SQLServerCentral Editorial

A Problem with Cowboy Coding

,

As we implement more automation and complexity in our systems, we increase the ability of every worker to accomplish more during their day. In many cases, this is a good thing, as many of us have no shortage of tasks to complete. Being able to do more is good. Some technology workers, and likely some managers, view these changes as opportunities to use less people to do the same work. That can be worrisome for many people, especially those concerned with job security. That's a difficult situation to find one's self in.

I'm not sure if this worker was concerned about losing employment or just wanted to get more work, but a contractor put logic bombs in his project that would cause issues at various times. He would then be called to fix issues, getting paid for his efforts. Recently he was sentenced to prison time for his actions.

Most workers make an effort to do good work and provide value for their time and knowledge. They may still make mistakes or cause problems, but it's often unintentional, not malicious. We also don't look to create future problems for others, though certainly we may accrue technical debt over time as we make trade-offs in our work.

Whether intentional or accidental, we want to limit mistakes and issues in our code. This is one reason that modern development in a DevOps environment isn't a wild west, cowboy style of work. We use code reviews and continuous improvement to help other developers learn to write better code. We use automated processes with instrumentation and logging to ensure we know what code is deployed when and by whom.

We may move faster with DevOps, but it's often safer, more secure, and more accountable than previous methods of building software. It doesn't prevent mistakes, but every part of the process should be more transparent and auditable, which is good for everyone involved.

Rate

5 (1)

You rated this post out of 5. Change rating

Share

Share

Rate

5 (1)

You rated this post out of 5. Change rating