Very Computer

Quote:> --
> Why Windows NT Server 4.0 continues to exist in the enterprise would
be a
> topic appropriate for an investigative report in the field of
psychology
> or marketing, not an article on information technology.
>      - John Kirch, Microsoft Certified Professional (Windows NT)

But I'll try to question author's technical knowledge of Windows NT.

First, there are some significant differences between NT version 3.51
(for which
the author has taken two exams and has the degree of Microsoft Certified

Professional) and NT version 4.0 (for which I've taken six exams and
have the
degree of Microsoft Certified Systems Engineer, as well as
MCP+Internet).

Good example is the argument for NT's "poor" shell scripting. MS
introduced very
powerful native command line (CMD.EXE) scripting with version 4.0
(missing in
version 3.51). I use this every day and there haven't been a task I
wasn't able
to execute using only the scripting language and the Resource Kit
utilities.
Let's mention some basic features of NT 4.0 shell scripting such as
string
tokens processing, numeric loops, arithmetical expressions.

Another example is the execution of applications on servers from remote
workstations (not the case of client/server apps)- Kirch claims this as
missing
for NT.
But in Windows NT 4.0 Resource Kit MS introduced "Remote Console". You
install
it as service on the "server" (actually you can install it on both NT
4.0
workstation and server). Using the client part, user can connect to the
server
and be on the server's shell. Then from the shell the user can any
program (even
GUI as far as it doesn't require remote user's intervention).
Simultaneous
connections from several workstations are possible as well.

Kirch's paper is updated on June 26 1998. But on June 17 MS announced
the
release of Terminal Server after almost a year in the last beta. It
provides
"X-window" type capability (not using X-window protocol) to the NT 4.0
server.
There is also third party tool like this that has been on the market for
long
time - Citrix's WinFrame.
Using one of these a user can connect even from UNIX workstation
(provided there
is a client for the specific UNIX flavor)

On the other hand, the question is not so simple as whether "Windows NT
Server
4.0 is no match for any UNIX operating system". There are many other
"matches"
in the real world one have to meet when choosing complex things, not
only
technical. It's a complex factor including the support, availability,
price,
training, performance. NT might be not the best in any of them, but when
you
calculate the "average" apparently it's one of the best for a specific
range of
tasks.
Definitely there are many features UNIX has NT server lacks, but stating
that NT
is not even close to UNIX is giving too much credit to some option
because you
don't know well enough the other.

At the end, I'd like just to add also that it's pretty strange to find
examples
with Novell in Kirch's paper. I don't think there is computer
professional who
would even consider Novell an "operating system" at all. Just one point
is
enough: it is this year Novell will release for first time protected
memory
kernel - something considered one of the most important "musts" of a
real
operating system. Something that has been implemented in Unix and NT (as
well as
all other OS-es) from the very beginning.
If there is something that does only file and print sharing, that's
Novell.

Kras Gadjokov,

MCSE, MCP+I (Windows NT 4.0)
M.Sc. in Computer Science

http://www.interlog.com/~kras

--
************************
* Enjoy the pane-Run NT*
************************

FWIW, I don't like that paper a whole lot either.  It's not very
scientific.  An objective paper should not start with "Unix is better"
as the conclusion and fill in the data to fit the conclusion as this
one does.

The kirch.net paper is cited at every mention of comparing Unix to NT.
I would like to see more articles.  Objective case studies are the
most interesting to me personally.

Lastly, I prefer Unix.  I have to say that even if NT were technically
superior, I would still prefer to use Unix.  I don't feel that I have
to justify my use of Linux with charts and graphs.  It works well for
what I need it for and has a lot of software available.

-Kevin

: release of Terminal Server after almost a year in the last beta. It
: provides
: "X-window" type capability (not using X-window protocol) to the NT 4.0
: server.

        Of course one has to wonder why they didn't just use the X
protocol.. they wouldn't have had to write ANY servers for the Unix
workstations to support it.  So, where can I download Terminal Server?  

: At the end, I'd like just to add also that it's pretty strange to find
: examples
: with Novell in Kirch's paper. I don't think there is computer
: professional who
: would even consider Novell an "operating system" at all.

        Same could be said about Win95. :-)    

--
-----------------------------------------------------------------------

Blinky lights are the essence of  |  Spam will be auto deleted
modern technology!                |  Caffeine underflow (brain dumped)

--
Tracy Reed      http://www.ultraviolet.org
Linux is harder to learn than Windows. But it is easier to use.

My biggest complaint is that there's a lot of Netware administration
that you cannot effectively do from inside of Netware - you have to do
it from tools that run on Windows or Macs (Nwadmin etc.)  This is
changing a *little* bit with Novell's new Java admin tools, but those
are so slow and clunky that I consider them completely unusable in
their current form.

For Netware to fit my definition of a complete OS, it would have to be
usable completely in and of itself, without support of other
environments.  Yes, I guess it's within the realm of potential that
someone *could* write an NLM word processor someday, but I don't see
that happening...  :-)

Here is how:
IF windows 95 is an OS then bring up your favorite shell. and type this
command:

deltree c:\dos

Then reboot your PC and tell me if it comes back up. Bet it doesn't
that's becaue winblows 95 is ONLY a GUI over the REAL OS, DOS. not that
that is a real OS either.

--

Richard Knechtel
email(richard dot knechtel at eds dot com)
EDS
(Systems Engineer/System Administrator)
(Aspiring AS/400 GURU)
(Aspiring Linux GURU)
(Aspiring Visual Basic Programmer)

       The contents of this message express only MY opinion.
       This message does not necessarily reflect the policy or views of
       my employer, EDS.  All responsibility for the statements
       made in this posting resides solely and completely with the
       ME.
       I Ex-Spaminate spammers!
       See US Code Title 47, Sec.227(a)(2)(B), Sec.227(b)(1)(C)
       and Sec.227(b)(3)(C).

--

Richard Knechtel
email(richard dot knechtel at eds dot com)
EDS
(Systems Engineer/System Administrator)
(Aspiring AS/400 GURU)
(Aspiring Linux GURU)
(Aspiring Visual Basic Programmer)

       The contents of this message express only MY opinion.
       This message does not necessarily reflect the policy or views of
       my employer, EDS.  All responsibility for the statements
       made in this posting resides solely and completely with the
       ME.
       I Ex-Spaminate spammers!
       See US Code Title 47, Sec.227(a)(2)(B), Sec.227(b)(1)(C)
       and Sec.227(b)(3)(C).

I think we can agree that the kirch.net report was not "even-handed"
objective.

That being said, are you saying that "even-handed" objectivity is better
than "truthfull" objectivity, or are you saying that the report was not
objective in either sense of the word- that it, and the references he
supplied, are apriori biased?

One of the reasons that report comes up as often as it does is that it is
a clearing house for links to all other peices of information that
supports the arguments we're advancing.

Brian

Quote:>YUP! Win 95 IS NOT an OS and the Winblows lovers can be proved wrong in
>one swoop!

>Here is how:
>IF windows 95 is an OS then bring up your favorite shell. and type this
>command:

>deltree c:\dos

>Then reboot your PC and tell me if it comes back up. Bet it doesn't
>that's becaue winblows 95 is ONLY a GUI over the REAL OS, DOS. not that
>that is a real OS either.

     Deleting c:\command.com will have the effect you're looking for.
(But then again, deleting any critical file will, too.)

--

WYSIWYG - What You See Is What You Get
WYGIWYD - What You Get Is What You Deserve

Quote:>Good example is the argument for NT's "poor" shell scripting. MS
>introduced very
>powerful native command line (CMD.EXE) scripting with version 4.0
>(missing in
>version 3.51). I use this every day and there haven't been a task I
>wasn't able
>to execute using only the scripting language and the Resource Kit
>utilities.
>Let's mention some basic features of NT 4.0 shell scripting such as
>string
>tokens processing, numeric loops, arithmetical expressions.

Quote:>Another example is the execution of applications on servers from remote
>workstations (not the case of client/server apps)- Kirch claims this as
>missing
>for NT.
>But in Windows NT 4.0 Resource Kit MS introduced "Remote Console". You
>install
>it as service on the "server" (actually you can install it on both NT
>4.0
>workstation and server). Using the client part, user can connect to the
>server
>and be on the server's shell. Then from the shell the user can any
>program (even
>GUI as far as it doesn't require remote user's intervention).
>Simultaneous
>connections from several workstations are possible as well.

I don't think you have come anywhere near answering Kirch's points.  True,
some of his points may be a little outdated, but the spirit of his letter is
correct.

I don't think that you have discreddited his crudentials at all.

Rob Hall

www.hcm.iinet.net.au

On Tue, 30 Jun 1998 10:51:13 -0400, Krassimir Gadjokov

I am a NT Admin with two tests to go for MSCE, and I have to tell you
I do ANYTHING that needs scripting in PERL, the CMD is only good for
running a simple few commands one after another.

Rgrds,
DH

Brian> That being said, are you saying that "even-handed" objectivity
Brian> is better than "truthfull" objectivity, or are you saying that
Brian> the report was not objective in either sense of the word- that
Brian> it, and the references he supplied, are apriori biased?

That they are a priori (presumed).  The paper isn't "I set out one day
to find out whether Unix or NT is better", it's "I believe Unix is
better and here is why".  That's all fine and good, but not objective
in the scientific sense, by which I mean seeking the truth.  (The
truth is assumed: Unix is superior).  Note that I don't think Kirch
intended it to be an objective analysis.

There are also many flaws -- the cost analysis is simplistic, the
discussion on Apache is a distraction, the blurring of client
vs. server issues, outdated criticisms, etc.  I don't think like
a manager though either :-).

Brian> One of the reasons that report comes up as often as it does is
Brian> that it is a clearing house for links to all other peices of
Brian> information that supports the arguments we're advancing.

To some extent.  Linux needs mind share and marketing, it's not technical
superiority that sells NT, or any flavor of Windows for that matter.

The paper by John Kirch is certainly not without merit, and it is a
good effort.  It's not the whole picture though.  There is one quote
that I like however:

 "The trend seems to be to have the "dummies" be the administrators."
        --Tom Moore

This seems very true from my experience.  The industry is becoming
watered down - why hire one expensive admin when you can hire ten
cheap mediocre ones that will stand around and say "I love Microsoft.
You're such a clever manager for following the herd."

-Kevin

--

Richard Knechtel
email(richard dot knechtel at eds dot com)
EDS
(Systems Engineer/System Administrator)
(Aspiring AS/400 GURU)
(Aspiring Linux GURU)
(Aspiring Visual Basic Programmer)

       The contents of this message express only MY opinion.
       This message does not necessarily reflect the policy or views of
       my employer, EDS.  All responsibility for the statements
       made in this posting resides solely and completely with the
       ME.
       I Ex-Spaminate spammers!
       See US Code Title 47, Sec.227(a)(2)(B), Sec.227(b)(1)(C)
       and Sec.227(b)(3)(C).