Running Adobe ColdFusion on ARM: An Adventure in Unsupported Territory
I tried to move Adobe ColdFusion to AWS Graviton. ColdFusion worked. The connector didn’t. Then CommandBox made the whole problem disappear.
Personal website for David Byers, a seaasoned ColdFusion developer with a soft spot for clean code, practical software advice, and the occasional well-earned rant about legacy systems.
I tried to move Adobe ColdFusion to AWS Graviton. ColdFusion worked. The connector didn’t. Then CommandBox made the whole problem disappear.
Because Reading About It Is Nice. Running It Is Better.
Because Production Doesn’t Care That It Worked on Your Laptop
Because a Message Isn’t Really a Conversation Until Your Application Says It Is
Because Accepting Random HTTP Requests at Face Value Is Poor Career Planning
Adobe has done a commendable job modernizing ColdFusion. The next challenge isn’t the language... it’s growing the community around it.
Everything You Need Before Writing a Single Line of CFML
Before you dismiss AI, remember what they said about calculators, word processors, and ColdFusion.
A blunt take on why calling older tech obsolete is condescending, and why experience with legacy tools still matters in software development.
Because Sometimes the Boring Solution Wins
Because writing your own chat administration interface sounded like a good idea right up until it didn’t.
A quick reference for the core ColdFusion AI concepts, when to use them, and what to watch for.
How to validate, block, redact, and safely fail AI inputs and outputs before your helpful assistant becomes a compliance incident with a loading spinner.
How to move beyond stateless AI calls by giving your ColdFusion assistant short-term memory, user-specific context, and preferences without accidentally turning it into a cross-user gossip appliance.
RAG lets your ColdFusion AI assistant answer from your own documents instead of guessing, while still respecting permissions, freshness, source quality, and the basic rule that the robot only reads what ColdFusion allows.
MCP lets your ColdFusion AI assistant connect to external tools, prompts, and resources through a standard protocol, without pretending the robot should be trusted with every system it can reach.
Letting your AI assistant call real ColdFusion application logic through carefully constrained CFC tools, while keeping validation, authorization, and production safety exactly where they belong: in your code.
A hands-on first step into ColdFusion AI, using ChatModel() to send a prompt, get a response, and understand where simple stateless AI calls fit before things get complicated.
A practical, jargon-light introduction to what LLMs are, what they are not, and how ColdFusion developers can start using AI without mistaking the robot for the application.
Every Chat Message Needs a Home
Your First Conversation with Slack