Serious investigations into UNIX and Windows

[Giorgos Keramidas]

On 2004-10-26 07:42, "Butterworth, Thaddaeus (UI Exploratory)" <thad.butterworth at hp.com> wrote:
> I work in a testing environment where I have set up both Windows and
> *nix type servers. The first time I set up a server it was Exchange 2003
> on Windows Server 2003. I was able to figure out how to securely set it
> up within two hours. On the other hand, setting up ldap on FreeBSD took
> me two days. All of these needed to connect to various computing
> platforms, including the embedded systems (using LYNX) that I was
> testing. I've worked with Windows, *nix, and Mac OS. I've found Mac to
> be the easiest to work with, Windows second easiest, and the *nix take
> far more skill than the other two combined. Part of the issue that you
> are facing from your description of the complications with Windows,
> comes from trying to make windows do what windows was not designed to
> do.

> I don't care what Bill Gates says, none of the windows server environments
> were ever designed with anything more than simple, small networks in
> mind. It's part of the culture of MS. They started out with personal
> computing systems, and then decided that they would get into the server
> market. They inherently approach all software from a personal computing
> standpoint. That's why there are so many "undocumented" procedures to make
> things work the way that they are supposed to.

Nonsense, if you ask me.  For many reasons:

a. Windows doesn't work nicely even for small networks most of the time.

It's not the size of the network that matters.  It's the nature of the
network.  Homogeneous, Windows-only networks will usually work somehow;
not optimally, mind you, but they can be coerced into working.
Heterogeneous networking environments, with many different types and
versions of operating systems, are not so easy to use from Windows.

b. The small-network culture has nothing to do with documentation.

Undocumented stuff is undocumented because Microsoft either didn't have the
time to document them all (rushing a new release out to gather a few more
billion dollars) or -- more importantly -- they don't _want_ them
documented, to have an edge over the rest of the software developers.

> It's not really a matter of what is better for everybody, but what is
> better for the context that you are working under. I've recommended both
> Windows and *nix solutions to people. It just depends on who I am talking
> to. It's the same thing with this subject. I cannot and will not
> emphatically state that one OS is better than the other. I can tell you
> which I prefer, but you have to look at the needs of the individual or
> company and try to determine the right solution from there. If you are
> having to mess around with undocumented procedures and do all this extra
> junk just to secure your windows servers, then I would say you need to
> take a serious look at changing your server OS.

I mostly agree.  Then, one day, eventually and also pretty unavoidably,
freedom suddenly matters.

That's about the same time that Windows starts to feel uncomfortable, with
all its undocumented "lock-in" stuff whose only purpose is not to make
computing easier but to make more money for Microsoft.  If it so happens
that some part of the every day experience of the average user is also made
easier, it's a happy coincidence in the Microsoft world; not the Ultimate
Goal(TM), but not unwelcome either ;-)

But this thread reminds me of far too many threads that I've seen this
topic discussed to death and beyond, some of them on this list too.  So
I'll stop writing.

</rant>

- Giorgos