Suggestion

[Chad Perrin]

On Sat, Mar 10, 2012 at 11:31:33PM +0700, Erich Dollansky wrote:
> On Saturday 10 March 2012 22:08:37 Alejandro Imass wrote:
> > 
> > ALL of Windows' problems are precisely based on poor design... just to
> > name a few:

Actually, I disagree with this statement.  Many of MS Windows' problems
are a result of poor management, too.  For instance, the policy of hiding
(known) grave security issues for years, and of launching smear campaigns
against security researchers who get tired of waiting for Microsoft to do
anything about such grave vulnerabilities and thus publish information
for end users to use in making technology decisions and trying to
mitigate their exposure, adds up to a whole lot of problem for MS
Windows, too.

It's certainly true that a lot of problems are based on poor design,
though.


> > 
> > - no clean separation of system and apps
> 
> it is very clearly separated.

Perhaps you can explain the pervasive spread of IE's tentacles throughout
the system for much of the lifetime of the MS Windows family of operating
systems, then.


> 
> > - apps re-write system libs at will
> 
> Isn't this another masterpiece FreeBSD is far off achieving?

I'm not aware of any cases where installing or firing up an editor, web
server, or mail user agent alters base system libraries.  I think you are
mistaken.


> 
> > - no lib versioning
> 
> I think that you are wrong here. It a long time ago but I think I
> remember they put a version number into the library name.

I read "no lib versioning" as meaning "we don't get the same support for
being able to use multiple versions of a library for different purposes,"
but maybe I'm mistaken.


> 
> > - there is not out of the box user / admin separation
> 
> Another point where FreeBSD is far behind. It is not possible to give
> every user on FreeBSD its own account and full administration rights.

1. Plan 9: some kind of next generation rights management and privilege
separation

2. FreeBSD: architectural privilege separation between user accounts

3. MS Windows: user-level restrictions on what users can do, trivially
bypassed by DRM software and malicious code

I think the way you try to paint situations 2 and 3 as being equivalent
is grossly off the mark.


> 
> > - no filesystem-based security
> 
> FAT rules!

Uh . . . what?


> 
> > - default network protocols are insecure
> 
> Windows has meanwhile default network protocols? I think, I have to do
> some catching up.

I suspect this was a reference to things like SMB/CIFS and other common
networking protocols and toolsets on MS Windows systems.

-- 
Chad Perrin [ original content licensed OWL: http://owl.apotheon.org ]