Posted on

Why “Polyglot Programming” or “Do It Yourself Programming Languages” or “Language Oriented Programming” sucks?

Last year we saw the launch of a new Web programming language Dart - Structured Web Programming from Google. A very interesting approach to support web application development. Not so long after Go, Groovy, Ruby, Scala, << Name your DSL here >>; we see Dart. Is it a good thing to have at least one programming language to solve one problem? The answer is, like we already know, it depends.

Stay Away From “Do it Yourself”

It is your choice as to if you will try to do things yourself or allow the truly seasoned professionals to help out. Some decide that they are going to try to go it alone when they are programming something new, but this often ends up in a less than desirable place. It may even be more expensive than just hiring an expert who can help you get it programmed for you in the first place.

Posted on

Evolving Domain-Specific Languages

When designing domain-specific languages (DSL), the most critical choice is the selection of concepts that form the basis of the language. Sometimes concepts for the language come from the customer directly or from domain traditions. Sometimes the DSL developers force customers to use what they are already familiar with. Implementing these concepts in the DSL as close to the domain as possible is usually a good choice so the language will be readily understood by domain experts. However, instead of sticking to the existing domain concepts, it is also possible to evolve domain concepts by designing higher-level concepts based on existing concepts. In this article, I’ll demonstrate how such an evolution could be done using a classic state machine language as an example.

State Machine Language

Martin Fowler wrote a classic book called “Domain-Specific Languages.” This is a really good book, and I recommended reading it if you have plans to go into DSL design. The state machine sample from that book is copied from an article, and a lot of DSL framework developers and language workbench developers demonstrate the capabilities of their tools based on this state machine language. This language has become a kind of a DSL tool benchmark language. The sample is in the following code block (one of many variants. I've tried to compare to the one that is implemented for Xtext version of the language at the blog post by Sven Efftinge, “Martin Fowler's State Machine DSL with Xtext 2.3”):