OK, that was a lot of work for a little bit of syntax sugar, and it's still not as elegant as Haskell code can be without any effort. A big flaw is the hardcoded nature of our little zip iterator--it only works on the std::vector ranges of double variables. With a bit of template code and some type traits, the zipper can be made more generic. This way, it could combine lists and vectors, or deques and maps, even if these are specialized on completely different container item types.

The amount of work and thought needed in order to really and correctly make such classes generic is not to be underestimated. Luckily, such libraries do already exist. One popular non-STL library is the Boost zip_iterator. It is very generic and easy to use.

By the way, if you came here to see the most elegant way to do a dot product in C++, and don't really care about the concept of zip-iterators, you should have a look at std::valarray. See for yourself:

#include <iostream>
#include <valarray>

int main()
{
    std::valarray<double> a {1.0, 2.0, 3.0};
    std::valarray<double> b {4.0, 5.0, 6.0};

    std::cout << (a * b).sum() << 'n';
}