Denote C is cipher text, m is message, and PRG(k) is a psuedo randomize generator, generate an key.
Because we have :
C1 = m1 xor PRG(k)
C2 = m2 xor PRG(k)
--> C1 xor C2 = m1 xor m2
So, this is a problem because if m1 and m2 is a real text (a text that is in normal life) and use ASCII, you can predict m1 and m2 if you know
m1 xor m2.
I don't understand last statement so much. Why we can predicted m1 and m2 when we just know
m1 xor m2 ? I have though some additional information, such as in normal text,
E is the most appearance character, after that is
T... but I still cannot explain.