My problem with Evolution is that I simply can't see how all this is the result of undirected mutations, and the best one wins. It would make a lot more sense if during a lifetime an animal stores some information in the DNA. If it lives in the desert, it stores perhaps some bits about heat, sun radiation, dehydration, that sort of stuff. And over time, if the descendants keep living in that area as well, and keep adding information to their DNA, slowly their descendants become more and more adapted. But just based on coincidence and selection, it doesn't make much sense for me.
What you propose is Lamark's early theory of evolution based in the inheritance of acquired characteristics, like a Giraffe stretching its neck throughout its life and having children with longer necks as a result. This is as fantastic as a creator god when you get down to it, because it relies on additional functions that simply aren't there. DNA doesn't "know" what genes do what, it can't look through your eyes and think "Oh hey, that long necked gene sure comes in handy, lets pump out a few more of them. MORE VERTEBRA FOR ALL!" It just is
Consider instead that mutation is constantly occurring and in a small and volatile population (one facing food shortages, predators, etc.) a single mutation can stand out much stronger than in a society like our own where a mutation barely makes a dent in the gene pool. If an early giraffe mutated an extra(or just longer, I'm not sure what the anatomical details of a giraffe's neck are) vertebra in its neck and the result was that his neck was that much longer he is going to have that much more success in eating food, surviving, and living to breed. If the mutation is passed on, his progeny now have it and a growing number within the Giraffe population acquire this trait. In selecting for this trait, evolution has also imbued the population with genetics more predisposed to this kind of mutation, and so this same or a similar mutation is more likely to present itself again, further increasing their long necks. As they reach the point of diminishing returns for further mutation(no longer beneficial to have an even longer neck, problems with circulation for example, or simply untenable without other mutations) The process corrects itself and the genetics of the population comparatively stabilize (at least for these particular genes).*
It is really quite intuitive.
* I am not anywhere near an expert on evolution or genetics, so if anyone can correct any mistakes I've made in this example, please do.