Interview - PHP's Creator, Rasmus Lerdorf

The membership of the SitePoint community forums recently got together and produced a bunch of questions for PHP's original creator, Rasmus Lerdorf. In reviewing his responses, I was pleased to discover that the man who originally put the PHP machine in motion maintains an unclouded vision of what the open source movement is all about.

He is quick to play down his contribution to what PHP is today, instead attributing most of PHP's success to the vast community of developers that have signed on to the project over the years. In a sense, Rasmus today is simply PHP's biggest fan.

But enough from me; let's hear what Rasmus had to say!

In the beginning...

SP: What was your first contact with the Open Source movement, and what is it about Open Source that got you hooked?

RL: Well, back in the early and mid-90's the term "Open Source" did not exist.

"Free Software" existed, of course, and I had been playing with Linux almost since the very first release in 1991. Previously I was using QNX and Xenix and then started to fiddle with Minix until Linux rescued me.

I don't think I was ever really "hooked" by a "movement". When you don't have the money to buy SCO Unix and you can download something that works and even find people who can help you get it up and running, how can you beat that? Religion never really played a part.

SP: What led you to develop PHP? And what do you think this language has to offer that others don't?

RL: The first version of PHP was a simple set of tools that I put together for my Website and for a couple of projects. One tool did some fancy hit logging to an mSQL database, another acted as a form data interpreter. I ended up with about 30 different little CGI programs written in C before I got sick of it, and combined all of them into a single C library. I then wrote a very simple parser that would pick tags out of HTML files and replace them with the output of the corresponding functions in the C library.

The simple parser slowly grew to include conditional tags, then loop tags, functions, etc. At no point did I think I was writing a scripting language. I was simply adding a little bit of functionality to the macro replacement parser. I was still writing all my real business logic in C.

In the end, what I think set PHP apart in the early days, and still does today, is that it always tries to find the shortest path to solving the Web problem. It does not try to be a general-purpose scripting language and anybody who's looking to solve a Web problem will usually find a very direct solution through PHP. Many of the alternatives that claim to solve the Web problem are just too complex. When you need something up and working by Friday so you don't have to spend all weekend leafing through 800-page manuals, PHP starts to look pretty good.

SP: Looking at the usage figures, there are now over 9 million domains using PHP. Did you have any idea that PHP was going to become this big? How does it feel to know that your product is probably the best alternative to Microsoft's solutions for the Web?

RL: First, to be clear, I did not develop the PHP we know today. Dozens, if not hundreds of people, developed PHP. I was simply the first developer.

PHP is very much a collaborative project. Think of it this way: you have a Web problem. You can either go to the store and buy an expensive shrink-wrapped product that may or may not solve most of your problem. Or you can get together with a couple of thousand people who have the exact same problem as you, and work out a solution that works for all of you.

Not only will you get a solution that addresses your exact problem, you'll also become part of a like-minded community where ideas and experiences flow freely. That beats any commercial product you can go buy at a store, and to me is the best way to develop this type of software.

So when people ask me what it feels like to have developed something that millions of people use, it doesn't really fit with how I view things. In the end, I am simply the first member of a community that has arisen around one approach to solving the Web problem.

SP: Who would you call your hero? Which people in or outside of IT have inspired you?

RL: I don't really get inspired by people in the metaphysical sense. But I definitely appreciate and respect a slick solution to a tough problem.

SP: During your years of PHP development, what do you think was the most important decision you had to make? Are there any decisions you made that you now wish you had decided differently?

It is tough to ask me to second-guess decisions that were made 6 or 7 years ago when PHP was used by a grand total of 1 person. Don't forget that I did not sit down to write a scripting language that would be used by 9 million domains: I sat down to solve a problem. Solving the problem by 5pm so you can go to a movie with your girlfriend leads to some aspects that aren't ideal 7 years later, when thousands of people have to work around that late-night hack you added.

The most important decision I made along the way was probably to give up control. To open up the project and give just about anybody who asked full access to the PHP sources. This brought in a lot of excellent talent, and people tended to feel a real sense of ownership. The PHP project is probably one of the biggest out there when it comes to the number of people with commit access to the CVS repository where the code and documentation lives.