Who's Online

We have 56 guests online

Statistics

Members: 34710
News: 52
Web Links: 57
Visitors: 13193935

Please install Flash and turn on Javascript.

Home
Jeroen van der Zijp Interview PDF Print E-mail
Written by Administrator   
Tuesday, 19 June 2007

Jeroen PhotoTell us a few words about yourself.

My name is Jeroen van der Zijp; was born in Amsterdam, the Netherlands. My hobbies have always involved building things; all through Grammar School, I was doing electronics projects of various kinds. Then while studying Physics at the University of Amsterdam, I got bitten by the computer bug, and from that point on, I have been building software rather than hardware.

After graduating I left the Netherlands for the U.S.A. to pursue a Ph.D. I landed at University of Alabama/Huntsville because a professor I knew in Amsterdam came here and invited me and a couple of other fellow students to come to the States. We landed in computer-nirvana:- we had the run of various models of Silicon Graphics, Ardent, Symbolics, and even LMI (Lambda Machines Inc.) LISP machines. [The Silicon Graphics machiness were my favorites, of course! ]

Later, I joined CFD Research, where I developed CFD-VIEW, the Fluid-Dynamics Visualization software, and later on started on the FOX Toolkit.

Right now I am at deciBel Research; another place using FOX... ;-)

 

When did you start developing Fox Toolkit and what where your initial goals?

I started FOX purely as a hobby in spring of '97. (Yes, that means FOX is now 10!!). My initial goals were kind of vague:- just to do some project that was totally different from the stuff I was doing for work. I picked a GUI Toolkit as the subject, because it seemed like a problem that wasn't solved in the right way by anyone yet. Little did I know that when a small project at CFD Research successfully started using it, FOX got such a boost that it developed into what it is today!

 

How Fox compares to other platform independent APIs? (like .NET, wxWidgets, Qt and others)

First and foremost, FOX is portable, and written in C++ from the ground up. It is not a wrapper that uses another GUI toolkit underneath; as such, the behaviour of each and every Control is totally defined in a C++ class, and be subclassed to redefine behavior or looks.

What distinguishes the FOX development process from that of other toolkits is the close cooperation I've always had with its users; and of course, I'm a user myself! One can not underestimate the "Eat Your Own Dogshit" approach to development. The greatest GUI blunders you see today in applications with large user-interfaces would be easily avoidable, had the developers been forced to use their own applications!

 

What were the turning points in Fox Toolkit development? How helpful was the community?

Good software libraries are not written in a vacuum. In order to grow, it needs to be used. FOX's main turning point was when CFD Research started to use it for a small project. Of course, any reasonable developer would have decided to sprinkle the code with #ifdef - programming technique that usually leads to less flexible applications - to accomodate a second GUI platform. Clearly, I must not be reasonable, because I just wrote an entire toolkit instead!

I wouldn't have done this for a simple "Hello World", of course. But for the massive code base we had at CFD Research, writing a toolkit just for this purpose may in fact have been actually less work than trying to get all those millions of lines of code working on multiple platforms.

 

We are watching currently big efforts made in the OS development. (There is MS Vista and MacOSX on one hand and the open source linux with a lot of different distributions. Google will also come into play with their own OS.) What do you think about this OS war and what are your predictions?

Since I'm a UNIX guy, I feel the O.S. stops at a lower level than Windows people do. As such, I don't foresee very dramatic developments. I'd like to see some nicer API's than POSIX but I'm not holding my breath. There is such a staggering amount of software resting on the foundations that were laid in the past decades, I'd be very surprised if we will see more than incremental developments in the next few years.

 

How do you see in general the development of open source software? What are the advantages and disadvantes of open source vs closed source and commercial SW?

The fundamental advantage of Open Source is that of control. There are basically no disadvantages, or at least, if you look at the greater picture, it will become clear that all disadvantages that are often cited by this or that trade-rag, are in fact short term ones. Long term, having control over your own destinity is paramount. Open Source gives you this.

 

There are many that say that software code development will eventually die or at least move to a very abstract level. Is this near?

No, I do not believe this. What's really going on is some sort of "telescoping effect", i.e. lots of applications building on layers, and those layers also building on yet lower down layers, and so on. But all these layers remain under continuous development. We will always need people adapting the kernel to new processors and architectural features; then GUI people will need to adapt to new graphics capabilities and new operating system facilities, and so on.

 

What do you do in your spare time? Favorite food? Movies?

Right now, I'm about to build a home. Planning has already been soaking up lots of time, but once things get underway things will really get rough. The good news is that I will have a nice computer room once I'm done, and I'm really looking forward to that! ;-)

 

Finally, how do you see the evolution of Fox in the future?

In the near future [1.8], the plans are to do an overhaul of OpenGL, Docking system, and further work on localization. Some of the current code for unicode probably will be redone a bit as well. Longer term [2.0], I will introduce theming support. At that stage we will have separated looks from feel completely. It is possible that I will also look into further abstraction of the "substrate" classes to isolate more code from the underlying systems.

During all this, I can not ignore the needs of my day-to-day usage of the library; indeed, lots of things are almost exclusively driven by this, and my "grand plans" sometimes need to take a back-seat to the things I need to add to the library for the applications I'm getting paid to write...

 

Contact Info

This e-mail address is being protected from spam bots, you need JavaScript enabled to view it

http://www.fox-toolkit.org/

 

Last Updated ( Sunday, 11 November 2007 )
 
< Prev   Next >
Download

Login Form






Lost Password?
No account yet? Register

Polls

What is the feature that you are missing most from Kerkythea?
 
Generated in 1.52872 Seconds