Home Messages Index
[Date Prev][Date Next][Thread Prev][Thread Next]
Author IndexDate IndexThread Index

Re: [News] UNIX-type Systems Inherently More Secure Than Others (Windows)

Roy Schestowitz wrote:

> 
> Linux vs. Windows: Which is Most Secure?
> 

I wasn't convinced that he wasn't doing any more than matter of opinion
testing here. It could have been done in a much more convincing way I would
have thought.

Quallity score. Where did the values come from, he didn't really say. Of all
of the things the nonLinux world can say about Linux they can't really
criticise it on a quality level. It is based on a robust well controlled
kernel. Some applications have been poor, but the majority are themselves
robust. But this man is only talking about the OS, so why not an A+ across
the board, when was the last time anyone here had a problem that was caused
by the OS itself?

> ,----[ Quote ]
> | True to UNIX.
> | Qualitative score: Windows gets a D+ while Linux gets an A-.

Why True To UNIX, UNIX is obviously the bee's knees of OS's as far as
stability and quality is concerned. But it is slow moving. That is because
it is still based on the strong code footing of an OS, i.e. every single
function must be tested with good and bad data, it must behave well, all
possible paths in code must be tested. What ever the applications do the OS
must not crash.

Linux was following that course, I hope it still does, the exceptionally
strong kernel suggests that it is still true, but Linux was able to move
much more quickly than UNIX in the applications/utilities area.

Still this particular part of his article doesn't make sense, who would
suggest a mark of Windows trueness to UNIX? It's just daft.

> | "Bummer of a birthmark"
> | Qualitative score: Windows gets an F while Linux gets an A.

This is probably the truest part, Windows has a bad press, even some of it's
own users would like to see MS fall, just to see what happens. But it is
also a bullshit mark, because had Vista been a great product it would have
flown off the shelves, it would have been adopted quickly by many a
business.

It wasn't a great product, so the Deer with a target on it's bum, as he
said, painted that target on itself.

> | User data confidentiality.
> | Qualitative score: Windows gets a B- while Linux gets a B+.

This encrypted /home that seems to be the latest fashion, it's pants. If
users are set up properly it doesn't matter that the user area isn't
weighted down with encryption. 

Like many others here, I have saved many a data area from a damaged drive, I
was only able to do that because the data wasn't encrypted.

The important part here is not that the data area isn't encrypted, but where
servers tend to be placed. Servers are meant to be locked away, in server
rooms. But very often these days you visit an office and see a server out
on, or under, the desk just like any other computer in the office. That is
the only weakness from nonencrypted file systems, that the thief can simply
pick up the server and walk out of the door.

Put the servers away in a safe place, with limited access, and in Linux
there is no risk.

> | Patch practices
> | Qualitative score: Windows gets an A- while Linux gets a B+.
> `----
> 

How much easier can updates be made. Whether you have a coporate server or a
home computer, is there a version of Linux distro in the top 20 or so that
doesn't allow for automated updates? I don't know of one.

For the Redhats and the SLES users you can also automate, or you can pick
and choose, whether at home or work you always have access to review the
updates before you allow them. They run so smoothly that personally I think
that it would be absolutely safe to allow them on the Live server in normal
working hours. Ok, I wouldn't actually do that, but it is so long since I
had an update go bad that it would be very easy to become complacent and
just let them update any time of the day.


So in the end I really don't think Kenneth did a particularly good job on
this comparison. He didn't go anywhere near deeply enough to convince
anyone when they was obviously so much scope for a much more believable
comparison.


[Date Prev][Date Next][Thread Prev][Thread Next]
Author IndexDate IndexThread Index