ALM Best Practices

What Do Developers Hate Most About Static Analysis?

  

 Email Article  

Tweet  

  

A couple weeks ago, we launched a survey about what developers hate most about static analysis. Here are the results as of Thursday July 21:

Some interesting comments so far...

• "I voted for the fact that fixing violations can produce other errors, because I have seen that happen. So while that is something that I wish could be completely avoided, that doesn't mean that I hate static analysis as a result. I have found enough real bugs through static analysis that the issue of producing other errors is an acceptable evil. In fact, it's a problem that exists any time you change code, not just because you're fixing a static analysis violation. "
  
  
• "...False positives are less of a problem than an unordered surplus of valid but low-priority bugs, what I’ve called the Embarrassment of Riches problem (Supplemental Proceedings of the 21st IEEE International Symposium on Software Reliability Engineering, November 2010,  Pitfall #8). More broadly, unfocussed deployment of static analysis can result in impressive-sounding metrics, but end up wasting the time of the software engineers who have to deal with the defects (#1 & #6)."
  
  
• "I cannot vote for any of the choices in the poll...Having experience with good tools I cannot say that I hate something about static analysis. I may hate something about lousy implementations but not the technology in general. So, I have to admit I failed to answer the question of the poll. As opposed to the empty hate list, the list of things I like about static analysis would be large :-)  "

What do you think? Vote for your greatest gripe and/or join the discussion...

***

To explore static analysis white papers, articles, and videos/webinars, visit our Static Analysis Resource Center. Or, visit Parasoft's static analysis page.