There is no such thing as can't in programming. It's only a matter of how expensive it would be in both programming time and to the hardware to get around the limits.
Although in this case, they could easily switch to unsigned integers to increase the limits to 4.2B. After that, they'd need to use something like the BigInteger class, or just switch to 64bit OS/code
