Alan Williamson

I like to think I've been instrumental in getting more awareness and more adoption of frameworks. I published the Macromedia CFML Coding Guidelines back in 2001 and then the Mach-II Development Guide back in 2003.

Both are woefully out of date now but the CFML community has come a long way since then too. We have Mach-II, Model-Glue, ColdBox and a seriously updated version of Fusebox. We have ColdSpring and LightWire and Reactor and Transfer. We have several unit testing frameworks and now we're starting to see mock object frameworks as well. We're growing up as a community and adopting better practices all the time - I like to take some of the credit for that.

One word: community. A successful project has an active and growing user base with a two-way flow of information as the users provide feedback, testing, documentation and even code. Look at how Ray Camden's BlogCFC has become the de facto standard in our community and look at how rich the feature set is now, because of community feedback. The frameworks I mentioned above are all successful projects because of their communities.

It's growing. The 'problem' is that the web market as a whole is still growing very fast and so we're seeing new technologies (or even old technologies in a new skin - Ruby on Rails) grabbing the headlines because there is so much market to share. CFML is an 'old' mature technology with multiple vendors providing CFML engines and it still suffers from the poor reputation it garnered early on as a proprietary technology that didn't really scale (regardless of whether or not that was ever true).

The CFML community is definitely getting better in this area but open source is still a very new concept for the vast majority of them. Many developers have released code in source form but for the most part those offerings have not really been "open source" - no way to build a community, no clear licensing, no public bug tracking, no contribution path for the users. As an active participant in the development of several frameworks, I've seen developers out there who think "open source" either means that the core project should accept any and all changes the developer wants or that the code is just put out there to take and modify - fork - and use without giving anything back. Open source is a two-way street.

I'm a long-time advocate of open source. I contributed to the GNU C++ Standard Template Library back in the day and several other open source C++ projects. I helped port and test a number of large open source projects as they first moved to Mac OS X. And of course I contribute to most of the CFML frameworks. For me, an open source CFML engine is a natural progression.

I also think that opening up CFML in this way has the potential to get the language out in front of a lot of people who would never have considered it. If I was set on building a large scale system based solely on open source components, I would not have even considered CFML until now. I think this has the potential to really change the game.

In my early career, I worked on compilers, interpreters and runtime systems (including a couple of virtual machine architectures, long before Java appeared). Later on I was very active in the ANSI and ISO C++ Standards process - I was a voting member of ANSI J16 for eight years and secretary of the committee for three. I think my experience in language design, development and standardization, along with my long track record in open source projects, will be very valuable to the steering committee as the project moves forward.

I'd like the language to be more consistent so it is easier to learn and easier to use. There are too many special cases and some strange choices of attribute names. Developers old and new get frustrated with this sort of thing.

I'm also leaning toward cfscript more these days - after years of railing against it and wanting it deprecated - so I'd like to see that become a more complete way write entire CFML applications.

A 17" MacBook Pro running Eclipse + CFEclipse and MySQL (and multiple CFML engines).

I think we can claim OpenBD a success when we see hosting companies offer it as a no-cost option on their JSP plans and when we see a couple of large, high-traffic, high-profile sites running on OpenBD. And of course a vibrant community of OpenBD users that are contributing back to the project!

Contact Sean @ sean.corfield@openbluedragon.org — http://corfield.org/