Analyst Forrester has come under fire from leading Linux distributors for suggesting that Windows and Linux are equally secure.
In a research note published last month entitled Is Linux More Secure Than Windows?, Forrester argues that the perception that Windows is far less secure than Linux is unfair.
In assessing the relative security of open source and proprietary platforms, Forrester looked at factors such as vendors' responsiveness to vulnerabilities, the severity of vulnerabilities and how thoroughly flaws are fixed.
But this analysis is flawed, according to a joint statement by Linux distros Debian, Mandrake, Red Hat, and SUSE issued this week.
"Despite the report's claim to incorporate a qualitative assessment of vendor reactions to serious vulnerabilities, it treats all vulnerabilities as equal, regardless of their risk to users. As a result, the conclusions drawn by Forrester have extremely limited real-world value for customers assessing the practical issue of how quickly serious vulnerabilities get fixed," they said.
The vulnerability data corrected by Forrester is accurate enough, according to Linux distros, it's just Forrester's conclusions they take issue with. They believe Forrester failed to treat open and closed source software the same way.
Forrester's point that closed and open source security are roughly equivalent is in line with those of distinguished academics like Ross Anderson (who is anything but a Redmond lackey). It's also fair to say that how technology is deployed more important than the platform a user chooses from a security perspective.
The four Linux distros helped Forrester create its report. The perceived failure of the analyst to present their case fairly in this instance appears to have provoked an unusually strong rebuke.
Forrester strayed into an area staked out by strongly held, sincere beliefs. It's difficult to strike the right balance in commenting on this area (we speak from experience and have the flames to prove it). If Forrester made a mistake it would seem to be one of drawing sweeping conclusions on security based only on looking at vulnerabilities. The majority of viruses (Windows security public enemy number one), after all, rely on tricking users into running malicious code rather than exploiting a specific vulnerability.
So risk or exposure are better metrics of overall security than vulnerability handling alone. Unfortunately risk is even harded to assess than vulnerability handling. ®
Open and closed security are roughly equivalent
.NET more secure than WebSphere
Security is our biggest ever challenge Gates
IT managers trust Microsoft on security (up to a point)
Human error blamed for most security breaches