the length of the snippet is somewhat arbitrary (determined by Wikipedia's search engine)
many snippets in foreign languages' (such as Chinese's) Wikipedia contain raw wiki formatting characters (Wikipedia's search engine returns bad results)
the pronunciations of names takes up a lot of space
probably need a new way to acquire text snippet to improve the situation.
probably need a new way to acquire text snippet to improve the situation.