Digital Ninja

The conference is over and I'm sitting in the hotel room trying to organize my thoughts. A four day conference on the most advanced ColdFusion techniques, frameworks, various design patterns, etc. can really make one's head spin. Without getting into specifics, let me see if I can take a breath and jot down what I thought of it all.

First, this was a first class meeting of the most brilliant minds in the ColdFusion community. As I understand it, there are programmers in other languages who would still scoff at that. To them I say they better take another look at ColdFusion. These experts are also experts in Java, design theory, etc. - I'm pretty impressed with all the speakers.

Second, I was somewhat correct in my last post about wanting to go home and change what we do and how we do it. It is mind-boggling when you see what can be done, what is being done, and you realize you could be doing the same. However, we have some things to consider before launching a bunch of new tools and practices. I was able to break everything down into a few categories:

• New practices - these are steps in the development process we don't even have, but can add to our pipeline with little-to-no disruption.
• Improved practices - We're doing some good things, but we can do them better and slightly different. There will be some disruption, but not too much.
• Replacements - Tools and techniques exist that simply blow away what we do. We should adopt some of these, and its going to be a rough ride to make it happen.

What needs to happen now is I must become an advanced user of each tool or technology I would like to see adopted across the team. I must then train the developers how to implement these technologies, and then we phase them in by either adopting each approach for new projects or blocking out time to refactor existing projects.

The bottom line is that we're accomplishing a good many of the right goals for a web application development team, but we're taking the long hard path through the desert with no water, no shoes, heading straight down razor-blade road and then wading across rubbing-alcohol river. This is no fun. What we do can be fun, and these new ideas and technologies can help us automate/eliminate the boring, the repetitive, the un-fun tasks.

Here goes nothing. :)

Here's a stroke of irony for you. I've spent the last year or so making leaps and strides in my understanding of web application development and architect-ing. I've amassed many tools that are on the newer edge of my trade, and based decisions on trends I've seen. It has been a tough road with no one to show me or teach me. I'd like a moment of self-appreciation to pat myself on the back for that one, if for no other reason than that the products I've helped launch and continue to support haven't crumbled under their own weight. :)

That's the tool part. The frustration? Now I have no time to work the tools. Years ago when someone came to me with a problem, I would find the next time I'm not busy and sit down with the code editor and a database query tool and author the code right on the production server in the very spot it will be launched. In a day or two, maybe hours, I would be done. Why, some of those applications are still standing! And boy are those a pain in the neck to support, for only I understand how they work.

Well, today I have wire-frame tools, documentation guidelines, wikis, a pattern for API development - heck, I have an API! - I have code libraries to choose from, frameworks to support the application, code repositories... the list goes on . Let me get to the point. Now I feel like I can't do anything in the time given to me. For anything! There are so many pieces to the puzzle - and I haven't had time to cross train the developers I lead. They haven't been asked, nor expected to, go off and make the same self-administered journey that I have. Talented and wonderful folks, every one of them, but I have some work to do to get the whole team ready to use the same tools I do. Or at least the members of the team that need to bother with such tools... I wonder if those are all in the same.


In a couple weeks I head off to cfObjective. I think I'll be dazzled with the latest and greatest development approaches and technologies. In fact, I suspect I'll come back wanting to rewrite and restructure everything we do! Don't worry - I'll resist in the name of good business practices. :) But I think my primary goal is going to be finding out what are the expectations, for an architect or senior developer, regarding how much preproduction and production goes into a project. Are there hordes of well behaved developers out there using all these tools and practices that lead to the road of easily-supportable application code? Or are the majority of the developers out there still winging it, writing code on their production servers without frameworks or source control?

What's the middle ground, and how far am I from it?

This January brings a few things to my plate. At the office we're performing one of the most significant upgrades to the enterprise that we've performed to date. It's not only the upgrade of the central database application (the student management software) but also the SQL Server version across the board. Since we engage in replication from the student DB system we need to make the subscriber a SQL 2005 machine as well. So much for trying to space out the challenges.

Luckily we've been able to hold off upgrading Coldfusion from 7 to 8 on all but the newest servers.

The more fascinating item on the board is that I've been invited to and am participating in the specification for the next major version of the Fuse Doc standard. It all started because I squawked on a FuseBox mailing list about the FuseDoc not providing a way to describe incoming or outbound objects (CFCs). As is apt in the community, this action turned into a call to arms to revise the standard, as it's mighty old and dusty.

The most rewarding aspect of this work has already been discussing these ideas in a form with people such as the team Fusebox members as well as the authors of books that I've acquired and make my entire development team read and learn.

I suspect 2008 to be an interesting, fun year.