What I Have Learned From Building Software for Real-World Problems

Bridging Physical Environment Realities with deterministic software code

Developing software to solve real-life problems has revolutionized my thinking about software development. I have found out that coding is just not the complete solution to a problem. The true test is to get software working effectively and reliably in real-world scenarios where hardware restrictions, network failures, storage space issues and unexpected user input may be present.

I listed here the most important things I've learned when I was doing software and hardware development.

1. Code Must Work in Real Environments

What is perfect in the dev environment, might not be perfect in the prod environment. The behavior of an application can differ depending on the server configuration, limitations, permissions and network environment.

2. Understand Hardware Constraints

For the ESP32-S3, I found that a correct ESP32-S3 flash partition and file system configuration is necessary for storage initialization. Problems that result from incompatible storage settings cannot be solved with software logic.

3. Troubleshoot the Root Cause

Changes in the code that happen randomly can be time-consuming. I have come to understand how to look at error logs, see what is the problem, test out any assumptions and strive to fix the actual cause and not the visible symptom.

4. Testing Must Go Beyond Simulation.

While logic is verified through simulations and local testing, the actual devices and production-like environments can uncover further issues. Testing should include hardware interactions, hardware integration, performance and failure modes.

5. Design for Failure

Networks go down, databases fail and operations can be lost. A reliable software should recover and store essential data in a safe manner, handle errors gracefully, and be safe in the event of an error and recovery.

6. Configuration Is Part of the System

Software operation is directly linked to the environment variables, database settings, file permissions, memory limits and hardware configurations. Correct configuration is as crucial as correct implementation.

7. Protect Data Integrity

Successful operation does not necessarily imply correct responses. Data validation, database constraints, transactions and duplicate checking ensure reliable and consistent data.

8. Build Maintainable Architecture

The separation of business logic, storage operations, hardware communication and network requests makes applications easier to test, debug and maintain. The architecture should be appropriate to the real needs of the project.

9. The student will grasp the purpose and intent of the problem in the real world.

Consider the users of the system, the environment in which it will run, the resources available, and what will occur if something goes wrong, before starting to write code. These details can be the deciding factor where it concerns the practicality of a solution.

10. Consider the Ways of the Written Word

The development process is more than just adding features. It includes tracking, auditing, evaluating assumptions, recording interdependencies, and planning for contingencies.

Conclusion

My biggest learning is that even if the code is correct, it doesn't mean that the system is reliable. Software has to function within a real-world environment and have a predictable reaction to changes or variations in the environment. Knowing about the environment, testing assumptions and planning for failures to ensure that systems are easier to maintain and more reliable when put into action.