1 / 3
May 2012

I think that this is not fair, and can be abused. It's not fair because not all the people started solving problems in the same time, so if someone start solving problems in 2050 even if he does all the problems he will not be at the first place. And it can be abused if someone makes 100 fake accounts and resubmits his solutions.

My suggestion is to change that rule in this manner: instead of "number_of_people_who_have_solved p" it should be "number_of_people_who_have_solved p in faster running time then yours" In that way also you will encourage users to submit more quality solutions wink

I sincerely hope that you will consider my suggestion,
My regards

Marko

  • created

    May '12
  • last reply

    Jul '12
  • 2

    replies

  • 511

    views

  • 3

    users

The scores are updated as new users solve problems. If you'll look on some of the retired/non-active top-rankers, you'll see their score is slowly dropping. The actual problem with the scoring system is the opposite. New users have newer problems available and those are solved by less people while still being easier than some of the older ones.

The scoring system is not perfect, but it will never be. Making any radical changes at this point will flip the scoreboard around and that will put-off some of the more active users, like me! If anything is to be changes, I would postulate to start giving points for tutorial and partial tasks. The scoring schemes would be a little different, of course, but you should be rewarded with some points for ANY problem you solve.

1 month later

I also wanted to suggest the following improvement:
In challenge problems where users gets points on score length,it should be for each language so any langauage like AWK or python cannot get more points than an c++ code.All points should be distributed among each language seperately.as the problem setter has done for the problem AGSCHALL he has not allowed any such language to make a chance for c/c++ coders to gain something.I would like that all languages should be seperated while giving points.comparison should be made with the smallest code of that language only in which the user submits it not with all languages.
Thank you