set of the scientist's most cited papers and the number of citations that they have received in other publications. The index can also be applied to the productivity and impact of a group of scientists, such as a department or university or country, as well as a scholarly journal.

Journals have also traditionally received "impact factor" ratings that reflect their influence/credibility. Journals that publish articles that are cited a lot, have a higher impact. Scholars who write articles that are cited a lot, have a higher impact. Usually many citations imply "positive impact", but this is not necessarily the case. The use of these metrics are also highly contentious among academics (these are often connected to future job growth and funding opportunities).

So in the offline academic community, they also use numbers to recognise contribution. If your on-line system wants to make use of academic "experts", then maybe you could attempt to piggy-back on one of these existing influence metrics. I know Thomson Reuters keeps record of the journal impact factors, but there used to be tools available to compute the h-index for an individual scholar.

Improve this answer
2
  • thats a great perspective that I don't have much insight to! thanks for the response. "The index works properly only for comparing scientists working in the same field; citation conventions differ widely among different fields." I was never stellar at math, but you'd have to 'normalize' the numbers across the scientific community to arrive at a score that could be universally displayed?
    – Shash
    Dec 19, 2012 at 18:32
  • 1
    Technically, yes, you may have to normalize the scores between fields. This may be a philosophical issue, but I can't help but wonder "what does the number really mean in the end?". The H-index isn't a linear score. People may have trouble interpreting it. I almost want to compare it to "statistical significant difference" vs "practical difference". Sure, user A has a 10-point difference to user B, but what does that mean? Is user A less of an authority? I suppose that is the trouble with quantifying any human ability. There should be a point where someone is simply "good enough" vs "better".
    – CJF
    Dec 20, 2012 at 18:35
5

You're missing LinkedIn. Here's what my profile looks like:

enter image description here

You've probably seen it already - when you log in to LinkedIn, they ask you "Is this person good at X?" and you just click to say "yes" or skip to someone else.

This works pretty well in practice, although it can obviously be "gamed" by filling in something in your profile and asking your mates to vote you up for that particular parameter. But in reality, no one who matters is going to do that because that stuff is visible in your network and people would be weirded out if I suddenly showed high number of votes for "brain surgery" or something.

This system is also better than the previous recommendations system because it requires much less effort by all involved parties. I just fill out my profile like normal, listing skills I have or use, and other people let the system know whether they agree. That's it. As a result, you have a pretty good set of real world skills attributed to people on a social network that is all about connecting people based on those skills (and other things).

Improve this answer
5
  • 1
    I added it to the bulleted list but I already included a description of that exact element in the background. That is the only system I know of which attempts tho, thanks!
    – Shash
    Dec 19, 2012 at 2:48
  • Oops. It's late, sorry about that.
    – Rahul
    Dec 19, 2012 at 2:50
  • I find those skills recommendations on LinkedIn useless. They suffer from the same problem that the OP mentions. The number is heavily dependent on how active you and your friends are on LinkedIn.
    – yarian
    Dec 19, 2012 at 19:41
  • @yarian, i'd go even further to say that unless you know and can judge the reputation of the person who upvoted the skill, or at least could see that person's score for the same skill, it is not meaningful to the profile viewer. the numerical upvote score would have to reflect something more like h-index
    – Shash
    Dec 19, 2012 at 21:39
  • @Shash Agreed. Also, there are very few people that you can honestly endorse for most things. Besides people you have worked with closely, it's hard to gauge anyone else's skill at a language/framework/os.
    – yarian
    Dec 20, 2012 at 0:05
1

I just thought of one. Call of Duty MW3 displays those badges for your ranks in the previous games so you can tell if somebody with a low rank is just new to this game. But still overall, the reputation system scores usage.

Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.