So, I guess that I should introduce myself first. That would be the polite thing to do, right? My name is Aaron Kendall, and I’ve been a software engineer and developer for almost two decades. (And, yes, I’m enjoying the expected mid-life crisis, but that topic is for another time.) During that time, I’ve been a member of various software projects, both at work and at home. Does that make me an unbelievable talent and an undisputed expert? Well…probably not. However, I am very good at what I do, and when it comes to a few specific topics, I know them well.

So, I decided to humor myself (and possibly you) by attempting to stoke the fire of my mind, in order to share some smoky tendrils of pine-flavored knowledge. Hopefully, some of these written points will be novel enough, and then the content here can perhaps be of some help to someone. If not, I invite you to read and mock any and all attempts at trying to be clever. Either way, it should make for a fun time.

2 thoughts on “About

  1. Dear Aaron, After reading your article about DSL. I got some substantiation on the path I took in a recent project for creating metadata driven generic solution. The validation layer I was trying to formulate with out having any knowledge about DSL consisted of extracting the rules into related SQL operators among data elements that consisted in a file and/or across the files. Though, I struggled to create a lexicon that might resemble an adhoc query generator, I have been thinking that there should be more definitive proven way of abstracting validation rules by business owners themselves. Do you have some example or an article that describes on creating metadata, approach, lexicon, functionality to enhance the layer with additional metadata and a method or tool to do it? There are many complex validation engines but DSL seems to be the right choice as you iterated. Any help in this regard is highly appreciated for me to fathom and create a very generic and simple scalable metadata driven engine.


    1. Dear Sudhakara,

      First, if I understand you correctly, you’ve had some success with MDD, and I’m glad to hear it! As for a verbose lesson or a comprehensive tool to help one create a DSL based on MDD…now that’s a tough one. 🙂 Of course, there are projects that can help you with DSL creation (like XText), but unfortunately, there still isn’t an abundance of material or tools out there regarding MDD. The general exercise of how one would use abstraction to create a DSL based on MDD…well, that could be an article by itself…or even a book! But if I could describe it in succinct terms, I would say this: it’s best to work backwards. Ask the business owners to write out the rules in a verbose, human language (like English). Then, like a natural language parser, I would break down their rules into nouns and verbs, and I would look for patterns. (Like if two operations were always combined together.) Eventually, you can create a finite, computer-legible language based on that analysis. If you like, we could try one of your rules as an exercise. Do you have one to show?


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s