DNA - Blueprint or recipe?

While explaining the function and workings of DNA to someone who doesn't have a clue, it is always helpful to use analogies. I remember Richard Dawkins pointing out that a recipe is a better analogy for how the DNA regulates development and function, than the perhaps more often encountered "blueprint" analogy.I think it would be interesting to discuss the respective qualities and deficiencies of these analogies. Is "recipe" indeed better, and why? In what ways would they still be misleading, or inadequate to really do justice? Are there any crucial differences that need to pointed out in order to avoid big misconceptions?Edited by Adminnemooseus, : Added the "DNA - " part to the topic title.

Would it be fair to say that the "recipe" analogy is better at illustrating that:- small changes in DNA do not necessarilly correspond with small changes in the organism, so a small change could have an enormous impact and completely change (or fuck up...) the end product(while small change to a blueprint is much more likely to only suggest a small local effect)
- not only the DNA instruction itself is important, but also at which moment it is called, or for how long it operates; i.e. the same instruction can have different impact depending on when and how long it is executed (while a blueprint has no 'time' dimension)At least those are the "high level" arguments that I would use when defending the "recipe" approach.