You should generally use plurals, they work better, and I like the form it forces you to use:
preceding texts of firsts ")" of following texts of firsts "(" of (it as string) of operating systems
I also try to keep the flow of relevance from right to left as much as possible.
In this case, doing following texts of firsts "(" of before preceding texts should be more efficient than the other way around due to it resulting in a much shorter string sooner in the parsing process.
One way to compare the relative performance of 2 relevance statements is to run them many times, like this which does both 5000 times:
Q: number of ( preceding texts of firsts ")" of following texts of firsts "(" of (it as string) of operating systems ) of integers in (1,5000)
A: 5000
T: 197.903 ms
Q: number of ( following text of first "(" of preceding text of first ")" of (operating system as string) ) of integers in (1,5000)
A: 5000
T: 203.179 ms
based upon running this many times, it seems like the performance difference in this particular case is negligible and could be due to randomness.