Many badges have fixed requirements, such as:

  • Answer score of 100 or more

However different websites means a different level of difficulty: the more active users a website has, the easier the badge is to get.

On a new and/or low-view website getting this badge may be impossible for several months; on a website like stack overflow now you can obtain it in a day, and it's relatively easy (compared to other SE sites).

Do you think it would be a good idea to implement a formula that allows the required score to scale? The more active users voting, the more votes are required to get it?

If yes how would you handle people who got it for a post that no longer meets the requirements?

  • 4
    Surely this is a self limiting problem. As the site gets more users there'll be more votes and it will become "easier" for an answer to get 10, 25 or 100 votes. Old answers (if they are good) do gain votes (albeit at a slow rate) so you might get that Great Answer badge eventually.
    – ChrisF Mod
    Commented Nov 8, 2010 at 12:34

2 Answers 2


While I understand your intention, I am not sure if the idea itself is good. The basic question here is: How do you want badges to increase as users increase? Do you want to limit it by a percentage or an absolute threshold?

An example for your idea might go like this: Imagine you got 1k users. This means it's very hard to get 100 upvotes. Now imagine you got 1000k users. You're saying now it's easy.

The thing not considered here, however, is that with 1000k users, there are also 1000 times as many items to rate - so on average, each only gets a 1000th fraction on average. That's why scaling the value up in a linear way (e.g. 10 votes for 1k users => 10k votes for 1000k users) would be bad.

Since you are certainly right that it is easier to get 100 upvotes on SO than it is on a fresh SE site, I believe that if scaling is introduced it would have to be logarithmic. "50 votes on a new site is like 100 votes on SO"

After all, I believe your idea is reasonable and I do vote for it to be implemented. Just make sure the scaling is done in a way that is easy to understand and does not have unwanted side effect (i.e. don't overdo it).

  • Maybe by using a value that's calculated from the number of users in the site, but the problem is, for new sites, the equivalent to SO's 100 upvotes could be 2 upvotes (in the sense of x% of the users upvoting). Commented Jan 16, 2012 at 22:01

I don't like it and this is why:

Image that a scaling is implemented. On day X, on Foo Overflow, the good answer badge requires (say) 15 votes. Fred Q. Programmer has 4 such answers.

On day X+1, the scaling algorithm kicks in and the threshold changes to 20 votes, and only one of Fred's answers still qualifies.

Then he posts a fabulous answer to new question and breaks his rep cap with that answer alone. Of course, he has a three Good Answer deficit, so his badge count doesn't change...

But did he not do four Good Answers worth of good for the Foo Overflow community when it was small? Why should he be punished now?

I'd just a soon see those badge be rare and difficult during the early epochs of each site, rather than introduce this odd deflation of work from the early ages.

Yes, in theory more users joining up means more votes, and that probably works for the questions on the fist few pages of the "Sort by votes" list. But the pure number of duplicates of really good fundamental questions shows that they are not getting the kind of repeat business that they deserve. I don't think it is fair to scale badges away until the "users don't search" problem is fixed (which, I'm not holding my breath for...).

  • Where can I sign up for Foo Overflow? I have a question to ask... Commented Jul 21, 2013 at 0:39

You must log in to answer this question.

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