I have a practical example to explain what, at first glance, doesn't seem to add up. Many genomics analysis contain a step to clean up DNA sequences. For this "

I have a practical example to explain what, at first glance, doesn't seem to add up. Many genomics analysis contain a step to clean up DNA sequences. For this "Read Trimming" step there exist no fewer than 40 open source tools. Let's say you need this step in your project and use one of these tools. You find an issue the original creator is unwilling to fix: just choose another tool, problem solved. You find that the tool does not perform well enough on your data: choose another tool, problem solved. So while the articles points are true, they in practice often don't lead to a real pain.