Moved

Moved. See https://slott56.github.io. All new content goes to the new site. This is a legacy, and will likely be dropped five years after the last post in Jan 2023.

Tuesday, October 4, 2011

"Hard Coding" Business Rules

See this: "Stop hard-coding business rules" in SD Times.

Here's what's exasperating: "Memo to developers: Stop hard-coding business rules into applications. Use business rules engines instead."

Business Rules Engines?  You mean Python?

It appears that they don't mean Python.

"Developers can use [a BPM suite or rules engine] and be more productive, so long as they don’t use C# or Java as a default for development".

I'm guessing that by "C# or Java" they mean "a programming language" and I would bet that Python is included in "bad" languages for development.

Python has all the simplicity and expressive power of a Domain-Specific Language (DSL) for business rules.

Don't hard-code business rules in Java.  Code them in an interpreted language like Python.

Also, don't be mislead by any claims that business analysts or (weirdly) users can somehow "code" business rules.  They can't (and mostly, they won't).  That's what SD Times wisely says "Developers".  That's how coding gets done.

3 comments:

  1. The underlying point about reducing the friction of change management and moving towards more of a self-service mentality for business users is actually a valid one.

    Python is Turing complete - to really use it to its full power, you need to be able to think algorithmically and in terms of data structures.

    The key purpose of a business rules engine is to provide a constrained environment for *non* developers to make changes to business logic directly. Just as the data in an application has traditionally been the domain of the business users rather than the developers, a rules engine aims to put some of the *logic* in their hands as well (with appropriate access controls, of course, again, just like data).

    It's similar to the reason why JSON or ini-style config files are a better idea than using Python for the same thing: you *don't want* that level of expressiveness in your configuration settings, you want things to be explicit rather than algorithmically generated.

    ReplyDelete
  2. "constrained environment for *non* developers"

    The "constrained environment" may or may not work depending on the client. Personally, I have seen people become infuriated by limitations and started using Access/Excel.

    At a certain Vanguard location, they first generate reports w/ Access/Excel to explore/document the business rules. Once it's stable (ie they have figured out what they really want), then it gets incorporated into their infrastructure via a programming language.

    ReplyDelete
  3. So it's not the language, but the change deployment mechanism? Since it's interpreted it doesn't need a compiled binary to be shipped? But it still needs a testing cycle, right? Seems to me, many languages could work (even compiled ones, since you can compile and load a .NET CLR assembly at runtime). Obviously some languages might be more suitable than others in expressiveness and simplicity, but how do you measure that? How important is the readability of a traditional language like VB.NET versus something more declarative yet opaque (i.e. requires more experience) like XSLT, SQL or Scheme? If users can't program business rules, what level of developers should?

    ReplyDelete

Note: Only a member of this blog may post a comment.