Embedly solution #3

gistfile1.py

from operator import add

if __name__ == "__main__":
    most_frequent = 2520
    freq_list = [most_frequent/n for n in xrange(1, 901)]
    total_w = reduce(add, freq_list)
    half = total_w/2
    cnt = 0
    for i, f in enumerate(freq_list, start = 1):
        cnt+=f
        if cnt > half:
            print "The first %d most frequent words make up half the text" %(i+1)
            break