Slashdot Mirror
More Mac Vulnerabilities Than Windows In 2007?
eldavojohn writes "A ZDNet blog reports stats from Secunia showing OSX averaged 20.25 vulnerabilities per month while XP & Vista combined averaged 3.67/month. Is this report card's implication accurate, or is this a symptom of one company turning a blind eye while the other concentrates on timely bugfixes? 'While Windows Vista shows fewer flaws than Windows XP and has more mitigating factors against exploitation, the addition of Windows Defender and Sidebar added 4 highly critical flaws to Vista that weren't present in Windows XP. Sidebar accounted for three of those additional vulnerabilities and it's something I am glad I don't use. The lone Defender critical vulnerability that was supposed to defend Windows Vista was ironically the first critical vulnerability for Windows Vista.'"
They're just looking for excuses to downplay the results of the report.
No artificial metric really matters in the security landscape.
In the end, what matters is the real-world security performance of these systems. Sure, it's not so easy to quantify and measure, but stories like this ZDNet fodder are just pageview generators, and nothing more.
I know you put a lot of work into what you feel is a clever post, but all you did was come across as the exact kind of poster you are describing. And your link is really irrelevant as it was Apple supporters (mostly) who over-played the outsider status, not Apple itself. What kind of half-baked value system do you employ when you decide who is cool by what OS they use? An OS is a tool and you should use what fits your needs best. I'm a media junky and like to dabble in editing, that makes OS X my best choice. If I were still a PC gamer, you can bet I would use Windows. But that doesn't excuse the long history of Windows security issues, and an article that spins a a year where Windows finally has fewer vulnerabilities than another OS as proof of progress is really just proof how many people don't get it. The bigger question is how those vulnerabilities were handled, from point of discovery to solution, and that is where MS always breaks down.
Bush is the best President in history because he has fixed fewer problems.
Well, it has never been successfully tested.
Absolutely. Vulnerability counts are worthless. Here's the simplest example I can think of:
My friend and I both maintain a tool of some sort. We both get ten security vulnerability reports sent to us each year. I patch ten security bugs ten minutes after they are reported and my friend sits on the first ten bugs for a year, then the next year, we both fix ten vulnerabilities in the second year. However, for a user that keeps their system patched, I have an average of slightly over zero exposed vulnerabilities, while my friend's software exposes slightly over ten. According to the vulnerability count, however, I had 20 and my friend had 10.
Check out my sci-fi/humor trilogy at PatriotsBooks.
I don't get it. You opened port 80 on different machines, and saw different traffic, none of which managed to exploit the web server.
I'm sceptical this tells us much about anything, beyond maybe the set up of your NAT/DMZ. Otherwise you should have received exactly the same traffic on both web servers. Bots don't check the OS before sending their exploitable GET requests.
So I took a look at a few sample vulnerabilities and it leaves me Flabbergasted. The person who wrote this article and composed the data should be beaten. The ones listed as OS X vulnerabilities are primarily holes in software that runs on OS X, much of which does not even ship with OS X by default. A lot of it is holes in various Web server modules, some of which do ship with OS X, but are disabled by default. Some of them are NOT EVEN VULNERABILITIES... like CVE-2007-3876 which is a number reserved for use by an organization for the next time they report a vulnerability, but they haven't assigned it to anything yet. Whole ranges of numbers listed are like that. I mean did the author even click on the links he's providing? I tried, I was more than twenty items into the list of "highly critical OS X vulnerabilities" before I found one that actually affected a default install of OS X, and it was a potential denial of service for SSL Web sites if you have a machine in the middle. Of the first 30, 12 were reserved for future use and not real vulnerabilities, 7 were holes in the same Perl library, and 5 were holes in tcpdump. Only one was a real, hole that could be exploited on a default install without additional software being added, or it being reconfigured as Web server or something.
Another question is, for the real vulnerabilities to the OS's, how do they decide what the danger level is for a vulnerability? For example, one low rated one for WinXP (CVE-2007-2228) was a possible remote exploit, whereas a Highly cCritical one for OS X (CVE-2007-0267) was a denial of service on a machine, requiring a local user account. Does this make any sense to anyone?
I'm all for pointing out security problems in OS X and other OS's and doing comparisons of relative security, but this is just a sad joke. Please, can we at least get articles by someone with the tiniest bit of a clue instead of the number game from someone who might be able to count, but apparently can't be bothered to read his subject matter.
... until there is a self-replicating Mac virus in the wild.
This kind of cavalier attitude is what gets people hacked. Clearly you aren't watching your logs very carefully (or you're blocking those ports externally with some kind of firewall), because anyone who runs an SSH server (which is presumably what you're doing on port 22) knows that you get TONS of dictionary attacks. Before I disabled password authentication (and switched to using key-based authentication exclusively), I would sometimes get 20-30MiB of logs, all failed PAM logins with common usernames and from a variety of hosts. Clearly I'm not alone either.
From your experience? How do you even know when Apple has a fix? How do you know when the vulnerability has been reported? Are you basing this opinion on fact, or is it your "feel" that Apple is better than Microsoft about this?
Microsoft releases most patches during the Tuesday release cycle.
As someone who works in IT, I can tell you that we don't want patches released "as soon as they are ready". Patches need to be tested, and they need to be tested with other patches. You may not think that Apple patches cause issues, and usually they don't - but even one incompatibility could result in thousands of our users being down for hours or even days. 1000 employees being down costs us $1000000 per day. That's a damn big incentive to get it right.
With the Tuesday cycle, we can test ALL of the critical patches at once, together (about 2 weeks of both automated and manual testing). Then we can roll them ALL out to a pioneer group for a week, and see if any problems arise. If they don't, everyone gets the patch on the 4th week - and the process restarts. Our IT department has people dedicated to doing this cycle.
Guess what? We use the same Tuesday cycle for Mac and Linux patches. So what does Apple's "when it's ready" release process buy us? More time for the script kiddies to reverse-engineer the patch and exploit the vulnerability.
Agreed. Why don't we compare something like Windows Vista? Oh, wait, they did. Vista has fewer reported vulnerabilities than XP now, and far fewer than XP had in its first year of release. Not to mention far, far fewer than Mac OS X.
So, what does this mean? Do these numbers mean that Vista is more secure than Mac OS? No. The number of vulnerabilities is a poor measure for how secure an operating system is.
What it does mean, though, is that all is not well in Wonderland. Security is a process, and that process needs to be well-developed regardless of the software used. Mac OS X is not a silver bullet. Neither is Linux.
Agreed, although not all the "vulnerabilities" listed in this so-called study do ship from Apple, many are third-party applications that just run on OS X. Also, OS X includes a lot of cool tools with their OS, because they are free. 99.99% of the time, these tools are never used, let alone exposed to the outside world. For example, almost a third of the first 30 CVE's listed in this study apply to the same Perl, regular expression evaluator. Now how many users do you suppose turn on Apache and this module and make use of it on a Web page they're hosting from their home computer? I mean these tools are great for Web developers that want to test stuff on their workstation, but that is likely about all they are used for, in the very rare cases that they are used. That particular module accounts for 8 of the "vulnerabilities" in OS X listed.
It is fine to list these as vulnerabilities, but for a comparison to vulnerabilities in Windows, well they're pretty useless because of the use case as well as the dozens of other things wrong with this study. I mean, the OSS team developing this module lists each and every potential hole they an find on a public Website and it is counted by Secunia. Their list for MS includes only holes that have been discovered by the public and which MS has acknowledged. Since MS does not publish most of the bugs they find, none of those are counted against MS, including the ones they don't bother to fix (more than 50% according to an ex-MS developer I know).
Secunia knows this. Every respectable security expert knows this. The only problem is, random bloggers don't seem to know this, and write "articles" about it which get widespread readership, misinforming large numbers of people and leading them to make incorrect decisions that end up causing problems for everyone.
If you read some of the OS X vulnerabilities, you'll see that they're often in non-Apple software, such as CVE-2007-5476 (Highly Critical) which describes a "vulnerability in Adobe Flash Player 9.0.47.0 and earlier, when running on Opera before 9.24 on Mac OS X". The Microsoft vulnerabilities tend to be referring only to the Microsoft software
Also, the way they rate vulnerabilities seems to be different. Microsoft "Highly critical" vulnerabilities seem to all be remote arbitrary code, and "Less critical" can be remote DoS, whereas "Highly critical" on OS X seems to sometimes include DoS. Infact, CVE-2007-4702 (less critical) doesn't even seem to be a security vulnerability. I thought it was discussed and found that the application firewall on OS X functioned as documented (though potentially not as a user would expect). CVE-2007-3036 and CVE-2007-0023 seem to describe similar vulnerabilities, but they're rated less critical on Windows than OS X.
Well technically Apollo 11 had more things go wrong than did Apollo 1, but guess which one I would have rather been on?
http://www.mhall119.com
Could that have something to do with the fact that "Linux" means tens of thousands of different applications? In fact, how exactly is a SquirrelMail a Linux security threat? Why not a Windows security threat? Doesn't it run on Windows too? It's a web app.
Please make a difference between security threats targeted at GNU/Linux itself (the kernel and GNU tools) and something targeted at a 3rd-party app which may very well run on other OS as well.
Are you actually dumb as a rock or just trolling? How can you say there aren't enough Linux machines out there? What do you think most of servers of all kinds run on? Don't you think that a virus or worm would have a lot more to gain by breaking into servers than personal desktop computers?
That settles it, you ARE as dumb as a rock. You seem to really believe that somehow Linux apps are staying out of harm's way by sheer luck and hiding behind the poor Windows computers. Has it ever crossed your brain that perhaps Linux apps are designed with security first in mind? Such as, I dunno, NOT ALLOWING BLOODY EMAIL ATTACHMENTS TO BE EXECUTED?
i ate crayons when i was a kid and now i have two braincells and the blue ones taste nicer