4

I wonder how the "next badge" options for tag badges are sorted. In the screen shot below, the tag is recommended, and shown first, even though I'm closer to getting the badge:

enter image description here

I need write another 3 x 9 answers for , expecting the same number of votes, to get that badge, whereas I need to write almost 4 x 12 answers for , again expecting the same number of votes, to get the that badge.

This means: I need 3x what I have for , but almost 4x what I have for . I am much closer to the badge.


A correct algorithm to see how far along you are for each tag would be:

min(p_score, p_answers)

This accounts for the fact that both bars need to be filled in, the slower one to be filled in is the one that determines how long it will take to get the badge.

I suspect that what currently is used is:

( p_score + p_answers ) / 2

This latter leads to the numbers:

( 21/100 + 12/20) / 2 = 0.4050
( 25/100 +  9/20) / 2 = 0.3500
( 17/100 +  9/20) / 2 = 0.3100
( 21/100 +  7/20) / 2 = 0.2800
(104/400 + 23/80) / 2 = 0.2737
( 22/100 +  4/20) / 2 = 0.2100

which is consistent with the order they are shown in. Using the min method you get a different view:

min( 21/100, 12/20) = 0.2100
min( 25/100,  9/20) = 0.2500
min( 17/100,  9/20) = 0.1700
min( 21/100,  7/20) = 0.2100
min(104/400, 23/80) = 0.2600
min( 22/100,  4/20) = 0.2000

Now we see that actually I'm likely to first get the silver C++ badge, then the bronze algorithm badge, and I'm as far ahead in the C++11 tag as in the MATLAB tag.

1

As a percentage of the total requirements you appear, to me, to be closer to getting the matlab badge. The way I read that you've got 21% of the score and 60% of the necessary answers (40.5% of total) versus 25% of the score but only 45% of the necessary answers (35% of total) for the algorithm badge.

Answer predates OP's own mathematical annotations.

  • You would be right if the requirement was "#of votes" OR "#of answers", but it is AND. I need to fill both bars. The top bar is furthest from being filled in both, and therefore the votes is the limiting factor. I will need to write way more than the minimum number of answers for both tags to get these badges. – Cris Luengo Aug 22 '18 at 18:51
  • 1
    @CrisLuengo Not really, well maybe on the particular site in question, good answers on a lot of stacks can pull down a dozen or more votes which should make score the "easy" part of making the grade. – Ash Aug 22 '18 at 19:00
  • On both the sites I frequent (Stack Overflow and Code Review, the screen shot is from the latter), this behaviour depends very much on the tag. For the MATLAB tag on both sites, the average number of votes per answer is (way) below 5, meaning that it is the number of votes that is the limiting factor. For other tags you might get more than 5 votes on average per answer, in which case the number of answers is the determining factor. – Cris Luengo Aug 22 '18 at 19:05
  • 1
    @CrisLuengo Ah in that case it would appear you are the victim of a "one size fits all" that doesn't. – Ash Aug 22 '18 at 19:08
  • 1
    On SO, the top 14 tags in the "I'm almost there" category all have the number of answers bar full, I'm waiting to get enough votes. – Cris Luengo Aug 22 '18 at 19:09
  • @CrisLuengo Interesting, I just had a closer look at where I'm at on a couple of sites, there's some niche tags where the score is done (and I have less than half the questions I need) but I have a lot of tags that I'm in the same position with too, very little in the way of equal progression. A lot of it will be, as you say, the popularity/specialty of those tags. – Ash Aug 22 '18 at 19:20

You must log in to answer this question.

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