Why Does Software Cause Security Vulnerabilities?
It seems as though massive malware infections and breaches are reported every day. They’re often blamed on software vulnerabilities, but why is software vulnerable.
Unless you are an astronaut or a fighter pilot, almost all the software you use has security vulnerabilities. When your phone, television, or content management system demands to be updated, it is a safe bet that the new version fixes some of the vulnerabilities in the old version. But why are their vulnerabilities in the first place?
Complexity, developer error, and budgets all have a role to play.
Vulnerabilities are caused by bugs, mistakes in the way the software is written or in the way it interacts with other software and hardware. It is possible to write code that has no bugs, but bug-free software is either very simple or very expensive.
Simple programs can be verified by a human. As software becomes complex enough to be useful, it is impossible for a single person to understand all the ways it might behave.
Lines of code isn’t a good measure of complexity, but it provides some idea of the scope of software. The average WordPress theme has over 36,000 lines of code. WordPress itself comprises about 400,000 lines. The Apache web server has just under 2 million lines of code. The Zend PHP framework clocks in at 3.2 million lines. The Linux Kernel has an astonishing 19 million lines of code. Microsoft Windows is somewhere north of 50 million lines. As a comparison, there are 157,760 lines in all of Shakespeare’s plays.
Software is ferociously complicated; the most complicated artifact ever created by humans according to some. It is beyond the capabilities of even the smartest developer to understand every line, function, variable, module, and the combinatorial complexity of their interactions.
And, of course, sometimes developers just make dumb mistakes. They’re rushed, they don’t think carefully about the code they’re writing or take the time to understand a library they’re using.
To develop software that is both complex and free of bugs, developers would have to spend a lot of time planning, testing, retesting, rewriting, and testing some more — and that’s what makes it expensive.
If you have NASA’s budgets and the software runs the navigation system of a space shuttle, you have the money and the motivation to make sure that there are no bugs — and even then it’s impossible to be 100 percent certain.
No one wants to pay $3000 for a weather app or web server. Nor do we want to wait five years for new features in our favorite Twitter client. So we accept that software has bugs.
To be sure, developers test software extensively. They run automated tests that catch bugs, they have long beta testing periods during which the software is used and monitored for problems. Responsible developers invest great effort to devise and implement processes that reduce the number of bugs. Irresponsible developers don’t bother. Regardless, bugs get through, and some cause serious security problems.
When security vulnerabilities are discovered, developers create patches to fix them. That’s why applications and operating systems demand to be updated so frequently.
Once it is understood that there’s a price to be paid for rapid innovation, software users have to play their part in reducing the risk that vulnerabilities can be exploited by criminals, and that means installing updates.