We earlier imagined DNA as an instruction book. Let's even make it a reference book. When you need information about something you make a copy of the pages (genes) you're interested in, returning the book to the library. This way you don't have to risk losing or destroying the book.
In all eucaryotic cells DNA never leaves the nucleus, instead the genetic code (the genes) is copied into RNA which then in turn is decoded (translated) into proteins in the cytoplasm. Why? Wouldn't it be smarter if DNA itself was translated into proteins in the cytoplasm instead of using a RNA intermediate?
The answer, for many reasons, is no. One important reason is security. The cytoplasm is a dangerous environment for the DNA and the daily transcription of genes to proteins would be very harmful to the DNA, which has to stay intact in order to maintain life. Therefore, RNA works as a sort of throw-away version of DNA (like the copies from the reference book) - good for limited work but not for long-term storage. Another reason is to regulate the rate of protein synthesis. This will be further discussed in the section about protein-translation.
