Re: Warning : WINDOWS sample NOT representative
OK, I can't let that stand unchallenged because this is the kind selective quoting of statistics Microsoft tends to pull when they are selling to governments and military (which is why you can never quite get a copy of the exact same presentation from them later).
TL:DR version: using your statistics (minus the creative bit) it appears that OSX has LESS THAN A THIRD of the exposures of Windows since 1999.
Well, Mr Microsoft marketing, here is a lesson for you: NEVER try to use statistics if someone can actually check on what you're doing. If you get found out (like in this case) it doesn't just destroy your argument, it takes down your entire credibility.
Let's have a look at what you were trying to do. The very fact that you even tried is an illustration that you knew damn well you had to fudge the facts, so let's get some reality in there.
You were not trying to compare like for like, but one SINGLE version of Windows against the whole lifespan of OSX. So, let's leave out the version numbers for Windows as well then, shall we? That way, you end with 1304 vulnerabilities for ALL versions of OSX versus 4135 for ALL versions of Windows. In non-creative maths, that means that OSX has LESS THAN A THIRD of the exposures of Windows since 1999, even when Prince had not been singing about that year.
But hey, let's be a bit more precise and compare them year by year using the full facts, just to see what remains standing of your "facts" when reality hits them. Ooooh, look, on those same pages: shockingly, OSX has consistently fewer issues, year after year after year. Oops.
Let me translate that for you: fewer problems, fewer risks, fewer resources spent keeping it safe, fewer interruptions from getting work done - etcetera etcetera, you know, the arguments we provided earlier which you have kindly validated for us. It's a shame the site doesn't show any Linux stats because I reckon they'll be similar, but let's stick to the facts we have at hand - facts you yourself suggested we should use for this comparison.
For further entertainment, let's not stop there but look at what exactly those bugs represent. As you can see from the raw, unmassaged, clean yet somehow painful facts, Windows fairs EXCEPTIONALLY bad on privilege escalation: you know, the very thing that will install malware and steal your data even if the vendor itself doesn't help it along (as planned in Windows 10, which gave it the name "Slurp").
Now crawl back to Redmond and tell them you failed: you need something that isn't actually based on facts. I'd use the phone if I were you - your Windows machine is probably hacked already.