Open BlueDragon Steering Committee Interview Series - Peter Amiri

In our continuing series of interviews with the Open BlueDragon Steering Committee, it is my honor to introduce you to Peter Amiri.

What would you say is your biggest contribution to the CFML community?
Sure I've been using CF since the 1.5 days, I've consulted with some large corporate users, and have written some CFML projects that are going into their seventh year of production. Now most people probably know me because of my involvement at MySpace and the easy answer would be to say we proved a CFML application could scale to millions of users. But in light of my current involvement on the OpenBD Steering Committee, I think perhaps my biggest contribution to the community was simply getting BlueDragon in the door at MySpace.

I first contacted New Atlanta in late 2003 and we eventually started to test the BlueDragon product. Initially the JX server and eventually the .Net server. Along the way we worked very closely with the New Atlanta engineers to optimize the engine even down to the tag level on a few occasions. I'd like to think that all that work is now going to benefit the entire community with the release of OpenBD.
Tell us a little now about the MySpace project? For example, why CFML in the first place?
Thumbnails by Thumbshots.orgWhy CFML in the first place? That's a interesting question cause the project was already a couple of months along by the time I join and we had already settled on CFML. But the history is that the first version was written by a consultant in Perl. It bombed shortly after release and it was ported over to CFML in a week. But I think the better question would be why did we stick with CFML for as long as we did? During those initial 18 months we were cranking out code at a feverish pitch.

We had multiple developers and it wasn't unheard of to push out new features several times per day. Basically as features were completed they were added to the site. I don't think any other web application environment would have allowed the developers to crank out code as fast as they did. Speed to market was the number one driving force and CFML allowed us to achieve it.
Do you think the CFML market is shrinking, growing, or just staying the same?
I don't think there is a simple answer to this. We now have more books on CFML and more web hosts that support CFML than we've ever had in the past. But does that mean that the CFML market place is growing? I don't know about that. Let's look at this from the perspective of the Developer, the Project, and the Product.

If we look at each of these on a sliding scale we may be able to gain more insight as to how the market place perceives CFML. The developer scale has Newbies on one end and Veterans on the other. The Veteran developers aren't going anywhere but the Newbies have so many choices in front of them. The are inundated with press coverage about the latest development environment on the block. We're probably not picking up too many of these new developers.

Next lets look at Projects. We have new projects on one end of the scale and the other end has projects that have been in productions for years. Here again the old existing projects are probably not going anywhere but managers deciding on what to build a new project in are once again inundated with press coverage of the newest development environment on the block and we are probably not picking up too many of these new projects.

Next lets look at the Product if self. The sliding scale for this one is a little more interesting, because we have new version 1.0 type products on one end and mature feature rich products on the other end. Let's face it CFML is now over 10 years old. We don't get too much new coverage in the press which is exactly what perpetuates the issues with the two other sliding scales we already talked about. But is there a benefit to being old? Well, perhaps.

The beauty of CFML has always been it's simplicity. If we can take that simplicity and make it available to a wider audience with OpenBD then I think we can insure the continued growth of the CFML marketplace.
You've proven CFML is scalable, but should developers think about scaling when developing?
I think they need to. Let's face it, there aren't too many applications that need to scale to millions of transactions or hundreds of thousands of concurrent users. But every developer needs to be concerned about optimization. For years ColdFusion suffered from the stigma that is was slow and couldn't handle too much load. I personally did many share of consulting engagements where I was rewriting someone else's code. Many times people would say CF can't scale when in reality it was the architecture of the application or the developer that was at fault. Every developer needs to take the available resources into account and try to optimize their use of these resources.
Why does Open BlueDragon interest you?
This is a very exciting time for CFMLers. I've been using CFML for nearly 12 years now. I'm passionate about my use of it and my advocacy for it. The OpenBD project has tremendous potential for becoming the catalyst that brings more developers into the CFML camp.
What are you hoping to bring to the Steering Committee?
There are certainly more capable developers on the committee and even more in the general community at large, but I hope that my experience with large scale systems both at MySpace and at my current day job will be beneficial to the committee. I also hope to dispel any scalability questions about OpenBD by making some announcements around launch date. Stay will be worth it.
What one thing would you like to change about CFML?
I'm going to leave the arguments about language modifications to others better qualified to discuss them, but I think what is perhaps even more important for me is what I don't want to change about CFML. I alluded to CFML's simplicity earlier, more that anything else I would like to keep CFML and it's tag based language simple and accessible to everyone.
Your development environment of choice?
Mac OS, Apache, MySQL, Eclipse, Perforce or SVN. I currently user my PowerBook G4 primarily for the train trip to and from work, a Mac Pro at home and a Mac Mini at the office.
Looking ahead 12months, what needs to happen before we can claim Open BD a success
Frankly I think once we make a code drop and the community at large has a mature Open Source engine then the project is somewhat of a success. But I have higher goals for the project, I would like to see integration into linux distros.

Can you imagine if every linux installation could server CFML code out of the box. Along the same lines would be integration into hosting control panels. So every shared hosting account could serve CFML similarly to PHP right now. Once we make head way on these two fronts, then we should see more CFML open source projects and modules becoming available. Like I said before these are very exciting times.

Contact Peter @ peter.amiri@openbluedragon.org

Let me thank Peter for taking the time to answer my questions and his contribution to the Open BlueDragon GPL project.


Recent Cloud posts

Recent JAVA posts

Latest CFML posts

Site Links