Planning ahead of time needs a crystal ball no developer has.

TL;DR: Don't guess things that might not happen.

Problems

• Coupling

• Testability

• Readability

• YAGNI

Solutions

1. Create great models and bijections first.

2. Create a conclusive benchmark once the model is working.

3. Programmers waste enormous amounts of time worrying about, the speed of noncritical parts of their programs, and these attempts at efficiency actually have a strong negative impact when debugging and maintenance are considered. Donald Knuth

4. Design for Performance.

5. Use Test Driven Development technique. It always favors the simplest solution.

Examples

• Weird data structures

• Caches

• Singletons

Sample Code

Wrong

class Person {
    ancestors() {
        cachedResults = GlobalPeopleSingletonCache.getInstance().relativesCache(this.id);
        if (cachedResults != null) {
            return (cachedResults.hashFor(this.id)).getAllParents();
        }
        return database().getAllParents(this.id);
    }
}

Right

class Person {   
  ancestors(){
     return this.mother.meAndAncerstors().concat(this.father.meAndAncerstors());      
  }
  meAndAncerstors(){
     return this.ancestors().push(this);
  }
}

Detection

This is a design smell so it can not be detected by mechanical tools (yet).

Tags

• Premature Optimization

• Antipattern

Conclusion

Defer performance decisions until functional models are mature enough.

Donald Knuth created/compiled the best/fastest algorithms and data structures. With great wisdom he warned us of abuse. Why do we think we are smarter than him?

Relations

More info

• c2 wiki

• Wikipedia

Credits

Photo by Markus Spiske on Unsplash

---

Premature optimization is the root of all evil.

Donald Knuth

---

This article is part of the CodeSmell Series.

Last update: 2021/07/02