Your statements about Dell’s infected motherboards, again, try to place blame on the Linux without citing explicit sources and ignore the fact that Dell used infected Windows software during hardware testing. If you take a look at these links:

http://www.theregister.co.uk/2010/07/21/dell_server_warning/

http://www.channelinsider.com/c/a/Dell/Dell-Replaces-Server-Parts-Infected-with-Virus-709631/

http://www.digitaltrends.com/computing/dell-accidentally-sent-out-malware-riddled-motherboards/

You’ll find that:

1. The virus is old; as old as 2003. This means any person running even outdated anti-virus software on a Windows system will be protected.

2. The virus was located in Flash memory of the motherboard and it got there as a result of Dell’s use of hardware diagnostic software on a system that had become infected with the virus in question. If Linux had any part to play in this its job would have to have been to function as a middle-man between the infected Windows-based benchmark machine and a new motherboard. This is very much similar to connecting a Windows machine to a motherboard over a network switch or some other dumb device whose sole purpose is to pass traffic/data between two devices. If you got a virus from another computer on a network, would you blame the Ethernet cable and the network you’re attached to, the other computer that got your computer infected, or your own computer for not having proper protection in the first place? Linux has protection in place for itself; it has no obligation to go out of its way to protect Windows machines unless an administrator sets it up to do just that. Besides this scenario seems a little far fetched; to use two computers IN SERIES to run benchmarks on hardware, what would be the point? And again, why blame the messenger/middle-man who’s just passing traffic?

3. They clearly state in the third link that it does not pose any threat to non-Windows systems, meaning it cannot execute itself on a Linux system (because it wasn’t compiled for Linux) and use it as a platform to attempt to infect non-Linux systems. So you cannot say that a Linux system actually had this virus running on it or that it used a Linux system as a platform to stage another attack against a Windows based machine.

I patiently await any rebuttal you have and ask that you source your claims.

In one case, Dell had an issue with infected motherboards shipping. Guess how the software got onto the mother boards? Hint: they weren’t using Windows interfaces to provision their system-boards.

So while Linux isn’t as susceptible to viruses as Windows is (Again, Linux can still get sick), there is always the issue of Linux passing a viruses along to other machines.

Even if you do contents control in Windows, the system cannot live without an anti-virus. I think he is not conscious that he’s mixing up virus with spyware, scams and malware. That is what Windows is able to offer at its best, no matter what kind of secure practises you are accustomed to. In the other hand, many Linux based systems *can be made very fool proof.

The only way for isolating a Windows machine is not using it at all (No Internet/Intranet, no data exchange with friends, nor executing/opening anything).

In a final comment, I don’t use Ubuntu because I think it is going the wrong way in the “fool-proof” race. Over-complicating simple things is the worst you can do in any environment (Have you checked out most default system settings carefully?)

I do agree with you; many computer users out there fall victim to little traps and schemes and so on. I didn’t mention that in the post here because it was a quick copy and paste from something I wrote in a forum thread that already had others saying the exact same thing and I didn’t want to sound redundant.

That being said, I would wager that the novice user likely doesn’t feel up to the task of messing around with software outside the official repos. Those that do probably have an understanding that says if you’re going to download a piece of software or some handy-dandy script from an untrusted source on the Internet then they’re kind of inviting trouble. There is rarely if any need for a new user to wander beyond the official repositories for their software needs, and if there ever were a need I can only hope they would venture to use something outside the boundaries based on the knowledgeable recommendations of a friendly expert. I mean, a simple rule of thumb any user should use is that if you’ve never heard of the software and you can’t confirm through others whether or not it’s worth a shit, then you should leave it alone. There is no way to completely safeguard any computer system from user negligence.

I really like your header image, but the .htm file extension is for Windows servers that (for historical reasons, aka DOS) were limited to a three-letter extension. Apache and other FLOSS apps aren’t limited with this, so they use .html file extensions.

God Bless

In short: What you say is mostly true but your conclusion that users don’t need an antivirus is completely wrong.

A distribution is fairly secure because, as you say, it has crowdsourcing to detect things like this. What you’re forgetting that behind every Linux desktop (just as with Windows) is a user… And users are idiots. As users, we’ve all done something stupid at some point.

Security is a skill. Without an active antivirus you can stay safe (even in Windows) by just acknowledging the providence of the things you install and run.

Some people might stick to what you can install from the repos but plenty of people install from PPAs, download little scripts like winetricks, add extensions to their browser… Crowd-sourcing does very little to protect these users.

The standard Linux user relationship with the rest of the system (that is: you run as your user 99% of the time) does help but if you’re in the position where you have a nasty file, it won’t need root permissions to completely ruin your day.

What if you downloaded and ran a script that bottled up your home contents, encrypted them and deleted the originals for a ransom. It wouldn’t need root permissions. This has been done on Windows.

Or a trojan. It only needs to run as the user to have full access to the user’s session and files and do things like SubSeven (BO, BO2K, etc) did for Windows.

Things can even propagate from the user to other computers over network shares, infect the users files…

My point is Linux doesn’t do that much to protect you from yourself. Sure you can’t wipe the whole system without root but installing Linux takes 30 minutes at a push. Years of photos/music/work/documents can’t be replaced as easy.

To stay safe, you either need to recognise the risks of untrusted sources, PPAs, random binary downloads (and put some research in) or, just as within Windows, have something there looking out for you.

You also seem to be under the misapprehension that most Windows users get infected through indirect means (ie through service worms)… Putting aside the fact Linux servers get worms through exploits this exact way, a huge number of infections are contracted through user idiocy: downloading what they think is a screensaver or a video Britney Spears blowing KFed.

If these idiots ever migrate, what do you suppose they’re going to do if they’re offered a screensaver as a .deb package? Yeah download, run, put their password in and hey! It has root access to their system. It could easily be a +x binary inside a zip with the instructions to run it.

The reason we’re not swamped in this crap is there isn’t the market to write it. As soon as it becomes viable our world gets a whole lot less secure.