What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 131 Stack Exchange communities.

Once in a while I stumble on some question or answer with HUGE amount of upvotes (thousands for a question and multiple hundreds for an answer).

Here is the perfect example:

How do I edit an incorrect commit message in Git?

What I found that most of such questions/answers are:

  • Quite old (dating back to 2008, 2009, sometimes to 2010)
  • Very often these super high voted answers could be googled in 10 seconds in some other resources (meaning that both the question and the answer aren't THAT valuable).

Another thing which I found that there are users (Old timer's) who answered literally one such question 5 years ago. And these users can get more reputation each week doing nothing than a newcomer (like me) answering couple of questions each several days.

So, I decided to investigate a little bit and wrote a query which calculates a ratio of Total reputation / number of posts (Answers + Questions). Here is it:

http://data.stackexchange.com/stackoverflow/revision/96666/115919/old-timers-luck

It's interesting to see how people got into 5k+ with 1 contribution to SO and there are even people with 10k+ (admin rights) with just 10 questions/answers.

Generally speaking, I have nothing against old reputation. However, a person doing nothing for 5 years and still beating me in daily reputation - kind of pisses me off :)

I just want to hear your opinion on this thing. Especially, it would be interesting to hear from other members who become active recently (I understand that for an admin with 30k reputation and thousands of answered questions it's not that big deal that somebody got 7k reputation from one answer).

share|improve this question
9  
This is why rep caps should be per post and not per day :P –  minitech Feb 10 '13 at 0:36
    
@minitech: I think so :) –  Victor Ronin Feb 10 '13 at 0:55
    
It's been proposed a bunch of times (including by Jon Skeet), but it's been shot down every single time. And no I'm obviously not in support of the idea either. –  Mysticial Feb 10 '13 at 0:57
    
@Mystical: BTW. Why was it shot down? –  Victor Ronin Feb 10 '13 at 0:58
2  
It's too drastic of a change and it hurts too many users. (I would be hit particularly hard by a per-post cap.) My personal reasons against it is that it takes away all incentive to post quality answers. –  Mysticial Feb 10 '13 at 1:05
2  
@Mystical: Sure, if cap is 200 then you are right. However, if cap is 2000, it's more then enough to incentivize to write good answers. –  Victor Ronin Feb 10 '13 at 1:09
    
To be fair, a 2k per post repcap isn't going to hurt me any less than 200 - just saying. –  Mysticial Feb 10 '13 at 1:31
1  
+1: This happens all over around SE. Coz - at the start, all guys vote like they are pissed off. It can be easily noticed in meta sites. The top voted questions in all sites can be easily googled for sure. I'm quite happy that someone asked this question. Thanks Vic. One more thing: You should've also noticed that new answers will get very less votes relative to the first responses, though they can be encouraged :-( –  Waffle's Crazy Peanut Feb 10 '13 at 3:37
    
@Mysticial: Do you mean that a per-post rep cap would hurt you if implemented to affect previous reputation, or that it would hurt you either way? (And taking 200/post as an example — a 20-upvote answer is not really bad. It's not as if the current rep cap is any different — you can have one good answer and get a maximum of 215 reputation, or you can have more and be more or less unlimited. With a per-post rep cap, more good answers would mean far more reputation…) –  minitech Feb 10 '13 at 15:45
    
@minitech A per-post persistent repcap will definitely hurt me much more than the current repcap. The current repcap puts a lid on viral spikes - which is just a temporary affect. But a per-post repcap will stop all of it after a while. Taking my 3rd best answer as a typical example: It has 798 votes. My rep audit page shows about 340 uncapped upvotes. So that's already 3400 rep. –  Mysticial Feb 10 '13 at 18:05
    
Basically I'm saying that while a per-post cap will stop one-liner answers from inflating rep, it has too much collateral damage. The system already favors quantity over quality with uncapped accepts. I don't think we need any more of that. –  Mysticial Feb 10 '13 at 18:06
    
You are also just pointing out the largest cases of this. A micro-version of this is happening every day. For example, my highest upvoted answer (answered Oct `11) was probably also the answer that I spent less time on than any other answer. Yet it has been helpful to the most people, so it's netted me the most rep. I think that's true all over the site where older and simpler posts across the board get more rep than later and more complex posts. –  Ben Lee Feb 14 '13 at 22:50

6 Answers 6

It's really no different than stock options in a startup company.

  • In a startup, the earlier you are there, the bigger portion of the company shares you get;
  • On Stack Overflow, the earlier you were here to snatch up all the common questions, the more passive rep you get in the long run.

Yes there's an effort imbalance. It was never entirely fair in real-life. And it definitely isn't entirely fair on Stack Overflow.

You are right that old users have a huge advantage in this rep-game. You basically learn to live with it. Try not to sweat too much about that number.


That said, the game isn't over for us.

Programming and (computers in general) is a fast changing field. Things get outdated very quickly and new things come in. If you stick around long enough, you may get lucky enough to snatch the next "common question" for the next technology.

Then you will be able to sit back and enjoy passive rep just like many of the old-timers before us.

Newcomers like us also have one advantage over the old-timers: we have much less to fear about the deletion of many highly-voted posts now that the rules have more or less stabilized.

share|improve this answer
4  
In startup, you take more risk to get bigger stake. However, in described situation they didn't have any risk (they invested 10 minutes in 1 question) and got quite nice stake. –  Victor Ronin Feb 9 '13 at 20:43
3  
People win the lottery every day just for buying the right ticket :-) –  Duncan Feb 9 '13 at 20:43
2  
2Duncan: But, do we want a StackOverflow be a lottery? –  Victor Ronin Feb 9 '13 at 20:44
1  
@VictorRonin Oh there was plenty of risk. There was always a possibility that SO wouldn't succeed. (Of course we know now that it isn't the case and SO is possibly the most successful programming Q/A in the world.) –  Mysticial Feb 9 '13 at 20:45
    
@Mystical: I agree, it applies to a person who replied to 100 questions, but it doesn't apply to a person who answered just 1 question (all he risked is 5 minutes). –  Victor Ronin Feb 9 '13 at 20:47
    
@Victor true, but you'll notice that all the people on page 1 are currently active on Stack Overflow. They are still contributing every day. And many have joined late in the game. –  Pëkka Feb 9 '13 at 20:54
    
@VictorRonin If you look at that "one-time-old-timer's" answer, there would be other answers with the nearly same content yet have far less upvotes because they posted 1 minute later or so. So even some old timers would be feeling out there they "lost" due to their slow INTERNET on that day or a slow typing or they had to "prove" they are a robot etc :) But if you think about "what's the solution?", there's really nothing as limiting upvotes would be a bad idea -- because votes indicate how many people find it useful. So just get used to it like many others ;) –  user210447 Feb 9 '13 at 21:20
    
@VictorRonin I am a relatively new user (11 months old today :D) as well and felt the same at times (if that makes you feel any better :D). When there's no reasonable practical solution for this issue, so the only way forward is to get used to it :) –  user210447 Feb 9 '13 at 22:03
    
@KingsIndian: I am not sure that there are no practical solutions. I would say there is no solution without downsides. As example, there could be limit to upvotes, there could be a decay for older question (offered by somebody) or wiki-fication of questions. –  Victor Ronin Feb 9 '13 at 22:08
    
2KingsIndian: BTW. Thanks for acknowledging that it's not only my issue. :) –  Victor Ronin Feb 9 '13 at 22:15
1  
@VictorRonin The "issue" here is what people want votes to indicate vs. what they actually indicate. A simple but common question may not deserve a lot of votes to many people. But it is certainly helpful to all the upvoters who came from Google looking for an answer to the same question. –  Mysticial Feb 9 '13 at 22:26
    
@Mystical: I agree. That's a core of this (and several other reputation related problems). –  Victor Ronin Feb 9 '13 at 22:30
    
@VictorRonin Frankly, I'm not too bothered by this type of "rep-inflation" caused by common basic questions. First of all, it's rare. There may be dozens of them, but SO is a very large community. So the over percentage is still very small. Once you have long passed 20k (by whatever means) it becomes a lot less interesting (for most people at least). Furthermore, it can look silly at times when someone comes to your profile and sees a one-liner with 1000 votes as your top answer. If anything, people may hate you for it. –  Mysticial Feb 9 '13 at 22:45
    
@Mystical: Probably it's the best argument from all this question. As soon as number of such lottery winners is less than some small percent, it's ok. –  Victor Ronin Feb 9 '13 at 23:50
    
@VictorRonin I count about 2 dozen of these "basic and common" questions with over 1000 votes. That's about 75% of all non-wiki answers with over 1000 votes. But compared to the entire population of SO users - I'd call that pretty insignificant. –  Mysticial Feb 9 '13 at 23:54

I don't think you need to worry. If somebody gains moderator powers despite answering only a couple of ancient questions, they are hardly likely to be a major influence on the site. I think it's telling that the top of your list hasn't been seen since Feb 2010.

Anyway, StackExchange lets you measure your reputation per quarter. Perhaps that's a number you can take more pride in? That certainly can't be "cheated".

share|improve this answer
1  
Thanks. Frankly, I am not worried about moderator powers. Mainly, I don't like that their old reputation is so big. –  Victor Ronin Feb 9 '13 at 20:45
    
Oh... well that was the only downside I could really think of. –  Duncan Feb 9 '13 at 20:47

Very often these super high voted answers could be googled in 10 seconds in some other resources (meaning that both the question and the answer aren't THAT valuable).

Stack Overflow is a beast in Google results, that's probably how the questions/answers got so many upvotes - because they turned up after a search and were useful. Sure they may be basic, but it doesn't mean they aren't valuable.

Generally speaking, I have nothing against old reputation. However, a person doing nothing for 5 years and still beating me in daily reputation - kind of pisses me off :)

Keep contributing, and you'll one day become an "old-timer" yourself and be gaining rep every day for doing nothing.

Don't think of Stack Overflow as a competition, think of it as a collaboration.

share|improve this answer
5  
Thanks. I think it's both collaboration and competition. And I am more than happy with collaboration part of SO. –  Victor Ronin Feb 9 '13 at 20:49
    
There is a luck factor involved but it's not limited to old questions. Anything on-topic that is simple and interesting can reward tons of rep. For example: stackoverflow.com/questions/7073484/… Really, just keep contributing good content and don't be afraid to post solid answers on older questions either. –  Wesley Murch Feb 9 '13 at 20:58
    
BTW. Look at the discussion under Mystical answer. There are a lot of comments, which are definitely about competition, rather than collaboration. –  Victor Ronin Feb 9 '13 at 22:34
    
I think Jeff Atwood would agree that the competition aspect was added, to capitalize on human nature, and provide a way to increase collaboration (which is what a free public Q+A site is). The goal should be collaboration, but competition I think is a healthy way to achieve that. –  Nate Mar 10 '13 at 21:08

I think a decay function would be good. Not a fast one, perhaps an exponential decay one like e^(-x/5), which yields observable decays in a contributor's rep after 1.5-2.0 years or so.

This way, old, experienced contributors would be stimulated to keep on contributing to answers and questions as well, not just moderating.

Unfortunately, it may be well the case that the users who are in the position of deciding to do this are also the ones which would be negatively influenced by such a decision, so they won't want to lose their advantage over the others.

Oh well, that's how it works in real life too.

On the other hand, such a mechanism would encourage new users too, who would be less intimidated by such high reps.

e^(-x/5) is not that much, actually. A question/answer would decrease by approx. 0.8 in one year, and it could only be applied to questions or answers over a threshold, not to every contribution.

share|improve this answer
1  
I suspect most experienced contributors are still busy answering questions, not just wagging their big moderator sticks. –  Duncan Feb 9 '13 at 20:50
    
@Bart I've edited my answer. –  Flavius A Feb 9 '13 at 20:51
    
@Flavius: I don't think it should decay (it's too rough), but I believe some super active questions should be wiki-fied. I believe it's done already, but there are plenty of questions which are still bringing such type of disbalance. –  Victor Ronin Feb 9 '13 at 20:53
1  
@DuncanJones Sure, I suspect that too. But such a mechanism would encourage that behavior. –  Flavius A Feb 9 '13 at 20:54
1  
The rep decay might also help with the problem of old, massively-upvoted but obsolete, answers swamping out new, correct answers. –  Awesome Poodles Feb 9 '13 at 21:52
    
The users are not in the position to decide anything. Also can you point to a top 50 user who is "old money", living off his old rep but no longer participating? I predict you won't be able to - A sort of decay is already in place: your fellow users accumulate more rep every day, and not keeping on accumulating rep means falling behind. –  Pëkka Feb 10 '13 at 0:47
    
@Pekka웃 Well, there's two actually. Alex Martelli and Pascal Thivent. But I'd call that insignificant. –  Mysticial Feb 10 '13 at 0:50

What counts are not the points but the road that you walk to reach them.

I would like to tell you that I have seen some users here that started after me and beat me, passed me and they are higher ranked than me, and they deserve it. They have given more help, more answers, and they worked a lot to help the community.

Second, the reputation is one metric, but there are more, like the many answers you have given, like the tags, the badges, the present on top of some language that you may be, and maybe you are in top 20 of some kind of language.

And now the question, what you do after your reputation reaches 30k with only ten questions?

I can tell you what you do before it reaches 30k with 2000 answers: you help people by giving answers, learning from other questions and improving your software too. Be part of a community that allows you to test your skills and your know how.

This is the road that I am talking about, this road that improves you as programmer, is not the numbers, is not the reputation, is not if you new or old.

share|improve this answer
    
First of all, thanks a lot for good and calm answer. I agree there are multiple benefits of SO: helping people, self improvement. However, one can't disregard competitive nature of SO. And mainly the question was about fairness of competition on SO (vs benefits of SO). –  Victor Ronin Feb 9 '13 at 23:31
    
@VictorRonin Well, I do not know about fair, but there is code and knowledge of one line that worth more than my 1000 answers, meaning that there are many king of knowledge, some worth more than other. And some they get the jackpot on the lottery and be millionairess, is that fair ? (similar to the jackpot on an answer that bring many votes) –  Aristos Feb 9 '13 at 23:40
    
@VictorRonin Do you ever see the serial two and a half men ? –  Aristos Feb 9 '13 at 23:46
    
I wrote somewhere else in comments. Do we want SO to be a lottery? :) I understand there is no way to "remove" influence of luck completely, but it would be nice to reduce a little bit lottery side of SO. –  Victor Ronin Feb 9 '13 at 23:47
    
Yes. I saw it and I like it :) –  Victor Ronin Feb 9 '13 at 23:47
    
@VictorRonin Maybe is not the best example, but I like to point you that the one there have "the points" the other have the house. The one is promote "the points" that have, the other promote the house. If you go to work somewhere is better to promote what you have done and not the point and the paper that you have earn. Both they are good, but what you have done is better. –  Aristos Feb 9 '13 at 23:57
    
@VictorRonin Some part can be lottery, why not ? Lottery is for few, is part of the game, let it be, is not for all, and this few let them have some more vote - and you know, they say that we make our luck. :) but to tell you the truth I believe that is not lottery and they worth it (the many votes). –  Aristos Feb 9 '13 at 23:59
    
@VictorRonin Look, is not lottery, they give one answer that help 100k people, is similar like to give 100 answers to help peoples that totally see that 100 answer near 100k times. As I say you, there is knowledge that worth more and they diverse the more votes. But you are not here for that. You here to be better. –  Aristos Feb 10 '13 at 0:04

My two most upvoted answers both took about five minutes to write.

Many answers that I spent ages on didn't get any rep at all - no upvotes, no acceptance. However, while the rep from the simple answers is great, I've gained a lot in thinking about how to answer the question.

In other words, if all you're getting out of StackOverflow is rep points, it's a waste of time - they're worthless. There's much more value to be had here.

share|improve this answer
    
2Blowski: Sure. Reputation isn't only thing which is important on SO. However, it's definitely one of key metric for competitiveness. Please read our discussion with Aristos. SO is about helping people and learning yourself, but it's also about competition. –  Victor Ronin Feb 10 '13 at 2:12

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .