Glickman的ELO算法被搬到众包中实现，可以参考这篇
Bashir, M., Anderton, J., Wu, J., Golbus, P. B., Pavlu, V., & Aslam, J. A. (2013). A Document Rating System for Preference Judgements. In Proceedings of the 36th International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 909–912). New York, NY, USA: ACM. http://doi.org/10.1145/2484028.2484170
但是文章里算法描述的部分似乎有写错的地方，可能是作者不小心，把公式(7)的分母部分写错了(原文中的字母j没有换成自己的B)~
Glickman的原文是：
Glickman, M. E. (1999). Parameter estimation in large dynamic paired comparison experiments. Journal of the Royal Statistical Society: Series C (Applied Statistics), 48(3), 377-394.
其中m为对手个数， 为与对手B进行比赛的次数； 为本局比赛A的结果（1-胜，0-负）
F是一个常量，(Bashir et al. 2013)中F=200, (Glickman, 1998)中F=400.
The rating algorithm is implemented as follows.
(a) Collect game outcome data over a rating period.
(b) At the end of the period, update players’ rating distributions due to game outcomes from their preperiod (prior) rating distributions.
(c) Subsequently update players’ rating distributions due to the passage of time.
F是一个常量，(Bashir et al. 2013)中F=200, (Glickman, 1998)中F=400.
The rating algorithm is implemented as follows.
(a) Collect game outcome data over a rating period.
(b) At the end of the period, update players’ rating distributions due to game outcomes from their preperiod (prior) rating distributions.
(c) Subsequently update players’ rating distributions due to the passage of time.