5 FaceBook Source Code SecretsBy Angsuman Chakraborty, Gaea News Network
Wednesday, August 15, 2007
I glanced over the Facebook source code for both index and search pages. As an enterprise developer of some experience, source code is always interesting to me as it provides an insight into the health of a company. The interesting highlights of the Facebook code are:
1. Facebook writes pretty clean code & maintains coding standards. The code is readable.
2. To Facebook coder: Calling a function to do the magic isn’t an example of clean code; in response to Facebook coders comment - “Holy shit, is this the cleanest fucking frontend file you’ve ever seen”
3. Facebook doesn’t do MVC. There isn’t any separation of concerns as far as I could see. It is also not a designer’s job. It is created by developers with tons of procedural PHP code.
4. Facebook also doesn’t do much object oriented coding either, at least it wasn’t apparent from these source files. It is plain and simple procedural code.
5. Facebook uses templates which is nice.
It does some interesting stuff with advertising, specifically estimates the length of page and attributes to determine advertisements to show. There are few interesting ideas here.
Did you expect more? There isn’t much more really, take a look yourself. I wonder if MySpace’s code is as clean.
Tags: Software Review