> The folks at Google are smart, but they're not experts at everything.
That might be the case, but you can bet they are experts at Subversion. C'mon -- Greg Stein, Ben Collins-Sussman, Brian Fitzpatrick?! These guys helped to start the Subversion project. And as former CollabNet employees, they know a thing or two about hosted, integrated developer tools, too.
> However, how much is that increased security really worth for an average home > user, when you break it down? According to me, fairly little
Hrm. When I saw this in the summary, I was expecting an entirely different sort of article. See, I think the "worth" of all that extra security (let's assume for a moment that Unix/Linux security > Windows security) involves more than just the salvation your data might see. There are whole other vectors of cost involved, and the informed consumer of an operating system needs to think about all of those vectors.
My brothers have recently been begging me to install Linux on their computers. Long-time Windows users, and completely ignorant of Linux/Unix-type systems, they come to me from a standpoint of fear: omigosh, if we hear of another Windows vulnerability we'll just die!
Now, I'm a big fan of Linux -- use it every day of my life to Get Stuff Done -- but I cannot in good conscience recommend this to them. Why not? Well, let's flip a few more pages in the story, to the night one of my brothers convinced me that no, he was sure, he wanted Linux.
The Fedora Core 4 install went pretty well[1] -- no glitches to speak of; the system came up as expected. But in moments, we found ourselves in trouble. There was no driver for my brother's rather common WiFi card, and his second video card was acknowledged but not being used. The provided tools didn't grok MP3 or WMA formats out-of-the-box. A large chunk of the user experience remains a thin veil over command-line tools with a thousand options. And so on.
This particular brother still thinks he has the guts to drive ahead with this, but I know him better than he knows himself. In a matter of weeks, he'll realize that all that extra security amounts to a system that *hurts him* to use. And that will be that. FDISK. Setup.exe. He has his life back.
(Turns out there's an ndiswrappers RPM I could plop onto my USB JumpDrive -- along with RPMs for a kernel upgrade to match the ndiswrappers kernel module -- and then install on his box -- that took a few hours of running up and down the stairs from networked box to not-so-networked-box to accomplish. All is well, now. I still haven't solved the dual monitor thing. I can't even really make myself care.)
My point is this: an article like this is just noise (and in this case, poorly executed noise, even). Arguments and debates in response to it citing whose OS can kick who else's OS's behind are equally noise. The decision of what operating system to invest time and energy into running calls for broader consideration, and at least today, calls for recognition by all parties that all things are not equal. Some systems better suit entirely different swaths of the user base; none is all all things to all people.
Slashdot Mirror
> The folks at Google are smart, but they're not experts at everything.
That might be the case, but you can bet they are experts at Subversion. C'mon -- Greg Stein, Ben Collins-Sussman, Brian Fitzpatrick?! These guys helped to start the Subversion project. And as former CollabNet employees, they know a thing or two about hosted, integrated developer tools, too.
> However, how much is that increased security really worth for an average home
> user, when you break it down? According to me, fairly little
Hrm. When I saw this in the summary, I was expecting an entirely different sort of article. See, I think the "worth" of all that extra security (let's assume for a moment that Unix/Linux security > Windows security) involves more than just the salvation your data might see. There are whole other vectors of cost involved, and the informed consumer of an operating system needs to think about all of those vectors.
My brothers have recently been begging me to install Linux on their computers. Long-time Windows users, and completely ignorant of Linux/Unix-type systems, they come to me from a standpoint of fear: omigosh, if we hear of another Windows vulnerability we'll just die!
Now, I'm a big fan of Linux -- use it every day of my life to Get Stuff Done -- but I cannot in good conscience recommend this to them. Why not? Well, let's flip a few more pages in the story, to the night one of my brothers convinced me that no, he was sure, he wanted Linux.
The Fedora Core 4 install went pretty well[1] -- no glitches to speak of; the system came up as expected. But in moments, we found ourselves in trouble. There was no driver for my brother's rather common WiFi card, and his second video card was acknowledged but not being used. The provided tools didn't grok MP3 or WMA formats out-of-the-box. A large chunk of the user experience remains a thin veil over command-line tools with a thousand options. And so on.
This particular brother still thinks he has the guts to drive ahead with this, but I know him better than he knows himself. In a matter of weeks, he'll realize that all that extra security amounts to a system that *hurts him* to use. And that will be that. FDISK. Setup.exe. He has his life back.
(Turns out there's an ndiswrappers RPM I could plop onto my USB JumpDrive -- along with RPMs for a kernel upgrade to match the ndiswrappers kernel module -- and then install on his box -- that took a few hours of running up and down the stairs from networked box to not-so-networked-box to accomplish. All is well, now. I still haven't solved the dual monitor thing. I can't even really make myself care.)
My point is this: an article like this is just noise (and in this case, poorly executed noise, even). Arguments and debates in response to it citing whose OS can kick who else's OS's behind are equally noise. The decision of what operating system to invest time and energy into running calls for broader consideration, and at least today, calls for recognition by all parties that all things are not equal. Some systems better suit entirely different swaths of the user base; none is all all things to all people.