HELP: Need reasons to use Linux / Unix (Please Read)

[Julian Regel]

I'm currently doing a degree in Computing and want to persuade my
college to do more Unix / Linux. Although I'm no expert, I have played
with Linux a bit and think that it would be extremely beneficial to
learn it. The problem is convincing the Field Board.

Basically, I need help from some Unix / Linux experts. Please give me
reasons why we should learn it. I know Unix experience is required in
a lot of jobs, but I don't have any statistics or comments from
potential employers. There must be some people that can provide
information out there. Please email if you can help. I've been asked
to bring the topic up at the next meeting on the 30th, and I would
like some ammunition :-)

Many thanks,

Julian
---
+---------------------------+------------------------------------------+
| Julian Regel (JSAS) : | The Unofficial Julia Sawalha Homepage : |
| jre...@melech.demon.co.uk | http://www.geocities.com/Hollywood/3465/ |
+---------------------------+------------------------------------------+

[Josh Stern]

Julian Regel <jre...@melech.demon.co.uk> wrote:
>I'm currently doing a degree in Computing and want to persuade my
>college to do more Unix / Linux. Although I'm no expert, I have played
>with Linux a bit and think that it would be extremely beneficial to
>learn it. The problem is convincing the Field Board.
>
>Basically, I need help from some Unix / Linux experts. Please give me
>reasons why we should learn it. I know Unix experience is required in
>a lot of jobs, but I don't have any statistics or comments from
>potential employers. There must be some people that can provide
>information out there. Please email if you can help. I've been asked
>to bring the topic up at the next meeting on the 30th, and I would
>like some ammunition :-)

In the case of Linux, one obtains freely available source code for
99% of everything that runs on the system. There are also
tons of well written development libraries and tools in
freely available code. If part of the mission of your university
is to teach people how to code and how computers work, then
tell them that Linux is one of the best tools available for
that purpose because looking at real (not toy) source code and
figuring out how and why it works is an essential part of
learning about computing and programming.

In contrast, what would the students do with, say, Windows NT out of
the box? Hone their solitaire/minesweeper skills?
For the sake of argument, suppose the Field Board that you
are dealing with thought it would be more beneficial for
their students to master WIN32 programming. If so they
need to realize that they wouldn't be making any progress
towards that goal by just obtaining the OS. They would
also need to provide development tools and documentation,
which involves the acquisition of many separate commercial
products. If the computers are networked, they would also
need many additional products to provide a comparable
level of networking.

You could also inform them that most academic CS researchers
work with Unix and frequently share source code with their
colleagues that requires a Unix environment.

- Josh

--
-------------------------------------------------------------------------------
jstern
jst...@primenet.com
-------------------------------------------------------------------------------

[Chris Petit]

In article <4o810m$s...@nnrp1.news.primenet.com>, jst...@primenet.com (Josh Stern) writes:
|> Julian Regel <jre...@melech.demon.co.uk> wrote:
|> >I'm currently doing a degree in Computing and want to persuade my
|> >college to do more Unix / Linux. Although I'm no expert, I have played
|> >with Linux a bit and think that it would be extremely beneficial to
|> >learn it. The problem is convincing the Field Board.
|> >
|> >Basically, I need help from some Unix / Linux experts. Please give me
|> >reasons why we should learn it. I know Unix experience is required in
|> >a lot of jobs, but I don't have any statistics or comments from
|> >potential employers. There must be some people that can provide
|> >information out there. Please email if you can help. I've been asked
|> >to bring the topic up at the next meeting on the 30th, and I would
|> >like some ammunition :-)

[ excellent reasons to persuade university to use Linux deleted ]

Also, don't forget that there is a large community of Linux users
who know a great deal and offer informal "support" via the net at no cost.

For Win NT or 95 newsgroups (ALL is strictly IMO), possibly more
users there have less in-depth experience with solving problems on this
OS'es.

And, that the resource requirements are similar to Win 95 --- not
too hefty for X-Windows included.

Also, if programmers can handle X-Windows programming smoothly,
using C++, or even XLib, they have an excellent handle on how to program
Win NT and 95 (although both are MUCH more kludgy API's IMO).

[Todd Needham [Microsoft]]

Julian,

All of the answers you're getting here are good (though not completely
informed). But please don't think of Linux vs. Windows as an either/or
situation. There is certainly room for both in universities today.

Taking the pro-Win32 side and building on the answers others gave you
regarding Linux/UNIX:

1. Source code for Windows NT is also available to universities at no
charge (currently 29 universities world wide are source code licensees and
we're rapidly expanding that).

2. There are a greater number of development tools available for Win95/NT
than Linux, covering a greater breadth of development needs. Certainly
the size of the Windows development market and it's sheer competitiveness
have delivered tool sets second to none.

3. Out of the box, NT (like Linux) contains a full networking environment
(including TCP/IP with sockets and Novell compatible IPX/SPX). Nothing
you need to add for networking unless you need NFS (available from a
number of vendors).

4 Add the compiler or development tool set of your choice (one of ours,
Borland's, Symantec's, the GNU ports, etc.) and you're up and running.

5. Microsoft and other tools vendors offer significantly discounted
educational pricing on their development tools. We also offer the same on
our OS's. Documentation for the OS and for our development tools is
available in electronic (read: full text searchable) form through the
Microsoft Developer Network.

6. Newsgroup support. Not only are there extensive usenet newsgroups on
'NT and Win95 use and programming, but Microsoft recently moved it's
CompuServe forums over to our own news servers -
news://msnews.microsoft.com

You and the others who responded are correct that today the lion's share
of university research occurs on UNIX variants. However, the commercial
world runs on Windows. It's a disconnect that I hope to see resolved....

-Todd

[Bryan Seigneur]

Todd Needham [Microsoft] wrote:
>
> Julian,
>
> All of the answers you're getting here are good (though not completely
> informed). But please don't think of Linux vs. Windows as an either/or
> situation. There is certainly room for both in universities today.
>
> Taking the pro-Win32 side and building on the answers others gave you
> regarding Linux/UNIX:
>
> 1. Source code for Windows NT is also available to universities at no
> charge (currently 29 universities world wide are source code licensees and
> we're rapidly expanding that).
>
> 2. There are a greater number of development tools available for Win95/NT
> than Linux, covering a greater breadth of development needs. Certainly
> the size of the Windows development market and it's sheer competitiveness
> have delivered tool sets second to none.

I thought Unixes had some spectacular (and spectacularly expensive) tools.
NT v. Linux starts to sound like Comm. Unix v. Linux here. How immense
can the Win32 tool market be compared to others, given Win32 is so
young?

> 3. Out of the box, NT (like Linux) contains a full networking environment
> (including TCP/IP with sockets and Novell compatible IPX/SPX). Nothing
> you need to add for networking unless you need NFS (available from a
> number of vendors).

Or telnet, or _X-window_, or a thousand other things you take for granted in
Unix/Linux.

> 4 Add the compiler or development tool set of your choice (one of ours,
> Borland's, Symantec's, the GNU ports, etc.) and you're up and running.
>
> 5. Microsoft and other tools vendors offer significantly discounted
> educational pricing on their development tools. We also offer the same on
> our OS's. Documentation for the OS and for our development tools is
> available in electronic (read: full text searchable) form through the
> Microsoft Developer Network.
>
> 6. Newsgroup support. Not only are there extensive usenet newsgroups on
> 'NT and Win95 use and programming, but Microsoft recently moved it's
> CompuServe forums over to our own news servers -
> news://msnews.microsoft.com
>
> You and the others who responded are correct that today the lion's share
> of university research occurs on UNIX variants. However, the commercial
> world runs on Windows. It's a disconnect that I hope to see resolved....

Installed commercial base and existing applications base is MS's only
advantage. I used to think MS had some other advantages. But real free
Unix changes the rules.

Most of the extremely plural and competitive commercial software industry
is dependent (read controlled) by one firm. It's a disconnect I (and
thousands of others) hope to see resolved.

> -Todd

Bryan
--
Y W8 4MS?
LINUXNOW!

[drsoran]

Todd Needham [Microsoft] (to...@microsoft.com) wrote:
: 3. Out of the box, NT (like Linux) contains a full networking environment

: (including TCP/IP with sockets and Novell compatible IPX/SPX). Nothing
: you need to add for networking unless you need NFS (available from a
: number of vendors).

How are NT's out of the box DNS and http servers? Do you guys
have any discounts where a student can obtain NT on a CD for $7? How
about $50? I was flipping through an Egghead catalog and they had a
really hilarious typo.. they had NT Server listed for $700.. sheesh..
they should proof read those things before they print them. Oh well... I
guess I'll have to settle for Linux until Egghead gets that price fixed.. ;)

: 5. Microsoft and other tools vendors offer significantly discounted

: educational pricing on their development tools. We also offer the same on
: our OS's. Documentation for the OS and for our development tools is
: available in electronic (read: full text searchable) form through the
: Microsoft Developer Network.

If the source is freely available to NT, why don't more
universities have it? Do they have to indenture 10% of their grad
students to become slaves at Microsoft for 10 years or something? Why
not put the source code on an ftp site? Oh.. I forgot.. you guys are
into licenses..

: You and the others who responded are correct that today the lion's share

: of university research occurs on UNIX variants. However, the commercial
: world runs on Windows. It's a disconnect that I hope to see resolved....

I do as well.. Keep up the good work.. I'm sure within the next
10 years you'll have the commercial world running back to UNIX..

--
----------------------------------------------------
drs...@ni.cba.csuohio.edu
M$-Win95 user: "Why is this running so slow today?"

[Alexandre Maret]

hello

I find it quite strange, people @microsoft.com read
the linux newsgroups !

I've some questions about this :

- is there employees at microsoft whose job is "newsreader",
or were you browsing comp.os.linux.advocacy for your
pleasure...

- what do microsoft think about linux (not you, microsoft)

- will microsoft release Internet Explorer for Linux

- is there a lot of microsoft employee running linux at home...
(yes... there may be some... I've seen credits...)

- did microsoft already tried to hire linus...

thanks in advance !

alex

[Joe Sloan]

>Todd Needham [Microsoft] wrote:

>> You and the others who responded are correct that today the lion's share
>> of university research occurs on UNIX variants. However, the commercial
>> world runs on Windows. It's a disconnect that I hope to see resolved....

I agree with you totally -

Let's be more aggressive about promoting Unix/Linux for business!

--
Joe Sloan - I have seen the future and it is Linux -
j...@netvoyage.net <a href="http://www.netvoyage.net/~jjs">

[A Shelton]

"Todd Needham [Microsoft]" <to...@microsoft.com> writes:

>Julian,

>All of the answers you're getting here are good (though not completely
>informed). But please don't think of Linux vs. Windows as an either/or
>situation. There is certainly room for both in universities today.

This is true..

>Taking the pro-Win32 side and building on the answers others gave you
>regarding Linux/UNIX:

>1. Source code for Windows NT is also available to universities at no
>charge (currently 29 universities world wide are source code licensees and
>we're rapidly expanding that).

29 universities? wow... So the universities are allowed to run modified
windows NT? can they distribute their improvements? can they sell the
improvements to MS....

Perhaps we can get linus a copy...Certainly the Wine people would be
delighted.

I get the feeling they're under lock and key though.

>2. There are a greater number of development tools available for Win95/NT
>than Linux, covering a greater breadth of development needs. Certainly
>the size of the Windows development market and it's sheer competitiveness
>have delivered tool sets second to none.

Admittedly they are subject to change without notice, cost a fortune and
all rely on a proprietary single source. I noticed the windows programmers
weren't delighted with you when you dumped win32(some letter) recently.

>5. Microsoft and other tools vendors offer significantly discounted
>educational pricing on their development tools. We also offer the same on
>our OS's. Documentation for the OS and for our development tools is
>available in electronic (read: full text searchable) form through the
>Microsoft Developer Network.

At a mere 800 dollars for the priviledge of getting the Urdu version of
windows NT :)

How are microsoft compilers on cross compiling by the way? Surely they
have at least as many targets as say, gcc?

Still, I haven't seen anyone offering discounts on Linux development
tools.

>6. Newsgroup support. Not only are there extensive usenet newsgroups on
>'NT and Win95 use and programming, but Microsoft recently moved it's
>CompuServe forums over to our own news servers -
>news://msnews.microsoft.com

hehehe.....can I get it on MSN too?

>You and the others who responded are correct that today the lion's share
>of university research occurs on UNIX variants. However, the commercial
>world runs on Windows. It's a disconnect that I hope to see resolved....

So do I, but perhaps not in the way you intend....

>-Todd

--
Linux, because raw power can be....addictive.
Andrew Shelton s940...@yallara.cs.rmit.edu.au
GCS(2.1)-d+H+sw+v-C++UL+>L+++E-N++WV--R++tv-b+D++e+fr*y?

[Todd Needham [Microsoft]]

> Bryan Seigneur <fr...@gramercy.ios.com> wrote in article
<31AE1765...@gramercy.ios.com>...

>
> I thought Unixes had some spectacular (and spectacularly expensive)
tools.
> NT v. Linux starts to sound like Comm. Unix v. Linux here. How immense
> can the Win32 tool market be compared to others, given Win32 is so
> young?

Age isn't the issue - market size is what's driving the phenomenal Windows
tools market. Simple capitalism.

> > 3. Out of the box, NT (like Linux) contains a full networking
environment
> > (including TCP/IP with sockets and Novell compatible IPX/SPX).
Nothing
> > you need to add for networking unless you need NFS (available from a
> > number of vendors).
>

> Or telnet, or _X-window_, or a thousand other things you take for
granted in
> Unix/Linux.

Yup. Left off X, sorry. "A thousand other things" might be an
exaggeration. Most of the GNU toolset has been ported and is available
from either Congruent Software or from the University of Texas microlab
ftp site.

> Installed commercial base and existing applications base is MS's only
> advantage.

You must believe this installed commercial base and incredible
applications supported happened by coincidence?

> I used to think MS had some other advantages. But real free
> Unix changes the rules.

Well, I agree that free UNIX is a good thing for end users anyways....

> Most of the extremely plural and competitive commercial software
industry
> is dependent (read controlled) by one firm. It's a disconnect I (and
> thousands of others) hope to see resolved.

hmmm...highly competitive commercial software industry....controlled by on
firm. You're right. We definitely have a disconnect....

-Todd

[Todd Needham [Microsoft]]

> How are NT's out of the box DNS and http servers?

The http server (available now on our web site) is excellent (hey, a
biased opinion for sure, but you can look at comparative reviews
yourself).

Peer web services (or personal web services) are being built into 'NT 4.0
(due this summer) as is a new DNS service with a graphical administrative
interface and WINS/DHCP integration.

> Do you guys
> have any discounts where a student can obtain NT on a CD for $7?

No, but they can obtain it for about $99 with documentation. Universities
can also purchase licenses without doc for about $50.

> How about $50?

See above.

> I was flipping through an Egghead catalog and they had a
> really hilarious typo.. they had NT Server listed for $700.. sheesh..
> they should proof read those things before they print them. Oh well...
I
> guess I'll have to settle for Linux until Egghead gets that price
fixed.. ;)

Price was probably right for Windows NT Server for corporates (not that I
believe we're talking about that in this particular thread).

> If the source is freely available to NT, why don't more
> universities have it?

Simple. We just started the program.

> Do they have to indenture 10% of their grad
> students to become slaves at Microsoft for 10 years or something?

First born child, sacrificed to you-know-who.

Actually, no. You probably missed this in my original post, but source is
licensed to universities for research purposes at no cost.

> Why not put the source code on an ftp site? Oh.. I forgot.. you guys
are
> into licenses..

Uhmmm... is that a point against us? Doesn't everyone (or 99.9%) who
sells software for a living license it? Perhaps I missed something as
regards Linux, but I thought that came very specifically with a license as
well.

> I do as well.. Keep up the good work.. I'm sure within the next
> 10 years you'll have the commercial world running back to UNIX..

I'm betting that in 10 years time UNIX will have less market share than it
does today (hey, I'm a shareholder ferchris'sake!)

-Todd

[Todd Needham [Microsoft]]

> >2. There are a greater number of development tools available for
Win95/NT
> >than Linux, covering a greater breadth of development needs. Certainly
> >the size of the Windows development market and it's sheer
competitiveness
> >have delivered tool sets second to none.
>
> Admittedly they are subject to change without notice

Yeah, it's a competitive market and tools are upgraded pretty regularly,
but I don't think this is a weakness, but a strength.

> cost a fortune

Assuming you're still on the same thread as I am (students, universities,
etc.), you'll discover that almost all the Windows tools vendors offer
huge academic discounts (including Microsoft)

> all rely on a proprietary single source. I noticed the windows
programmers
> weren't delighted with you when you dumped win32(some letter) recently.

I've no clue _what_ you're talking about. We haven't dumped Win32. Where
did you hear that one?!?

> >5. Microsoft and other tools vendors offer significantly discounted
> >educational pricing on their development tools. We also offer the same
on
> >our OS's. Documentation for the OS and for our development tools is
> >available in electronic (read: full text searchable) form through the
> >Microsoft Developer Network.
>
> At a mere 800 dollars for the priviledge of getting the Urdu version of
> windows NT :)

Hmmm. Lost me again. Academic pricing on Windows NT Workstation 3.51 is
about $99. Half that for universities who want to use it but don't need
the shrink wrap package.

> How are microsoft compilers on cross compiling by the way? Surely they
> have at least as many targets as say, gcc?

(nice lead) Our Visual C++ cross compiler for the Mac is selling quite
briskly, thank you.

> >6. Newsgroup support. Not only are there extensive usenet newsgroups
on
> >'NT and Win95 use and programming, but Microsoft recently moved it's
> >CompuServe forums over to our own news servers -
> >news://msnews.microsoft.com
>
> hehehe.....can I get it on MSN too?

Yes. And MSN is moving completely onto the Internet. You'll be able to
think of it as simply another resource on the 'net.

> >You and the others who responded are correct that today the lion's
share
> >of university research occurs on UNIX variants. However, the
commercial
> >world runs on Windows. It's a disconnect that I hope to see
resolved....
>
> So do I, but perhaps not in the way you intend....

Careful of that either/or trap...!

-Todd

[Todd Needham [Microsoft]]

> Alexandre Maret <ama...@infomaniak.ch> wrote in article
<31AF19...@infomaniak.ch>...

> hello
>
> I find it quite strange, people @microsoft.com read
> the linux newsgroups !

You'd be surprised...

> I've some questions about this :
>
> - is there employees at microsoft whose job is "newsreader",
> or were you browsing comp.os.linux.advocacy for your
> pleasure...

No, there's no one with that title. I follow this group because I run
into Linux occasionally in the universities I work with.

> - what do microsoft think about linux (not you, microsoft)

Hmmm. I don't speak for "Microsoft" in that sort of capacity, but to the
best of my knowledge, it's "not really on the radar screen", if you know
what I mean. I probably deal with it as much as anyone in an official
capacity - and I don't deal with it that much.

> - will microsoft release Internet Explorer for Linux

Nope.

> - is there a lot of microsoft employee running linux at home...
> (yes... there may be some... I've seen credits...)

I'm sure there are. We're certainly encouraged to experiment here.

> - did microsoft already tried to hire linus...

I'll defer to recruiting on that one!

-Todd

[Bob Nelson]

On Fri, 31 May 1996 23:31:53 -0700, Todd Needham [Microsoft] wrote:

>> > Bryan Seigneur <fr...@gramercy.ios.com> wrote in article
>> <31AE1765...@gramercy.ios.com>...

>> > Most of the extremely plural and competitive commercial software

>> industry
>> > is dependent (read controlled) by one firm. It's a disconnect I (and
>> > thousands of others) hope to see resolved.

>> hmmm...highly competitive commercial software industry....controlled by on
>> firm. You're right. We definitely have a disconnect....

Todd, you ignorant slut. Your heresy that you posted here will soon
be met with an onslaught of well-reasoned, articulate arguments that
are reserved for infidels such as yourself.

To assist in your much-needed re-education project, I now present
the "cola-composite", a digest of the typical Linux-loving,
mickeysoft-bashing advocate. It is hoped that this will help you
in finding the One True Way:

=========================================================================

LINUX, an unprecedented worldwide phenomenon, began in the sparkling
clean, sunny city of Helsinki, Finland by the erudite yet brawny LINUS
TORVALDS. The universally-acclaimed name of MR. TORVALDS is generally
spoken in the same reverential tone reserved for such esteemed figures
as Mother Theresa, Albert Einstein, Michael Jordan and Jordan Hubbard
and L. Ron Hubbard. The exponential growth and sweeping success of
LINUX is fueled by a worldwide community of developers, cheerfully
producing robust, error-free software without any thought of financial
gain. LINUX itself is a powerful 64-bit operating system (a few
versions are released for 32-bit platforms) known for its years of
trouble-free operation from Antarctica to the vast regions of outer
space.

By contrast, the micro$oft "corporation", based in the rain-drenched
shanty town of redmond, washington, hires hapless wage slaves to work
in sweatshop conditions to produce bug-ridden "software" such as
windoze "nt". The corporation's "chairman", shifty-eyed, pale-skinned,
anemic college-dropout bill gates, has been proven to be a member of
the same species that produced Adolph Hitler, Benedict Arnold, Lee
Harvey Oswald and Milli Vanilli. Empirical data show that gate's family
tree can be traced to Cain, son of Adam and Eve, who was found guilty
of murdering his own brother! Unlike the unquestioned purity of those
gladly contributing to the Righteous LINUX Cause, the micro$haft
"corporation" is constantly under government scrutiny for its shady
business dealings. Faced with a dwindling market share because of the
onslaught of the UNIX jauggernaut, micro$sloth is thought to bribe
unsavory PR flacks who then have their "articles" published in
PeeCee-oriented magazines.

Impartial statistics show LINUX is a considerable force that marketers
simply can't ignore. Over 30,000 LINUX CD-ROMS are shipped monthly with
each CD-ROM, then used by 11.3 others -- offering a monthly increase in
the LINUX universe of 339000. Extrapolating this figure along with
those from the 'net's LINUX Counter, it is estimated that there are now
82.7 million LINUX users. The sheer volume of big circulation
periodicals, including _LINUX Journal_ among many others too numerous
to mention is further proof of the explosive growth of LINUX.
Furthermore, there are at least 3.8 billion UNIX users worldwide --
counting those making use of a telephone -- since all telephone systems
are UNIX-based. They have to be! After all, UNIX developers Ken
Thompson and D.M. Ritchie worked for The Phone Company.

--
=============================================================================
Bob Nelson: Dallas, Texas, U.S.A. - bne...@netcom.com
Linux for fun, M$ for $$$...and the NFL for what really counts!
=============================================================================

[Bob Nelson]

On 31 May 1996 10:21:44 -0700, Joe Sloan wrote:
>> >Todd Needham [Microsoft] wrote:

>> >You and the others who responded are correct that today the lion's share
>> >of university research occurs on UNIX variants. However, the commercial
>> >world runs on Windows. It's a disconnect that I hope to see resolved....

>> I agree with you totally -

>> Let's be more aggressive about promoting Unix/Linux for business!

>> j...@netvoyage.net <a href="http://www.netvoyage.net/~jjs">

Joe...why not begin in your own backward with business promotion of
UNIX? It seems you have the power to make it happen.

Here are some quotes from http://www.toyota.com under the aegis of:

bnelson@netcom17$ whois toyota.com
Toyota Motor Sales, USA, Inc. (TOYOTA-DOM)
19001 Western Avenue
Torrance, CA 90509

Domain Name: TOYOTA.COM

Administrative Contact:
Broscow, David (DB990) bro...@TOYOTA.COM
(310)618-4287
Technical Contact, Zone Contact:
Sloan, Joseph (JS2446) j...@TOYOTA.COM
310-618-4029

Record last updated on 20-Mar-96.
Record created on 29-Dec-94.

Domain servers in listed order:

TOYOTA.TOYOTA.COM 206.3.39.6
NOVO1.NOVOMEDIA.COM 204.162.123.10
SEC1.DNS.PSI.NET 38.8.92.2
SEC2.DNS.PSI.NET 38.8.93.2

"Shockwave plug-in, you can play special animated files in your Mac
or Windows Netscape 2.0 window. UNIX users might have to wait for
a solution".

* So, joe, why treat "UNIX users" as second class citizens who have to
"wait for a solution". The solution is simply not to offer the service
until it available for UNIX. The web page needs a typo corrected since
an uppercase 'W' was used in the spelling of "windows." Your site might
gain extra friends if you used the c.o.l.a.-preferred spellings of
"winblows" or "windoze".

"Get in the driver's seat of every Toyota vehicle -- virtually
speaking. Our free PhotoBubble Viewers give you an inside
perspective in both Macintosh or PC platforms."

* But, joey, you have a typo above: You've taught us well in this
newsgroup that it is "PeeCee". Among the PeeCee platforms for
PhotoBubble at www.toyota.com are *none* of the UNICES. As "Technical
Contact", are you now empowered to change this short-sighted policy?
Do you have PhotoBubble support for a real graphics platform like
the SGI?

"Electronic media

Please indicate (*) Windows or ( ) Macintosh.
( )'96 Toyota CD-ROM (the complete line of Toyota vehicles with
walk-arounds, videos and more)"

* Why offer product information about 1996 Toyota products only to
people using these "operating systems"? Is Toyota trying to only
go after the "low end" market now?

As you said....

>> Let's be more aggressive about promoting Unix/Linux for business!

C'mon, joe, let's see what you can do to make your corporation
less hostile to "UNIX/Linux"!

[Ariel Mazzarelli]

"Todd Needham [Microsoft]" <to...@microsoft.com> wrote:
>> - will microsoft release Internet Explorer for Linux
>Nope.

So what will microsoft port on linux?

Ariel

[Joe Sloan]

In article <bnelsonD...@netcom.com>,
Bob Nelson <bne...@netcom.com> wrote:

>Joe...why not begin in your own backward with business promotion of
>UNIX? It seems you have the power to make it happen.

>As you said....
>
>>> Let's be more aggressive about promoting Unix/Linux for business!
>
>C'mon, joe, let's see what you can do to make your corporation
>less hostile to "UNIX/Linux"!

ROTFL - that was a good one... Bob, bob, ever the comedian...

Imagine confusing various Sloans like that -

At any rate, those who administer an internet domain are quite often
not the people who provide fodder for the web pages...

But you have pointed out a curious fact - Unix is often taken for
granted. Take my ISP for instance. Free BSD powered pentiums are the
life blood of the business, capably serving numerous virtual www
domains, providing full Unix shell accounts, and ppp connections,
with a very small number of hosts. Yet there is no mention of Unix
to be found in their advertisements - to look at their site, you'd
think Unix didn't exist - it's windows this and microsoft that...
But don't let appearances deceive you, Bob, and don't beleive the
hype - Unix is far more important than you'd ever guess just from
reading the trade press or surfing the web, and that's not going
to change anytime soon, regardless of how many people buy win95.

--
Joe Sloan - I have seen the future and it is Linux -

[Anti-Smerp]

In article <01bb4f87.1c7bdf40$1913369d@toddn40>,

Todd Needham [Microsoft] <to...@microsoft.com> wrote:

>> Why not put the source code on an ftp site? Oh.. I forgot.. you guys
>are
>> into licenses..
>
>Uhmmm... is that a point against us? Doesn't everyone (or 99.9%) who
>sells software for a living license it? Perhaps I missed something as
>regards Linux, but I thought that came very specifically with a license as
>well.

Yeah, the GNU General Public License which is pretty much the exact
opposite of 99.9% of commercial software licenses. Read
it sometime if you haven't.

Ravi
--
Ravi K. Swamy http://www4.ncsu.edu/~rkswamy/www/
rks...@eos.ncsu.edu ro...@genom.com

[John Morris]

Todd Needham [Microsoft] (to...@microsoft.com) wrote:

> > Alexandre Maret <ama...@infomaniak.ch> wrote in article
> <31AF19...@infomaniak.ch>...
> > hello
> >
> > I find it quite strange, people @microsoft.com read
> > the linux newsgroups !

> You'd be surprised...

I'm not. back when I lived over in Dallas, I knew a few people who
worked at an outpost of the 'beast' over there, and it seems many of
them had a small Linux partition so they could use LILO to manage their
various boot configurations. I have also seen more than a couple of
serious posts into the Linux newsgroups originating from microsoft.com.
Lets face it, love or hate M$, they have a lot of bright young computer
using types which are the perfect profile of the GNU Generation.

> > - will microsoft release Internet Explorer for Linux

> Nope.

It would be the ultimate admission of defeat for BG to admit there are
users who will never be fully assimilated. So long as Linux and the
other free UNIX users can be marginalized, they can be safely ignored.
On the other hand, IE for Linux might get mentioned in some of their ZD
rags, and then they might even notice Linux.

> > - is there a lot of microsoft employee running linux at home...
> > (yes... there may be some... I've seen credits...)

> I'm sure there are. We're certainly encouraged to experiment here.

> > - did microsoft already tried to hire linus...

> I'll defer to recruiting on that one!

Never happen. Linus is most certainly a Real Programmer, and while M$
used to have a few of those, the only one left is BG himself... and
I somehow doubt he even knows how to code anymore. I'd bet money that
you could not lock the dude in a room all night with an NT box, Visual
C and a case of Jolt and expect any finished code the next morning.

--
John M. This post is 100% M$ Free!
Geek code 3.0:GCS d- C+++ UL++++$ P+ L+++ W+ N++ w--- Y+>+ 5+++ R tv- b++ e*
============================================================================
The views expressed may not reflect those of DataExchange Internet Services,
so don't bother flaming them. I'm the postmaster anyway, so just flame ME!

[A Shelton]

"Todd Needham [Microsoft]" <to...@microsoft.com> writes:

>> >2. There are a greater number of development tools available for
>Win95/NT
>> >than Linux, covering a greater breadth of development needs. Certainly
>> >the size of the Windows development market and it's sheer
>competitiveness
>> >have delivered tool sets second to none.
>>
>> Admittedly they are subject to change without notice

>Yeah, it's a competitive market and tools are upgraded pretty regularly,
>but I don't think this is a weakness, but a strength.

What I meant was that the temptation of utilising programming tools
for market manipulation. Such as prematurely retiring old tools to
force upgrades, making premier tools available only on the desired
platforms (VC++ 32 is only on NT?), upgrading tools primarily to
react to marketing needs and needless incompatibility to existing
standards.

However I have had a change of heart recently and realised that
microsoft is simply doing what any company with a market monopoly
would do ( other companies might even be much worse) and pushing
progress in the market it makes it's money from.

Of course I would still love to know how America would react if
microsoft were a japanese company, perhaps then there ani-monopoly
investigations would be a bit more rigourous.

>> all rely on a proprietary single source. I noticed the windows
>programmers
>> weren't delighted with you when you dumped win32(some letter) recently.

>I've no clue _what_ you're talking about. We haven't dumped Win32. Where
>did you hear that one?!?

One of the windows.programming groups, surf for it your self. One of the
MFC programmers was trying to explain to all the people that they should
have expected it to happen. The feeling I got was that it was an attempt
to make win3.1 development difficult...but I really don't follow the
many varieties of windows library...I prefer the concept of real standards
like posix.

>> >5. Microsoft and other tools vendors offer significantly discounted
>> >educational pricing on their development tools. We also offer the same
>on
>> >our OS's. Documentation for the OS and for our development tools is
>> >available in electronic (read: full text searchable) form through the
>> >Microsoft Developer Network.
>>
>> At a mere 800 dollars for the priviledge of getting the Urdu version of
>> windows NT :)

>Hmmm. Lost me again. Academic pricing on Windows NT Workstation 3.51 is
>about $99. Half that for universities who want to use it but don't need
>the shrink wrap package.

Oh, you're not on MSDN then?

>> How are microsoft compilers on cross compiling by the way? Surely they
>> have at least as many targets as say, gcc?

>(nice lead) Our Visual C++ cross compiler for the Mac is selling quite
>briskly, thank you.

wwowowow...one platform....golly.

Amazing that it produced such a crap version of microsoft word 6.0 for
the mac...does it not work very well?

>> >6. Newsgroup support. Not only are there extensive usenet newsgroups
>on
>> >'NT and Win95 use and programming, but Microsoft recently moved it's
>> >CompuServe forums over to our own news servers -
>> >news://msnews.microsoft.com
>>
>> hehehe.....can I get it on MSN too?

>Yes. And MSN is moving completely onto the Internet. You'll be able to
>think of it as simply another resource on the 'net.

The joke is that it had to... It was meant to be the other way round
wasn't it...

>> >You and the others who responded are correct that today the lion's
>share
>> >of university research occurs on UNIX variants. However, the
>commercial
>> >world runs on Windows. It's a disconnect that I hope to see
>resolved....
>>

>> So do I, but perhaps not in the way you intend....

>Careful of that either/or trap...!

Sorry, you've lost me :)

I don't really care what system dominates the future, I just want it to
be non-proprietary so that there will be no limits on my ability to
be an independant developer. Linux, unlike microsoft, has nothing to
gain by incombatibility.

Moreover the source code availability and GPL mean that every element
of linux, from API to kernel functions, is open to me and is guaranteed
to remain so. If i don't like the way linux is going I am free to spawn
my own variant, thus I consent to be a user of linux.

Microsoft is supportive of developers only to the extent that it needs
its operating system extended. Should microsoft continue to grow the
areas in which developers can roam free without competing with a micro
-soft product can be expected to shrink.

>-Todd

Thank you for posting...

[Bob Nelson]

Joe Sloan (j...@digital.netvoyage.net) wrote:
: In article <bnelsonD...@netcom.com>,
: Bob Nelson <bne...@netcom.com> wrote:

: >Joe...why not begin in your own backward with business promotion of
: >UNIX? It seems you have the power to make it happen.
: >As you said....
: >
: >>> Let's be more aggressive about promoting Unix/Linux for business!
: >
: >C'mon, joe, let's see what you can do to make your corporation
: >less hostile to "UNIX/Linux"!

: ROTFL - that was a good one... Bob, bob, ever the comedian...

Do understand that *nothing* personal was intended. It does strike
me ironic that someone as fervent about unix as you can't use your
influence to make it better publicized at Toyota's web site. joe,
glad you did get a kick out of it, though!

Suggestion: Maybe a one-liner on the opening screen: "This site
powered by <insert unix vendor/OS>." This segues nicely into the
next point you make:

: But you have pointed out a curious fact - Unix is often taken for

: granted. Take my ISP for instance. Free BSD powered pentiums are the
: life blood of the business, capably serving numerous virtual www
: domains, providing full Unix shell accounts, and ppp connections,
: with a very small number of hosts. Yet there is no mention of Unix
: to be found in their advertisements - to look at their site, you'd
: think Unix didn't exist - it's windows this and microsoft that...

[Todd Needham [Microsoft]]

Hey, I actually haven't seen this one before. Rest assured, it's in my
HUMOR archives!

-Todd

[Todd Needham [Microsoft]]

Ahhh. Thanx for the clarification. From his original statement, I
thought he was pinging us for HAVING a license.

-Todd

[Todd Needham [Microsoft]]

> What I meant was that the temptation of utilising programming tools
> for market manipulation. Such as prematurely retiring old tools to
> force upgrades, making premier tools available only on the desired
> platforms (VC++ 32 is only on NT?), upgrading tools primarily to
> react to marketing needs and needless incompatibility to existing
> standards.

While a number of competitors have claimed MS has a monopoly on OS (and it
seems no one in this newsgroup thinks so), I've don't believe I've ever
heard anyone say we have a monopoly on development tools. Certainly
multi-million dollar companies like Borland, Symantec, Watcom, Metaware
(even IBM) would argue against this. In the absence of a monopoly, you're
stuck with competition <G>. "Market competition" is what's driving the
tools market. Not Microsoft manipulation.

As for 16-bit tools, yes, there is still a market here. And we've
upgraded our Visual Basic 4.0 to include a new 16-bit as well as 32-bit
version. As for Visual C++, there is a 32-bit and 16-bit version shipped
in every package. You are correct however in that the 16-bit version is
not being updated (other than maintenance type stuff, I believe). This is
pretty much true of the rest of the compiler market as well - developers
on our platforms have made the move to 32-bit and that's where they're
spending their development tool dollars.

Your phrase above "upgrading tools primarily to react to marketing needs"
is accurate. We're a market driven company. We deliver what the market
asks for. As for "needless incompatibility", I would have to disagree. I
think MS has bent over backwards to provide compatibility - especially in
(1) Microsoft Foundation Class (MFC) libraries, (2) the Win32 API coming
from Win16, (3) Windows Sockets 1.1 to 2.0, and (4) binary compatibility
for 16-bit applications on our 32-bit (and mostly 32-bit) operating
systems.

> However I have had a change of heart recently and realised that
> microsoft is simply doing what any company with a market monopoly
> would do ( other companies might even be much worse) and pushing
> progress in the market it makes it's money from.

Since we're talking tools here (at least I think we are), Microsoft
doesn't have a monopoly.

> Of course I would still love to know how America would react if
> microsoft were a japanese company, perhaps then there ani-monopoly
> investigations would be a bit more rigourous.

So let's see: The FTC and the Justice Department investigate Microsoft
for 5 years and don't find anything to bring charges against. How many
other companies have undergone such a scrutiny and come up clean?

> >> all rely on a proprietary single source. I noticed the windows
> >programmers
> >> weren't delighted with you when you dumped win32(some letter)
recently.
>
> >I've no clue _what_ you're talking about. We haven't dumped Win32.
Where
> >did you hear that one?!?
>
> One of the windows.programming groups, surf for it your self. One of the
> MFC programmers was trying to explain to all the people that they should
> have expected it to happen. The feeling I got was that it was an attempt
> to make win3.1 development difficult...but I really don't follow the
> many varieties of windows library...I prefer the concept of real
standards
> like posix.

Still no clue. Someone off on a tangent most likely.

> >> >5. Microsoft and other tools vendors offer significantly discounted
> >> >educational pricing on their development tools. We also offer the
same
> >on
> >> >our OS's. Documentation for the OS and for our development tools is
> >> >available in electronic (read: full text searchable) form through
the
> >> >Microsoft Developer Network.
> >>
> >> At a mere 800 dollars for the priviledge of getting the Urdu version
of
> >> windows NT :)
>
> >Hmmm. Lost me again. Academic pricing on Windows NT Workstation 3.51
is
> >about $99. Half that for universities who want to use it but don't
need
> >the shrink wrap package.
>
> Oh, you're not on MSDN then?

Assuming we've settled the question of cheap academic pricing here, I'm
happy to move on to a new subject: Microsoft Developer Network (MSDN).
In this instance (and if I understand the thrust of your one liner), we're
in agreement. Microsoft should have academic pricing (for those both
interested and unfamiliar with MSDN, check out
http://www.microsoft.com/MSDN).

> >> How are microsoft compilers on cross compiling by the way? Surely
they
> >> have at least as many targets as say, gcc?
>
> >(nice lead) Our Visual C++ cross compiler for the Mac is selling quite
> >briskly, thank you.
>
> wwowowow...one platform....golly.

wowowowow...90% of the commercial personal computer market....golly.
Guess it depends on your perspective.

> Amazing that it produced such a crap version of microsoft word 6.0 for
> the mac...does it not work very well?

I'm not a Mac person, but I do remember both performance and memory issues
with the initial release of Word 6.0 for the Mac, both long since fixed
and made available to customers at no charge.

> >Yes. And MSN is moving completely onto the Internet. You'll be able
to
> >think of it as simply another resource on the 'net.
>
> The joke is that it had to... It was meant to be the other way round
> wasn't it...

Correct. Who'dathunkit?!?

> I don't really care what system dominates the future, I just want it to
> be non-proprietary so that there will be no limits on my ability to
> be an independant developer. Linux, unlike microsoft, has nothing to
> gain by incombatibility.

Stepping outside MS for a moment...

I don't really care what system dominates in the future, I just want it to
be cheap, widely available, well supported by the industry, easily
expandable, constantly evolved and to leverage my existing programming
skills.

> Moreover the source code availability and GPL mean that every element
> of linux, from API to kernel functions, is open to me and is guaranteed
> to remain so. If i don't like the way linux is going I am free to spawn
> my own variant, thus I consent to be a user of linux.

More power to you. As a commercial software vendor this isn't a lot of
help. I remember talking with some CAD vendors who solved a problem by
simply making a couple minor modifications to the (particular) UNIX kernel
to support their app. Very powerful to have this capability. Completely
antithetical (sp?) to the commercial world where you can't get away with
this.

> Microsoft is supportive of developers only to the extent that it needs
> its operating system extended.

Capitalism at it's finest. We support developers BECAUSE THEY'RE OUR
CUSTOMERS. We support them because they buy our tools, our operating
systems, our applications. This is perfectly normal and unquestionably
correct business behavior: support your customer (if you want to
survive).

> Should microsoft continue to grow the
> areas in which developers can roam free without competing with a micro
> -soft product can be expected to shrink.

Let me rephrase to make sure I understand what you're saying: Microsoft
should grow some areas to allow independent software developers a customer
base because if Microsoft enters the same areas, the area can be expected
to shrink. If I didn't get that right, ignore the next paragraph.

Horsepucky. With few exceptions, the markets Microsoft is in have only
grown (exceptions - building compression into the operating system).
Especially the developer market. I've seen nothing to indicate the
developer tools market has shrunk over the last 10 years that Microsoft
has been in the business. In fact, it's UNPRECEDENTED growth that I see.

FYI, I'm not here to convert anyone. Microsoft and Windows NT are
religion for me. My original post was simply to correct a few
misinterpretations about Microsoft, it's products and to inform someone
about some of the things we're doing in the educational space.

(but I really do enjoy this, though...thanx!)

-Todd

[Todd Needham [Microsoft]]

> So what will microsoft port on linux?
>
> Ariel

Pretty much nothing as far as I know. Linux isn't even a blip on the
corporate radar screen today (not meant to be insulting or derogatory,
just to express the market we're currently targeting).

-Todd

[Todd Needham [Microsoft]]

> MS does have a monopoly on the OS, it's simply an established fact
> unworthy of repeating.

To each his own. I personally don't consider it a monopoly.

> Being in this situation you have a heavy advantage in any application
> development, a heavy advantage in marketing and a massive advantage
> in development tools tied to the specific environment (Visual Cobol
> perhaps?).

Okay, we're betting ourselves on a single path so that gives us an
advantage over those attempting to support more platforms. So regardless
of the reason, there's an advantage.

> You honestly believe that the manufacturers you name are playing on
> level ground? you really believe their market share is eating into
> the visual products? .....yeah right, I guess the illusion of
competition
> is important to keep the troops motivated.

No offense, but what you (or I think) is irrelevant: their profitability
speaks for itself. Do you see Borland Delphi loosing market share because
we make Visual Basic? Do you see Symantec going out of business anytime
soon (or hell even becoming unprofitable) because we're developing java
tools as well?

Sorry, but what you "believe" just isn't what's happening.

> No problem, but because windows and visual C++ are closed proprietary
> standards from a monopoly vendor there is no possibility of an
independant
> company supplying the demand for win3.1 tools no matter how much the
> market wants it. That is not competition.

Again, this is so far from reality as to hardly be worth arguing. There
are _big_ companies who made their mark on almost nothing BUT development
tools for Windows 3.X. Come on, forget your personal bias for a moment
and LOOK AT THE HEALTY, COMPETITIVE INDUSTRY HERE.

> personally the rest of the market is probably amazed that 16bit tools
> survived that long in any market save embedded controllers.

Yeah, poor Borland, Watcom, Symantec, Powersoft and IBM. What ever
happened to those 16-bit tools vendors anyways?

> The particular example I was thinking of however was the fact that
visual
> C++ (I think, I don't use these products) for 32bit only gets sold as
> an NT version.

Nope. 16- and 32-bit. See previous post this thread for details.

> You do not have a monopoly on C or C++, but you do have a monopoly
> on the OS, Visual whatever, MFC and all the other buzzwords like
> OLE.

Jeez, with all the monopolies on acronyms you claim we have, it's hard to
see how anyone can stand against us.

> >So let's see: The FTC and the Justice Department investigate Microsoft
> >for 5 years and don't find anything to bring charges against. How many
> >other companies have undergone such a scrutiny and come up clean?
>

> They did find things to bring up, and microsoft was forced to sign
> a pissweak agreement to stop some unfair practices. The fact that
> this was largely meaningless to the microsoft monopoly does not
> mean you can rewrite history.

The fact that after 5 years of investigation by two separate agencies that
Microsoft was never charged with anything is also history and you can't
rewrite that.

> The best argument I read at the time was the suggestion that the
> government could not really do anything because restricting microsoft
> would basically stuff their own information services.

Oh please.

> >More power to you. As a commercial software vendor this isn't a lot of
> >help. I remember talking with some CAD vendors who solved a problem by
> >simply making a couple minor modifications to the (particular) UNIX
kernel
> >to support their app. Very powerful to have this capability.
Completely
> >antithetical (sp?) to the commercial world where you can't get away
with
> >this.
>

> Why? I've heard this silly argument before and it doesn't make sense.
> Source availability guarantee's infinite expandability, costs nothing
> to the performance or stability of the binaries and immunises me against
> threats from the vendor......oh..I see the problem.

Come on. Talk to a systems administrator at a Fortune 500 company and
tell them they should switch to an OS with source widely available so that
they'll be able to make their own changes to the kernel or other parts of
the system whenever they need to. See how far you get.

> It is the point at which your CUSTOMERS become your COMPETITORS that
> microsoft must decide which segment it values more. Do you make more
> money out of visual C++ or microsoft office...that make it clearer?

How many industries do you know of where there isn't partner/competitor
overlap? Do you find this unusual? Shall we simply decree that companies
over X dollars big can't compete in any areas with their customers?
You're not being realistic.

Competition is what's driven the growth of this industry like it has.

> Especially since only a very small subset of your developers fall
> into this category.

I think Microsoft's board of directors would quickly be replaced if we
arbitrarily decided to, say, get out of the development tools market.
Contrary to your belief, this is a HUGE market.

> I'm hear to make sure that microsoft has its fair percentage of the
> applications market, 100% sounds about fair.

Wouldn't surprise me. We have some very aggressive people here. I'm sure
I can find people with attitudes like that at every software company.

> An old Dr Dobbs I have said something like
> First microsoft will dominate the markets that sell millions of
products,
> then they'll move into the markets that sell hundreds of thousands of
> products, then into the markets that sell tens of thousands of products
> and then they'll fire a lot of programmers..

I guess that could be a reasonable scenario if we were talking about a
static, saturated market. But I don't think that realistic.

> no problem with that....but are you CERTAIN that all of your information
> is accurate and fair?

Do I make mistakes? Certainly. Do I believe anything I've posted in this
thread is unaccurate or unfair? No.

> Try to imagine for a moment being an independant software developer
> developing a new word-processor for windows. Would you take the job?

Hell no and I'd tell them they were a complete idiot for trying it (see
Dilbert comic from a few months back on the topic). With Microsoft, Lotus
and Wordperfect in the (general-purpose) word processing market, it's
saturated. Pick another example.

-Todd

[Todd Needham [Microsoft]]

Well, I read through his post and his reasoning and workarounds look sound
to me. And they're consistent with the reasoning I've espoused in this
thread.

No arguments here I guess.

-Todd

[Michael Dillon]

In article <01bb4e4e.a4061be0$ed8d389d@toddn40>,

Todd Needham [Microsoft] <to...@microsoft.com> wrote:

>You and the others who responded are correct that today the lion's share
>of university research occurs on UNIX variants. However, the commercial
>world runs on Windows. It's a disconnect that I hope to see resolved....

Where do you get this crazy idea? The commercial world runs primarily on
MVS using IBM mainframes. Second most common is UNIX on workstations and
servers. Windows only rules the desktop, like a glorified typewriter and
calculator. But the real work of business is done on the UNIX or MVS
database systems.

The "disconnect" problem you have is that you are like the Detroit
executives in the early sevnties who drove gas guzzlers to work every day
and parked in auto company parking lots full of gas guzzlers. Meanwhile,
in California, everybody was buying Toyota, Datsun, Honda, etc...

Your disconnect problem will be solved all right, but it won't be in the
way you want it to be solved.

--
Michael Dillon ISP & Internet Consulting
Memra Software Inc. Fax: +1-604-546-3049
http://www.memra.com E-mail: mic...@memra.com

[Michael Dillon]

In article <01bb4f85.6183c190$1913369d@toddn40>,

Todd Needham [Microsoft] <to...@microsoft.com> wrote:

>> I thought Unixes had some spectacular (and spectacularly expensive)
>tools.
>> NT v. Linux starts to sound like Comm. Unix v. Linux here. How immense
>> can the Win32 tool market be compared to others, given Win32 is so
>> young?
>
>Age isn't the issue - market size is what's driving the phenomenal Windows
>tools market. Simple capitalism.

I don't think you understand. There are *MORE* installed Linux systems
than there are installed NT systems. The Linux market is actually larger.

The thing that is driving the NT tools market is hype fed directly to the
developpers by Microsoft's Developpers Network and their Developper's
conferences. BTW, the recent JAVA developpers conference in San Francisco
was larger than the MS developpers conference held at the same venue a few
weeks earlier.

The market has changed and MS is struggling to catch-up.

>> Installed commercial base and existing applications base is MS's only
>> advantage.
>
>You must believe this installed commercial base and incredible
>applications supported happened by coincidence?

Nope. We know that it happened due to three factors. One was marketing
hype causing people to buy it because PC magazine said so. One was
developpers buying NT in order to produce products for a market that they
hope materializes. One was corporate types buying one or two NT systems to
try out.

None of these things create a "base" because these sales do not imply that
the product will be in wide demand. In fact, the competition improves
their product (OS/2 WARP server and Merlin, CDE, etc) and now the market
really knows what NT can do (or not do) as opposed to reading marketing
hype.

NT is currently a niche market with about 1 million installations. Much,
much smaller than any other server OS.

[A Shelton]

"Todd Needham [Microsoft]" <to...@microsoft.com> writes:

>> What I meant was that the temptation of utilising programming tools
>> for market manipulation. Such as prematurely retiring old tools to
>> force upgrades, making premier tools available only on the desired
>> platforms (VC++ 32 is only on NT?), upgrading tools primarily to
>> react to marketing needs and needless incompatibility to existing
>> standards.

>While a number of competitors have claimed MS has a monopoly on OS (and it
>seems no one in this newsgroup thinks so), I've don't believe I've ever
>heard anyone say we have a monopoly on development tools. Certainly
>multi-million dollar companies like Borland, Symantec, Watcom, Metaware
>(even IBM) would argue against this. In the absence of a monopoly, you're
>stuck with competition <G>. "Market competition" is what's driving the
>tools market. Not Microsoft manipulation.

MS does have a monopoly on the OS, it's simply an established fact
unworthy of repeating.

Being in this situation you have a heavy advantage in any application

development, a heavy advantage in marketing and a massive advantage
in development tools tied to the specific environment (Visual Cobol
perhaps?).

You honestly believe that the manufacturers you name are playing on

level ground? you really believe their market share is eating into
the visual products? .....yeah right, I guess the illusion of competition
is important to keep the troops motivated.

>As for 16-bit tools, yes, there is still a market here. And we've

>upgraded our Visual Basic 4.0 to include a new 16-bit as well as 32-bit
>version. As for Visual C++, there is a 32-bit and 16-bit version shipped
>in every package. You are correct however in that the 16-bit version is
>not being updated (other than maintenance type stuff, I believe). This is
>pretty much true of the rest of the compiler market as well - developers
>on our platforms have made the move to 32-bit and that's where they're
>spending their development tool dollars.

No problem, but because windows and visual C++ are closed proprietary

standards from a monopoly vendor there is no possibility of an independant
company supplying the demand for win3.1 tools no matter how much the
market wants it. That is not competition.

personally the rest of the market is probably amazed that 16bit tools

survived that long in any market save embedded controllers.

The particular example I was thinking of however was the fact that visual

C++ (I think, I don't use these products) for 32bit only gets sold as
an NT version.

>Your phrase above "upgrading tools primarily to react to marketing needs"

>is accurate. We're a market driven company. We deliver what the market
>asks for. As for "needless incompatibility", I would have to disagree. I
>think MS has bent over backwards to provide compatibility - especially in
>(1) Microsoft Foundation Class (MFC) libraries, (2) the Win32 API coming
>from Win16, (3) Windows Sockets 1.1 to 2.0, and (4) binary compatibility
>for 16-bit applications on our 32-bit (and mostly 32-bit) operating
>systems.

>> However I have had a change of heart recently and realised that
>> microsoft is simply doing what any company with a market monopoly
>> would do ( other companies might even be much worse) and pushing
>> progress in the market it makes it's money from.

>Since we're talking tools here (at least I think we are), Microsoft
>doesn't have a monopoly.

You do not have a monopoly on C or C++, but you do have a monopoly

on the OS, Visual whatever, MFC and all the other buzzwords like
OLE.

>> Of course I would still love to know how America would react if

>> microsoft were a japanese company, perhaps then there ani-monopoly
>> investigations would be a bit more rigourous.

>So let's see: The FTC and the Justice Department investigate Microsoft
>for 5 years and don't find anything to bring charges against. How many
>other companies have undergone such a scrutiny and come up clean?

They did find things to bring up, and microsoft was forced to sign

a pissweak agreement to stop some unfair practices. The fact that
this was largely meaningless to the microsoft monopoly does not
mean you can rewrite history.

The best argument I read at the time was the suggestion that the

government could not really do anything because restricting microsoft
would basically stuff their own information services.

>> >> all rely on a proprietary single source. I noticed the windows

>> >programmers
>> >> weren't delighted with you when you dumped win32(some letter)
>recently.
>>
>> >I've no clue _what_ you're talking about. We haven't dumped Win32.
>Where
>> >did you hear that one?!?
>>
>> One of the windows.programming groups, surf for it your self. One of the
>> MFC programmers was trying to explain to all the people that they should
>> have expected it to happen. The feeling I got was that it was an attempt
>> to make win3.1 development difficult...but I really don't follow the
>> many varieties of windows library...I prefer the concept of real
>standards
>> like posix.

>Still no clue. Someone off on a tangent most likely.

Read the post my friend....They would get multiple, and harried, responses
from an MFC core developer for piss in the wind.

Still, I may go hunting later....

<MSDN snip>

>they
>> >> have at least as many targets as say, gcc?
>>
>> >(nice lead) Our Visual C++ cross compiler for the Mac is selling quite
>> >briskly, thank you.
>>
>> wwowowow...one platform....golly.

>wowowowow...90% of the commercial personal computer market....golly.
>Guess it depends on your perspective.

>> Amazing that it produced such a crap version of microsoft word 6.0 for
>> the mac...does it not work very well?

>I'm not a Mac person, but I do remember both performance and memory issues
>with the initial release of Word 6.0 for the Mac, both long since fixed
>and made available to customers at no charge.

Yes, It is not quite as broken. However most people accept the fact that
microsoft uses an emulation layer to avoid the requirement of actually
writing macintosh code.

(For those who don't use macs, the original word6 was slow enough that
people could out-type it).

>> I don't really care what system dominates the future, I just want it to
>> be non-proprietary so that there will be no limits on my ability to
>> be an independant developer. Linux, unlike microsoft, has nothing to
>> gain by incombatibility.

>Stepping outside MS for a moment...

>I don't really care what system dominates in the future, I just want it to
>be cheap, widely available, well supported by the industry, easily
>expandable, constantly evolved and to leverage my existing programming
>skills.

Microsoft do provide cheap, widely available and well supported products.
It must be realised that there is significant potential for this to
change as their market monopoly nears absolute (not a certainty, but a
real possibility).

No microsoft product is anywhere near the expandibility of Linux, never
will be. The source code license guarantee's this.

Microsoft product evolution is inferior to linux at this time. Admittedly
linux does not have such a broken heritage but then it has far fewer $$$
and people. Added to this is the likelihood that increasing market
dominance may lead to a reduced need to compete.

>> Moreover the source code availability and GPL mean that every element
>> of linux, from API to kernel functions, is open to me and is guaranteed
>> to remain so. If i don't like the way linux is going I am free to spawn
>> my own variant, thus I consent to be a user of linux.

>More power to you. As a commercial software vendor this isn't a lot of
>help. I remember talking with some CAD vendors who solved a problem by
>simply making a couple minor modifications to the (particular) UNIX kernel
>to support their app. Very powerful to have this capability. Completely
>antithetical (sp?) to the commercial world where you can't get away with
>this.

Why? I've heard this silly argument before and it doesn't make sense.

Source availability guarantee's infinite expandability, costs nothing
to the performance or stability of the binaries and immunises me against
threats from the vendor......oh..I see the problem.

>> Microsoft is supportive of developers only to the extent that it needs
>> its operating system extended.

>Capitalism at it's finest. We support developers BECAUSE THEY'RE OUR
>CUSTOMERS. We support them because they buy our tools, our operating
>systems, our applications. This is perfectly normal and unquestionably
>correct business behavior: support your customer (if you want to
>survive).

It is the point at which your CUSTOMERS become your COMPETITORS that

microsoft must decide which segment it values more. Do you make more
money out of visual C++ or microsoft office...that make it clearer?

Especially since only a very small subset of your developers fall
into this category.

>> Should microsoft continue to grow the

>> areas in which developers can roam free without competing with a micro
>> -soft product can be expected to shrink.

>Let me rephrase to make sure I understand what you're saying: Microsoft
>should grow some areas to allow independent software developers a customer
>base because if Microsoft enters the same areas, the area can be expected
>to shrink. If I didn't get that right, ignore the next paragraph.

good heavens no... I'm saying that if microsoft enters the same market
segment their sales will shrink. I remember a really cool quote, which
I will have to dredge from memory, attributed to your apps sales head.

(unsubstantiated quote)

I'm hear to make sure that microsoft has its fair percentage of the
applications market, 100% sounds about fair.

An old Dr Dobbs I have said something like

(same again, but I can dig up the source for this one if people want it)

First microsoft will dominate the markets that sell millions of products,
then they'll move into the markets that sell hundreds of thousands of
products, then into the markets that sell tens of thousands of products
and then they'll fire a lot of programmers..

Of course there is no reason you shouldn't...Business has given you
the right to do so.

>Horsepucky. With few exceptions, the markets Microsoft is in have only
>grown (exceptions - building compression into the operating system).
>Especially the developer market. I've seen nothing to indicate the
>developer tools market has shrunk over the last 10 years that Microsoft
>has been in the business. In fact, it's UNPRECEDENTED growth that I see.

Certainly microsoft have been at the forefront of an expanding market.
Whether they are responsible for it is another question.

>FYI, I'm not here to convert anyone. Microsoft and Windows NT are
>religion for me. My original post was simply to correct a few
>misinterpretations about Microsoft, it's products and to inform someone
>about some of the things we're doing in the educational space.

no problem with that....but are you CERTAIN that all of your information
is accurate and fair?

Try to imagine for a moment being an independant software developer

developing a new word-processor for windows. Would you take the job?

>(but I really do enjoy this, though...thanx!)

You're most welcome

[A Shelton]

"Todd Needham [Microsoft]" <to...@microsoft.com> writes:

I've been working in the news fields, all the live long day...

>> What I meant was that the temptation of utilising programming tools
>> for market manipulation. Such as prematurely retiring old tools to
>> force upgrades, making premier tools available only on the desired
>> platforms (VC++ 32 is only on NT?), upgrading tools primarily to
>> react to marketing needs and needless incompatibility to existing
>> standards.

>While a number of competitors have claimed MS has a monopoly on OS (and it
>seems no one in this newsgroup thinks so), I've don't believe I've ever
>heard anyone say we have a monopoly on development tools. Certainly
>multi-million dollar companies like Borland, Symantec, Watcom, Metaware
>(even IBM) would argue against this. In the absence of a monopoly, you're
>stuck with competition <G>. "Market competition" is what's driving the
>tools market. Not Microsoft manipulation.

Found this for you....

From comp.os.ms-windows.programmer.misc Mon Jun 3 12:38:17 1996
From: jl...@access2.digex.net (Joe Lapp)
Newsgroups: comp.os.ms-windows.programmer.misc
Subject: Help! Visual C++ 4.x Constantly Crashes!

I am peeved. I just lost a half hour's worth of resource editing.
Visual C++ 4.1 couldn't handle me clicking the radio button resource
icon for the 10th time. According to the debugger: stack overflow.
MSDEV terminated and threw away all my work.

[snip...more of the same].

That is, before MSDEV. I've always worked with MS tools on the
job and always with Borland at home. It went that way for ten
years. Now I am a consultant. I have to hone my skills for the
market. What does the market want? It wants VC++ and MFC.

For about eight months I deliberated over abandoning Borland for my
personal work. I saw enough stupid bugs in VC++ 1.5x to know that
it was a significant backward step in terms of quality. But I have
to be salable. Eventually I decided that the mature decision was
to go with Microsoft. And now Microsoft slaps me in the face.

[ This is the important bit. This guy is pro-windows, pro-borland
and actually suggests that VC++ is inferior...yet he believes that
only microsoft products are saleable...and he's probably right..

Yet you still think the playing field is level? I tempted to say
the reason borland are doing well with Delphi is because it is
not (yet) challenged directly by an MS `standard' product]

[snip...sorrow,trapped, paid good money etc..]

<End of Quoted post>

[snip of todd's post]

>> >I've no clue _what_ you're talking about. We haven't dumped Win32.
>Where
>> >did you hear that one?!?
>>
>> One of the windows.programming groups, surf for it your self. One of the
>> MFC programmers was trying to explain to all the people that they should
>> have expected it to happen. The feeling I got was that it was an attempt
>> to make win3.1 development difficult...but I really don't follow the
>> many varieties of windows library...I prefer the concept of real
>standards
>> like posix.

>Still no clue. Someone off on a tangent most likely.

yeah? sure you checked up on that one?... Unfortunately only the
remnants are left but here's some of the summary...

You will concede to me that microsoft has a vested interest in
killing/promoting the win16 market won't you?

From comp.os.ms-windows.programmer.win32 Mon Jun 3 13:04:40 1996
From: Brett Miller <bmi...@comshare.com>
Newsgroups: comp.os.ms-windows.programmer.win32
Subject: Re: Win32s not supported in next VC++ version
Date: Fri, 31 May 1996 20:52:49 -0400

Vinnie Finn wrote:
>
> Has anyone heard that Microsoft is planning not to support WIN32s any
> longer in future versions of MFC. Supposedly only a very small
> percentage of developers are using Win32s. The group I work in develops
> Win32 and Win32s applications and drivers and would dearly miss it. Is
> there anyone else out there that feels the same ?
>
> I guess I'll have to stick with old versions of VC++.
> --
> Vinnie Finn
> vf...@kodak.com

Yup, it's true. You missed the big storm a few weeks back.
MFC 4.2 will not support
Win32s. A patch release of 4.1 *might* be done to fix problems
with the current MFC.
Here's the "official word" from Steve Ross, Visual C++ Product Manager
for those that missed the mid-May posting.

Brett Miller
Comshare, Inc.

=========================================================================

~Subject:
Win32s in Future Versions of MFC
~Date:
17 May 1996 20:52:08 GMT
~From:
Steve Ross <stev...@microsoft.com>
Organization:
Unknown
~Newsgroups:
microsoft.public.vc.mfc

The rumors people have been talking about regarding Win32s are true.
Visual C++ will not be carrying forward support for Win32s as of MFC
4.2. We'd like to explain the reasoning for this, and help you
understand what your options are.

TECHNICAL REASONS

The technical reasons for not continuing Win32s support are simple:

1)

[reduced overhead...from an API library? weird.]

2) The major thrust of the new features supported by MFC 4.2 and
later are targeted exclusively at 32-bit platforms. There is no
added value continuing Win32s support because these features are
not available on the 16-bit platform. At the Microsoft Internet
Professional Developers Conference, we provided a preview of MFC
4.2, and how it supports our client-side Internet technology, or
"Sweeper," as it is code named. This technology is not available
on 16-bit clients.

WHAT WAS THE COST?

Our surveys indicated less than 2 percent of our customers were using,
or were planning to use, Win32s. Since the Microsoft systems group
plans no more work on the libraries, it didn't make sense to continue
to support them in future, non-compatible, versions of MFC.

Although at first blush this might seem like a significant blow to
those customers who _do_ use Win32s, in reality it shouldn't be. All
the Win32s functionality there ever would have been is built into
MFC 4.1. That product will continue to ship on a separate CD, along
with instructions for how to install VC++ 4.1 and future versions
on the same machine.

Please realize that continuing Win32s support in MFC 4.2 would have
provided NO ADVANTAGES over 4.1 and would have instead reduced
performance of almost everyone's Win32 executables.

WHAT ARE YOUR OPTIONS?

Developers who must still support legacy code still have two choices:
Visual C++ 1.52 or Visual C++ 4.1 with Win32s. Developers who want to
have a single code base and use the latest MFC 4.2 classes (Sweeper)
and still create Win32s executables will have to fork their code using
conditional compile pragmas. Then that code base will have to be built
on Visual C++ 4.2 for pure Win32 clients, and rebuilt using #define on
Visual C++ 4.1 for Win32s clients. Both products support the same
level of conformance with the ANSI/ISO working papers on C++
standardization. Two separate executables will result, of course, from
this process, which will have to be distributed.

SUMMARY

The long and the sort of it is that we are committed to providing you
the best tools for emerging technologies. At the same time, we also
want to provide you with the tools you need to support legacy
platforms. These tools are Visual C++ 4.2 and beyond, and Visual C++
1.52c/Visual C++ 4.1, respectively.

Steve Ross
Visual C++ Product Manager

<End of included post>

In other words, if you have to support legacy 16bit programs, like for
the massive win3.1 base, or OS/2 compatible stuff, then you are on your
own..

You even have to install the compile environment twice...and from
memory VC++ ain't small..

PLEASE don't flame me for being ignorant of what this actually means...
I am not a windows programmer and all I wanted to show is that some
percentage of developers wanted a variant from the official MS product.

It may well be that the technical reasons given are strong enough to
justify the change (although a posting later in the thread spoke of
how `interesting' the conversion process was) but it must be said
that microsoft gains from this move, and that there is no danger of
a competitor keeping the platform viable.

[Craig Maloney]

On Fri, 31 May 1996 23:31:53 -0700 in comp.os.linux.advocacy Todd Needham [Microsoft] (to...@microsoft.com) wrote:

: > Installed commercial base and existing applications base is MS's only
: > advantage.

: You must believe this installed commercial base and incredible
: applications supported happened by coincidence?

No... Microsoft was preaching interoperability at a time when people were sick
of light weight "*works" packages and non-interoperable heavyweights (like
Word Perfect). MS put together Office applications that work together. People
like the idea and buy into it. MS doesn't port to other platforms. People
buy Windows for neat interoperable package. Simple supply and demand. If
MS ported Office to Unix, you'd see less people running Windows.

: > I used to think MS had some other advantages. But real free
: > Unix changes the rules.

: Well, I agree that free UNIX is a good thing for end users anyways....

: > Most of the extremely plural and competitive commercial software

: industry
: > is dependent (read controlled) by one firm. It's a disconnect I (and
: > thousands of others) hope to see resolved.

: hmmm...highly competitive commercial software industry....controlled by on
: firm. You're right. We definitely have a disconnect....

Three letters, Todd: IBM.
: -Todd

--
||| Craig Maloney | Phone: [313] 390-8096 | Automotive Safety Center |||
||| Opinions expressed are my own. | Ford Motor Company |||
||| "This message rated [R]. May contain violent punctuation, explicit |||
||| grammatical errors, misspelled language and a shocking ending with |||
||| a preposition." -- Michael Patterson | Censor Congress. |||

[Craig Maloney]

On Sat, 1 Jun 1996 00:04:15 -0700 in comp.os.linux.advocacy Todd Needham [Microsoft] (to...@microsoft.com) wrote:

: > Alexandre Maret <ama...@infomaniak.ch> wrote in article
: <31AF19...@infomaniak.ch>...

: > - will microsoft release Internet Explorer for Linux

: Nope.

That's a pity... Your closest competitor in the web browser arena (Netscape)
does. And it's free. (For Linux users, anyway)

[Darin Johnson]

> I don't think you understand. There are *MORE* installed Linux systems
> than there are installed NT systems. The Linux market is actually larger.

This is mostly because of the way NT is marketted. Microsoft seems to
want this to be a purely server and high end system. Not for use on
the desktop, except by developers, system admins, or other people
unrelated biologically to "real users". Win95 is supposed to be that
machine, never mind that it's only real improvements of 3.1 are a
nicer GUI and having win32s builtin. The users are supposed to use
Win95, and WinNT is supposed to be hidden in the basement.

Linux on the other hand, is a desktop system. You don't have to be in
the elite to use it, you just need a PC. You can run it on a much
smaller machine than is needed for NT (I use an 8meg machine at work
for it).

So of course, it's only natural that the desktop system turns out to
be more popular...
--
Darin Johnson
djoh...@ucsd.edu O-
"Particle Man, Particle Man, doing the things a particle can"

[Todd Needham [Microsoft]]

> >Age isn't the issue - market size is what's driving the phenomenal
Windows
> >tools market. Simple capitalism.
>

> I don't think you understand. There are *MORE* installed Linux systems
> than there are installed NT systems. The Linux market is actually
larger.

I don't think you understand: we're not talking about 'NT here, we're
talking about 32-bit Windows tools market and that's Windows NT, Windows
95 and Windows 3.1 (with Win32s). Order of magnitude difference here.

> The thing that is driving the NT tools market is hype fed directly to
the
> developpers by Microsoft's Developpers Network and their Developper's
> conferences.

The thing that is driving the Win32 tools market is CUSTOMER DEMAND. Some
things that are facilitating that market are the Professional Developer's
Conferences, and the Microsoft Developer's Network (MSDN).

> BTW, the recent JAVA developpers conference in San Francisco
> was larger than the MS developpers conference held at the same venue a
few
> weeks earlier.

Not that "Mine is bigger than yours" is a great argument, but the last I
saw, both conferences maxed out the Moscone center at roughly 5K (MS also
had an additional 60+ theaters downlinked nationwide which REALLY pumps up
the numbers).

> The market has changed and MS is struggling to catch-up.

Quite true. And I applaud the incredible progress the Internet teams here
at Microsoft have made in the last 9 months.

> Nope. We know that it happened due to three factors. One was marketing
> hype causing people to buy it because PC magazine said so. One was
> developpers buying NT in order to produce products for a market that
they
> hope materializes. One was corporate types buying one or two NT systems
to
> try out.
>
> None of these things create a "base" because these sales do not imply
that
> the product will be in wide demand. In fact, the competition improves
> their product (OS/2 WARP server and Merlin, CDE, etc) and now the market
> really knows what NT can do (or not do) as opposed to reading marketing
> hype.

You seem to be focusing on only one Win32 platform and this is skewing
your numbers.

> NT is currently a niche market with about 1 million installations. Much,
> much smaller than any other server OS.

Rest assured, the installed base of 'NT is _well_ above this.

-Todd

[Todd Needham [Microsoft]]

> No... Microsoft was preaching interoperability at a time when people
were sick
> of light weight "*works" packages and non-interoperable heavyweights
(like
> Word Perfect). MS put together Office applications that work together.
People
> like the idea and buy into it. MS doesn't port to other platforms.
People
> buy Windows for neat interoperable package. Simple supply and demand. If
> MS ported Office to Unix, you'd see less people running Windows.

This doesn't really hold up in light of Office for the Mac. On the Mac,
we actually have a larger percentage of the suite market than we do on the
PC.

> : hmmm...highly competitive commercial software industry....controlled
by on
> : firm. You're right. We definitely have a disconnect....
>
> Three letters, Todd: IBM.

Excellent example. I like to think Microsoft learned a lot from their
failures in the PC space.

-Todd

[Michael Dillon]

In article <bnelsonD...@netcom.com>,
Bob Nelson <bne...@netcom.com> wrote:

>Impartial statistics show LINUX is a considerable force that marketers
>simply can't ignore. Over 30,000 LINUX CD-ROMS are shipped monthly with

Sorry you got that wrong. A year ago, someone from one of the Linux CD-ROM
distributors revealed that they were shipping an average of 40,000 Linux
CD's every month. That's just one company! And it's also a year out of
date.

[Amit Chatterjee]

"Todd Needham [Microsoft]" <to...@microsoft.com> wrote:

>>
>> At a mere 800 dollars for the priviledge of getting the Urdu version of
>> windows NT :)
>
>Hmmm. Lost me again. Academic pricing on Windows NT Workstation 3.51 is
>about $99. Half that for universities who want to use it but don't need
>the shrink wrap package.
>

(1) Does the $99 Windows NT Workstation 3.51 include software development
tools, networking tools, games, internet browsing tools, etc. etc. etc.... and
gives you the freedom to copy the software to as many systems you want to ?

(2) In order to run all the above software packages simultaneously, how much
system memory is required ?

Todd, if your answer to question (1) is "NO" and answer to question (2) is
"around 32MB", then NT is not worth it even with a $99 price tag. I suggest
that you buy a copy of Redhat 3.0.3 Linux (sells around $40 with a reasonably
well-written manual, $20 without the manual and the installation is simple
enough even for microsoft users) and install it at home and check it out. You
will be surprized at what it can do with so little resource.

--
/***************************************************************************
Amit Chatterjee
E-mail me at as...@ix.netcom.com

All opinions are mine, not BNR's.
****************************************************************************/

[Amit Chatterjee]

"Todd Needham [Microsoft]" <to...@microsoft.com> wrote:
>
>>

>> - is there employees at microsoft whose job is "newsreader",
>> or were you browsing comp.os.linux.advocacy for your
>> pleasure...
>
>No, there's no one with that title. I follow this group because I run
>into Linux occasionally in the universities I work with.
>

Have you installed linux on your home computer and tried it out ? OR are you
trying to compare Linux with NT on the basis of what you hear from the people
whom you run into occasionally ?

[Todd Needham [Microsoft]]

Okay, so right after I post my last message saying no, I find this post
come across my email system:

"Microsoft Develops Internet Explorer for Unix (PC Week)
From PC Week Online for May 31, 1996 by Maria Seminerio

Microsoft Corp., taking aim at Netscape Communications Corp.'s
cross-platform browser market, is developing a Unix version of its
Internet Explorer browser.

Customer demand led the company to consider the Unix browser idea, in a
bid to capture some of Netscape's overwhelming lead in the browser
market, said Dawn Leonetti, a spokesperson for Microsoft.

The company is investigating several Unix vendors but has not yet
signed any deals, Leonetti said.

Creating the Unix browser would give Internet Explorer a place on the
only major desktop it lacks. The browser is currently available for
Windows and the Macintosh.

No time frame has been set for the release of the product."

In all honesty, I haven't seen anything come around officially on this so
it's "grain of salt" until it does...

-Todd

[Todd Needham [Microsoft]]

> Where do you get this crazy idea? The commercial world runs primarily on
> MVS using IBM mainframes. Second most common is UNIX on workstations and
> servers. Windows only rules the desktop, like a glorified typewriter and
> calculator. But the real work of business is done on the UNIX or MVS
> database systems.

Should have been more clear, sorry. This thread was following development
tools. "Windows only rules the desktop" is several orders of magnitude
larger installed base than MVS mainframes (and at least one or two greater
than UNIX on the desktop). The disconnect is the demand for Windows
programmers vs what schools are using to teach.

> The "disconnect" problem you have is that you are like the Detroit
> executives in the early sevnties who drove gas guzzlers to work every
day
> and parked in auto company parking lots full of gas guzzlers. Meanwhile,
> in California, everybody was buying Toyota, Datsun, Honda, etc...

Really bad analogy. "PC's are to mainframes what gas guzzlers are to
Hondas/Toyotas/etc.".

> Your disconnect problem will be solved all right, but it won't be in the
> way you want it to be solved.

Everyone's entitled to their opinion. I'm betting it will be.

-Todd

[Todd Needham [Microsoft]]

> (1) Does the $99 Windows NT Workstation 3.51 include software
development
> tools, networking tools, games, internet browsing tools, etc. etc. etc

The answers are: no, yes, yes, yes and probably yes,yes,yes again
depending on what your "etc.'s" end up being. If you're looking for the
GNU toolset, that's available too.

> gives you the freedom to copy the software to as many systems you want
to ?

No. If you read my post again, you'll find that copying to additional
machines is about $50 each.

> (2) In order to run all the above software packages simultaneously, how
much
> system memory is required ?

Depends on (a) what development tools you're running, (b) what networking
tools you're running, (c) what games you're running....

If you're a developer, you're typically looking at a 24 meg or better
machine for acceptable performance.

-Todd

[Todd Needham [Microsoft]]

> Have you installed linux on your home computer and tried it out ? OR are
you
> trying to compare Linux with NT on the basis of what you hear from the
people
> whom you run into occasionally ?

Actually, if you backtrack this thread, I was merely pointing out
incomplete or mis-statements about 'NT/Win32. Have I mis-stated anything
here about Linux?

-Todd

[Todd Needham [Microsoft]]

> This is mostly because of the way NT is marketted. Microsoft seems to
> want this to be a purely server and high end system. Not for use on
> the desktop, except by developers, system admins, or other people
> unrelated biologically to "real users". Win95 is supposed to be that
> machine, never mind that it's only real improvements of 3.1 are a
> nicer GUI and having win32s builtin. The users are supposed to use
> Win95, and WinNT is supposed to be hidden in the basement.

If you're interested in the real story, check out
http://internet/NTWorkstation/NTWN9500.htm

-Todd

[William Edward Webber]

Todd Needham [Microsoft] (to...@microsoft.com) opined:
: Okay, so right after I post my last message saying no, I find this post

: come across my email system:
:
: "Microsoft Develops Internet Explorer for Unix (PC Week)
: From PC Week Online for May 31, 1996 by Maria Seminerio
:
: Microsoft Corp., taking aim at Netscape Communications Corp.'s
: cross-platform browser market, is developing a Unix version of its
: Internet Explorer browser.

:
The man gets results! Keep the good work up, Todd, I always thought this
newsgroup needed a resident insider at Microsoft to get that company
rolling the right direction :-).

William Webber.
---
William Webber Postgrad. Dip. in CS, RMIT, Australia
w...@yallara.cs.rmit.edu.au http://www.ozemail.com.au/~wew/
Interests: linux, chess, hermeneutics
"Well, I'll say this for you, the quality of your stupidity is rising"
- Lucy, from Peanuts.

[Tim Smith]

A Shelton <ashe...@yallara.cs.rmit.EDU.AU> wrote:
>for market manipulation. Such as prematurely retiring old tools to
>force upgrades, making premier tools available only on the desired
>platforms (VC++ 32 is only on NT?), upgrading tools primarily to
>react to marketing needs and needless incompatibility to existing
>standards.

VC++ 32-bit compilers run fine on Windows 95. However, why assume that
people doing Windows development are using Microsoft compilers? Borland,
Symantec, and Watcom all license from Microsoft the most important parts
of VC++ (MFC and the online documentation).

>>(nice lead) Our Visual C++ cross compiler for the Mac is selling quite
>>briskly, thank you.
>
>wwowowow...one platform....golly.

One platform with more users than all versions of Unix put together and
integrated over time from the big bang to the present. Gee, I wonder why
they picked that one, rather than one of the numerous insignificant platforms
that, say, gcc targets? (BTW, gcc does not target the Mac).

--Tim Smith

[Bruce Ide]

In article <01bb4e4e.a4061be0$ed8d389d@toddn40>,

"Todd Needham [Microsoft]" <to...@microsoft.com> writes:

>Julian,
>
>1. Source code for Windows NT is also available to universities at no
>charge (currently 29 universities world wide are source code licensees and
>we're rapidly expanding that).

Great. Where can I FTP a copy?

>
>2. There are a greater number of development tools available for Win95/NT
>than Linux, covering a greater breadth of development needs. Certainly
>the size of the Windows development market and it's sheer competitiveness
>have delivered tool sets second to none.
>

This sounds like marketroid hype. Have you counted on either system?

>3. Out of the box, NT (like Linux) contains a full networking environment
>(including TCP/IP with sockets and Novell compatible IPX/SPX). Nothing
>you need to add for networking unless you need NFS (available from a
>number of vendors).

X/Windows perchance? I do like to NFS mount wuarchive over PPP and I gotta
have that remote windows support?

>
>4 Add the compiler or development tool set of your choice (one of ours,
>Borland's, Symantec's, the GNU ports, etc.) and you're up and running.
>

IMHO the Windows and OS/2 message passing archetecture is a lot less
elegant and far less intuitive than the X Paradigm of callbacks. Nor
do I particularly like the API of the week club you tend to see
Which WIN* API do you want to develop for today?

>5. Microsoft and other tools vendors offer significantly discounted
>educational pricing on their development tools. We also offer the same on
>our OS's. Documentation for the OS and for our development tools is
>available in electronic (read: full text searchable) form through the
>Microsoft Developer Network.
>

Todd. Babe. Kinda hard to beat FREE. And you wanna talk educational,
try setting up a complete UNIX system or better yet a complete UNIX
TCP/IP network from the ground up, configuring Web servers, mail,
news, X/Windows and DNS. Keeping in mind that UNIX is TCP/IP's native
home. Configure TCP/IP on UNIX and you can do it anywhere. Be it
FreeBSD, Solaris, AIX, Linux or whatever, nearly everything remains
the same.

>6. Newsgroup support. Not only are there extensive usenet newsgroups on
>'NT and Win95 use and programming, but Microsoft recently moved it's
>CompuServe forums over to our own news servers -
>news://msnews.microsoft.com

<sarcasm>
Oooh Ahhh. Newsgroup support. What a novel concept. MS must have invented
it.
</sarcasm>

>You and the others who responded are correct that today the lion's share
>of university research occurs on UNIX variants. However, the commercial
>world runs on Windows. It's a disconnect that I hope to see resolved....
>

It's a what, Todd? I'd like to see it resolved, too, whatever it is.
To wit, everyone's learned UNIX in college and UNIX gets you the highest
paying jobs out of college ($70K for entry level X programmers? Yow!)
So why not run UNIX at home? At work? Anywhere Windows does it, UNIX
can do it better, better and in less memory.

--
-----------------------------------
Bruce Ide gre...@gate.net, gre...@vnet.net, or ro...@greyfox.org
http://www.greyfox.org

[Anthony D. Tribelli]

A Shelton (ashe...@yallara.cs.rmit.EDU.AU) wrote:
: How are microsoft compilers on cross compiling by the way? Surely they

: have at least as many targets as say, gcc?

Currently only x86, 68K, and PPC. I don't know about 68K or PPC, but
Microsoft has produced much better x86 code (in terms of execution speed)
than gcc on the code I've run through both.

Tony
--
------------------
Tony Tribelli
adtri...@acm.org

[Anthony D. Tribelli]

Todd Needham [Microsoft] (to...@microsoft.com) wrote:

: > weren't delighted with you when you dumped win32(some letter) recently.
:
: I've no clue _what_ you're talking about. We haven't dumped Win32. Where

: did you hear that one?!?

PCWeek recently reported that Visual C++ 4.1 is the last release to offer
Win32s support, 4.2 and later will only target Win95 and WinNT. The
article claimed Microsoft viewed Win32s as an interim solution for
developer's until Win95 was released. However, as a developer, I have a
different perspective. Win32s is viewed by developers as an interim
solution until users have switched from Win3.1 to Win95. Microsoft
releasing Win95 is not sufficient.

: (nice lead) Our Visual C++ cross compiler for the Mac is selling quite
: briskly, thank you.

Briskly for a $2000 compiler. :-)

If you want something to brag about (beatup gcc users with) I would
suggest the Developer Network CD subscription.

[Anthony D. Tribelli]

A Shelton (ashe...@yallara.cs.rmit.EDU.AU) wrote:
: The particular example I was thinking of however was the fact that visual

: C++ (I think, I don't use these products) for 32bit only gets sold as
: an NT version.

Visual C++ has supported Win95 for quite a while, older versions were
WinNT specific.

Just out of curiousity, why is it that you keep commenting on topics (DOS
extenders, game development, Visual C++, ...) you have no knowledge of or
no experience with? I think this is the third time in recent history I've
seen you use qualifications like: I don't have, I don't use, leads me to
believe, ... You could do a little research before commenting, or at least
let us know when your guessing in the original posts. :-)

: Yes, It is not quite as broken. However most people accept the fact that

: microsoft uses an emulation layer to avoid the requirement of actually
: writing macintosh code.

Actually, it's not too terrible an approach. A big chunk of software is
not system dependent (the core application code) or not performance
critical (dialog box waiting for a user to type or click). Where
Microsoft may have screwed up with this approach is that they may have
skipped the performance tuning step where you identify the performance
critical areas and add conditional compilation. For fun I recompiled some
Windows based molecular visualization code for the Macintosh and after I
replaced Win GDI polygon and circle calls with Apple QuickDraw calls the
program ran twice as fast as the untuned original recompile, and as far
as my stopwatch relexes permit I believe the tuned code runs about as
fast as our native Mac version compiled with MetroWerks.

I know some long time Mac developers who are using this approach while
porting some games from PC to Mac. Of course only things like a game's
configuration and map editor is a Windows-to-Mac recompile. The game's
themselves are done with native Mac tools and have lots of rewritten and
special purpose code.

[Anthony D. Tribelli]

Michael Dillon (mic...@memra.com) wrote:
: I don't think you understand. There are *MORE* installed Linux systems

: than there are installed NT systems. The Linux market is actually larger.

: The thing that is driving the NT tools market is hype ...

Incredibly naive and very wrong with respect to commercial software.
There are more linux user's, but they don't constitute a "market" since
most will not spend any money. Most linux users will get along quite well
with the freely available software. In this context, WinNT does have a
larger "market" in spite of fewer users.

[Anthony D. Tribelli]

Michael Dillon (mic...@memra.com) wrote:
: In article <bnelsonD...@netcom.com>,
: Bob Nelson <bne...@netcom.com> wrote:

: >Impartial statistics show LINUX is a considerable force that marketers
: >simply can't ignore. Over 30,000 LINUX CD-ROMS are shipped monthly with

: Sorry you got that wrong. A year ago, someone from one of the Linux CD-ROM
: distributors revealed that they were shipping an average of 40,000 Linux
: CD's every month. That's just one company! And it's also a year out of
: date.

And how many of those 40,000 users also purchased a WordPerfect,
Mathematica, or some other piece of commercial software? There is a big
"market" for linux operating system CDs, but there is a small "market"
for commercial application software. Too many users get along very well
with the free software.

[Todd Needham [Microsoft]]

Can you (or anyone else reading this message) tell me roughly how many
Linux users there are out there?

-Todd

[Todd Needham [Microsoft]]

> PCWeek recently reported that Visual C++ 4.1 is the last release to
offer
> Win32s support, 4.2 and later will only target Win95 and WinNT. The
> article claimed Microsoft viewed Win32s as an interim solution for
> developer's until Win95 was released. However, as a developer, I have a
> different perspective. Win32s is viewed by developers as an interim
> solution until users have switched from Win3.1 to Win95. Microsoft
> releasing Win95 is not sufficient.

You read correctly. See another branch in this thread for the details.

> : (nice lead) Our Visual C++ cross compiler for the Mac is selling quite
> : briskly, thank you.
>
> Briskly for a $2000 compiler. :-)

All things in perspective. :)

> If you want something to brag about (beatup gcc users with) I would
> suggest the Developer Network CD subscription.

I would agree. But to be honest, I'm not perusing this newsgroup to beat
anyone up. Working in the university research area, I just like to keep
up with the competition and correct misunderstandings about our offerings
along the way.

Success beyond my wildest dreams wouldn't convert a fraction of the folks
who browse this newsgroup (or justify the effort required).

-Todd

[Todd Needham [Microsoft]]

Depending on how you actually define cross compiler, we have MIPS and
ALPHA support as well.

> a...@netcom.com (Anthony D. Tribelli) wrote in article
<adtDsG...@netcom.com>...

[Joe Sloan]

In article <adtDsG...@netcom.com>,

Anthony D. Tribelli <a...@netcom.com> wrote:

>Incredibly naive and very wrong with respect to commercial software.
>There are more linux user's, but they don't constitute a "market" since
>most will not spend any money. Most linux users will get along quite well
>with the freely available software. In this context, WinNT does have a
>larger "market" in spite of fewer users.

Tell me, Tony, how did you arrive at this conclusion? I mean where you
proclaim that "most linux users" will not spend any money.

Could you share your research with me? I'd really like to examine the
statistics you collected.

Meanwhile, Caldera and Red Hat are prospering, somehow managing to
sell an awful lot of value-added CDROMs in a market where Linux is
freely available from ftp sites and BBSes, even free CD giveaways...

I spent $500 on a Linux network spreadsheet a year or so ago, and
I will probably spring for either the Caldera network desktop or
a Red Hat system with the Applix suite, once the 2.0 kernel is
released.

Don't kid yourself Tony - Linux users will pay good money for software
if they think the quality is up to snuff. Personally speaking, I don't
use Linux just because it's less expensive than windows "nt" or 95 -
I could have either of those on my desktop if I wanted them -

No, Tony, I use Linux because I prefer it.

--
Joe Sloan - I have seen the future and it is UNIX -
j...@netvoyage.net <a href="http://www.netvoyage.net/~jjs">

[Todd Needham [Microsoft]]

> >1. Source code for Windows NT is also available to universities at no
> >charge (currently 29 universities world wide are source code licensees
and
> >we're rapidly expanding that).
>
> Great. Where can I FTP a copy?

Happy to send the license to your university CS Chair or Dean for
signature. Please email me the shipping details.

> X/Windows perchance? I do like to NFS mount wuarchive over PPP and I
gotta
> have that remote windows support?

See my previous post.

> >5. Microsoft and other tools vendors offer significantly discounted
> >educational pricing on their development tools. We also offer the same
on
> >our OS's. Documentation for the OS and for our development tools is
> >available in electronic (read: full text searchable) form through the
> >Microsoft Developer Network.
>
> Todd. Babe. Kinda hard to beat FREE.

Quite true. Guess it kind of depends on whether you want to prepare for
the ever burgeoning commercial software market (and we all know what the
majority there is). Individual choice to make the investment.

> And you wanna talk educational,
> try setting up a complete UNIX system or better yet a complete UNIX
> TCP/IP network from the ground up, configuring Web servers, mail,
> news, X/Windows and DNS. Keeping in mind that UNIX is TCP/IP's native
> home. Configure TCP/IP on UNIX and you can do it anywhere. Be it
> FreeBSD, Solaris, AIX, Linux or whatever, nearly everything remains
> the same.

We can argue back on forth on this all week and never convince the other.
Setting up a Windows NT machine on a TCP/IP net is a breeze. And it's
just as easy to add Netware support. Or Vines support. Or NFS support.
Certainly the native DHCP support we've had in 'NT since 3.1 is a huge
plus in ease of configuration.

Web servers? Hey, check out the recent reviews on, say, Microsoft IIS on
'NT vs. some of the web servers available for UNIX. While folks will
argue performance vs. features endlessly, we win the ease of installation
category every time.

DNS? Okay, same level of difficulty both sides for the current release of
'NT. 'NT 4.0 (now in beta 2) includes a much praised graphical
administration interface for DNS/WINS integration.

Cross platform. Hmmm. More difficult argument and kind of depends on
how you define cross platform. If you define it as running on different
computer architectures, 'NT wins (set up is identical across Intel, DEC
Alpha AXP, MIPS and PowerPC processors running 'NT). If you define it as
different operating systems, you win as long as it's a UNIX variant.

> >6. Newsgroup support. Not only are there extensive usenet newsgroups
on
> >'NT and Win95 use and programming, but Microsoft recently moved it's
> >CompuServe forums over to our own news servers -
> >news://msnews.microsoft.com
>
> <sarcasm>
> Oooh Ahhh. Newsgroup support. What a novel concept. MS must have
invented
> it.
> </sarcasm>

Oooh. Ahhh. Newsgroup AND commercial services(MSN, AOL) AND telephone
support AND 24 service options AND support from major systems integrators
like DEC, HP, COMPAQ, etc......

Guess I wasn't really certain what the sarcasm was for (and yes, I
recognize this is a pretty lame "come back", but sarcasm was never my area
of expertise :).

> >You and the others who responded are correct that today the lion's
share
> >of university research occurs on UNIX variants. However, the
commercial
> >world runs on Windows. It's a disconnect that I hope to see
resolved....
>
> It's a what, Todd? I'd like to see it resolved, too, whatever it is.
> To wit, everyone's learned UNIX in college and UNIX gets you the highest
> paying jobs out of college ($70K for entry level X programmers? Yow!)
> So why not run UNIX at home? At work? Anywhere Windows does it, UNIX
> can do it better, better and in less memory.

So tell me: why hasn't UNIX taken over the world? Why do so many people
run Windows at home? At work? And anywhere UNIX is running? Why is the
PC market with Windows racing so far ahead of the UNIX market in
practically every measure? I've already got an annual run rate on 'NT
that's larger than the entire installed base of Solaris. Hell, we've even
got Sun releasing software development kits for Windows!
(http://internet.microsoft.com/tools/tools.htm)

Yes, there are good paying programming jobs available in UNIX. But what
programming market is growing more quickly?

[Darin Johnson]

> That's a pity... Your closest competitor in the web browser arena (Netscape)
> does. And it's free. (For Linux users, anyway)

Is it free? I don't think so, except for educational users, or if
you're using it in a non-profit institution. But home use does not
apply here, and use at most companies doesn't apply either.

What's the scoop here? I've been thinking of getting the windows
version and not using it just to pay for the linux version.

--
Darin Johnson
djoh...@ucsd.edu O-

The full name of the compiler is "Compiler Language With No Pronounceable
Acronym", which is, for obvious reasons, abbreviated "INTERCAL".

[Hoi Yan Lydia Lam]

On Thu, 30 May 1996, Todd Needham [Microsoft] wrote:
>You and the others who responded are correct that today the lion's share
>of university research occurs on UNIX variants. However, the commercial
>world runs on Windows. It's a disconnect that I hope to see resolved....

It might, if the commercial world realises its huge mistake. (At least a
few companies have intelligent IS departments, relying on UNIX of one type
or another. And in the U.S., state, and local government levels, UNIX is
used for mission-critical work, because they cannot rely on NT.)

[Hoi Yan Lydia Lam]

On 1 Jun 1996, Ariel Mazzarelli wrote:
>So what will microsoft port on linux?

Nothing. M$ cannot make any money selling their technologically-inferior
hash to us. They do not want to play a game they will definitely lose.

[Hoi Yan Lydia Lam]

On Mon, 3 Jun 1996, Todd Needham [Microsoft] wrote:
>
>> Your disconnect problem will be solved all right, but it won't be in the
>> way you want it to be solved.
>
>Everyone's entitled to their opinion. I'm betting it will be.

How noble of you to go down with the ship, Captain!

[Anti-Smerp]

In article <01bb5237.0530e530$ed8d389d@toddn40>,

Todd Needham [Microsoft] <to...@microsoft.com> wrote:
>Depending on how you actually define cross compiler, we have MIPS and
>ALPHA support as well.

Can you be using an x86 box and produce an Alpha executable? Vice
versa, all the other combinations, etc.

>> a...@netcom.com (Anthony D. Tribelli) wrote in article
><adtDsG...@netcom.com>...
>> A Shelton (ashe...@yallara.cs.rmit.EDU.AU) wrote:
>> : How are microsoft compilers on cross compiling by the way? Surely they
>> : have at least as many targets as say, gcc?
>>
>> Currently only x86, 68K, and PPC. I don't know about 68K or PPC, but
>> Microsoft has produced much better x86 code (in terms of execution
>speed)
>> than gcc on the code I've run through both.

Ravi
--
Ravi K. Swamy http://www4.ncsu.edu/~rkswamy/www/
rks...@eos.ncsu.edu ro...@genom.com

[Maynard Handley]

In article <01bb523a.3fd9a6b0$ed8d389d@toddn40>, "Todd Needham
[Microsoft]" <to...@microsoft.com> wrote:

> > >1. Source code for Windows NT is also available to universities at no
> > >charge (currently 29 universities world wide are source code licensees
> and
> > >we're rapidly expanding that).
> >
> > Great. Where can I FTP a copy?
>
> Happy to send the license to your university CS Chair or Dean for
> signature. Please email me the shipping details.
>

I suspect the reason many people are leary of NT in spite of offers like
this is because it is owned by MS and MS has a proven track record of
being willing to cripple the OS to achieve some (revenue) goal. Obvious
examples are the use of VB all over Office---with the corollary that
Windows has no scripting language like AppleScript/Apple Events, or the
use of OLE with no concession to technically superior alternatives.

The fact is that right NOW (IMHO) NT is a better designed OS than the
random flavors of UNIX. But how long will it stay that way? Meanwhile
(again IMHO), once MkLinux is up and running it'll provide a decent
technical base for expansion directions beyond the suffocating constraints
of UNIX' simple-minded paradigms like fork() or process-based scheduling?
The fights with DEC over who gets to add 64bit, or the refusal to go
big-endian certainly don't make one confident that MS' only interest is in
the technical quality of the OS.

Various forms of spin and hype about the wonderful things MS is doing/will
do with NT may convince the suits, but they're not going to convince many
people in this newsgroup. I suspect the only thing that would really
convince people MS is willing to let NT live its life as a high quality OS
defined by the best technology available, not the technology MS wants it
to use, is to spin NT into a separate company whose only like to MS is
something like royalty payments for each copy shipped. Fat chance that'll
happen.

Maynard Handley

--
My opinion only

[Todd Needham [Microsoft]]

> It might, if the commercial world realises its huge mistake. (At least
a
> few companies have intelligent IS departments, relying on UNIX of one
type
> or another. And in the U.S., state, and local government levels, UNIX
is
> used for mission-critical work, because they cannot rely on NT.)

The commercial world hasn't made a mistake. UNIX still has a large share
of the mission-critical server market. Windows NT hasn't been on the
market THAT long :). And yes, the US state and local governments are some
of these users. Just like they're also some of the Windows NT users (and
yes, they can rely on Windows NT and yes, the US Gov't does buy Windows
NT).

-Todd

[Todd Needham [Microsoft]]

> >Everyone's entitled to their opinion. I'm betting it will be.
>
> How noble of you to go down with the ship, Captain!

Somehow with stock at an all time high, I don't feel like my ship is
sinking, but thanx for the sentiment!

-Todd

[Todd Needham [Microsoft]]

> >So what will microsoft port on linux?
>
> Nothing. M$ cannot make any money selling their
technologically-inferior
> hash to us. They do not want to play a game they will definitely lose.

Interesting. Why do you believe it's a game we would lose?

-Todd

[Todd Needham [Microsoft]]

> I suspect the reason many people are leary of NT in spite of offers like
> this is because it is owned by MS and MS has a proven track record of
> being willing to cripple the OS to achieve some (revenue) goal. Obvious
> examples are the use of VB all over Office---with the corollary that
> Windows has no scripting language like AppleScript/Apple Events, or the
> use of OLE with no concession to technically superior alternatives.

Just so I understand this clearly:

1. We crippled Windows by not including Visual Basic. Never mind that
customers have their choice of scripting language from a number of
vendors. Never mind that a third party industry grew up around scripting
languages. Never mind that someone would have claimed we were trying to
force VB down their throat by including it in the OS.

2. Where do you get "no concession to technically superior alternatives"?
What are you talking about?

> The fact is that right NOW (IMHO) NT is a better designed OS than the
> random flavors of UNIX. But how long will it stay that way? Meanwhile
> (again IMHO), once MkLinux is up and running it'll provide a decent
> technical base for expansion directions beyond the suffocating
constraints
> of UNIX' simple-minded paradigms like fork() or process-based
scheduling?
> The fights with DEC over who gets to add 64bit, or the refusal to go
> big-endian certainly don't make one confident that MS' only interest is
in
> the technical quality of the OS.

What? 'NT is technically inferior because it's little-endian? Where do
you get that one?!

> Various forms of spin and hype about the wonderful things MS is
doing/will
> do with NT

Uhmmmm. "Spin". "Hype". With 'NT. Such as...?

> may convince the suits, but they're not going to convince many

> people in this newsgroup. I suspect the only thing that would really...

Never expected it too. Not my purpose here.

> convince people MS is willing to let NT live its life as a high quality
OS

> defined by the best technology available...

We're doing that (with obvious limitations like what the market will
handle).

> , not the technology MS wants it
> to use, is to spin NT into a separate company whose only like to MS is
> something like royalty payments for each copy shipped. Fat chance
that'll
> happen.

Well, I agree with your last sentence anyways.

-Todd

[Michael Dillon]

In article <01bb5197.2ee56e00$ed8d389d@toddn40>,

Todd Needham [Microsoft] <to...@microsoft.com> wrote:

>> Where do you get this crazy idea? The commercial world runs primarily on
>> MVS using IBM mainframes. Second most common is UNIX on workstations and
>> servers. Windows only rules the desktop, like a glorified typewriter and
>> calculator. But the real work of business is done on the UNIX or MVS
>> database systems.
>
>Should have been more clear, sorry. This thread was following development
>tools. "Windows only rules the desktop" is several orders of magnitude
>larger installed base than MVS mainframes

No it's not. The number of users served by MVS is roughly equal to the
number of users served by Windows. That's because a single MVS machine can
serve several hundred users whereas a Windows desktop can only serve one
user.

> (and at least one or two greater
>than UNIX on the desktop).

UNIX is not really a desktop OS. Most UNIX systems are still servers that
have dozens of users per computer.

> The disconnect is the demand for Windows
>programmers vs what schools are using to teach.

Schools are supposed to teach computer science, not Windows. The
programmers that they graduate are equally capable of programming for any
OS. If they only knew Windows they would have a hard time finding a job.
The majority of programmers in the world still program in COBOL. But CS
grads tend to want to do more interesting stuff than COBOL or Windows
office stuff. Maybe that's why they use UNIX so much.

>> The "disconnect" problem you have is that you are like the Detroit
>> executives in the early sevnties who drove gas guzzlers to work every
>day
>> and parked in auto company parking lots full of gas guzzlers. Meanwhile,
>> in California, everybody was buying Toyota, Datsun, Honda, etc...
>
>Really bad analogy. "PC's are to mainframes what gas guzzlers are to
>Hondas/Toyotas/etc.".

Nope. Honda/Toyota wiped out the gas guzzler. But the PC hasb't wiped out
the mainframe. Mainframe sales are rising at between 50 to 60 percent per
year right now. Like TV didn't wipe out radio either.

The analogy I was getting at was you guys up in Seattle really don't have
a clue what is really going on in the world of Information Systems because
you have too much "Windows Everywhere". But in the real world, Windows is
just a glorified calculator/typewriter/terminal. It's an office machine.
If people want to do computing they run UNIX or OS/2 or MVS.

One more question. How on earth does anything you have to say here
relate to Linux. You seem to be talking about Windows, about development
tools, about the education system. None of this is relevant to Linux.
Maybe you should be over in a Windows newsgroup or hanging out in the
college newsgroups...

--
Michael Dillon ISP & Internet Consulting
Memra Software Inc. Fax: +1-604-546-3049
http://www.memra.com E-mail: mic...@memra.com

[Todd Needham [Microsoft]]

You're right. Should have just said "cross platform support". Color me
stupid.

-Todd

[Evan Leibovitch]

In article <adtDsG...@netcom.com>,
Anthony D. Tribelli <a...@netcom.com> wrote:

>Michael Dillon (mic...@memra.com) wrote:

>: There are *MORE* installed Linux systems

>: than there are installed NT systems. The Linux market is actually larger.

>Incredibly naive and very wrong with respect to commercial software.

But that isn't the only necessary context.

>There are more linux user's, but they don't constitute a "market" since
>most will not spend any money.

We shall see. If there was no (potential) market then why are a growing
number of companies such as Corel, Delrina and McAfee creating Linux
ports? This market is indeed not very mature, it's hardly been tapped.
Only recently have ISVs even been formally approached, the Linux community
has never done much to sell itself (outside of occasional fervent personal
evangelism :-).

>Most linux users will get along quite well
>with the freely available software. In this context, WinNT does have a
>larger "market" in spite of fewer users.

I don't follow the logic behind this. If there are more Linux
installations than NT, a smaller proportion of Linux sites need
to require commercial software XYZ to have it sell as well as an NT
version. Then again, the fact that Linux software is most often free
of cost doesn't mean there's no market, it just means that the
competition is particularly firce and margins are *very* slim :-).
There are plenty of gaps in what one can get in freeware, most
significant that comes to mind is a robust SQL Server. There is
money to be made from Linux installations if the product and the
price is right.

In any case, the real point (which is better described as "installed
base" rather than "market") is that Linux has a significant number
of users; this has benefits, not only for software vendors, but also
for those looking to market their own Linux skills and services. Even a
site running free software needs administration and care. this point
also helps counter the Microsoft publicity juggernaut once people
look behind the Rolling Stones voice-overs.

--
Evan Leibovitch, Sound Software Ltd., located in beautiful Brampton, Ontario
Caldera Business Partner / SCO Authorized VAR / ev...@telly.org / (905) 452-0504
Economists have successfully predicted 14 of the last 2 recessions

[Doug Ridgway]

In article <01bb523a.3fd9a6b0$ed8d389d@toddn40> "Todd Needham [Microsoft]" <to...@microsoft.com> writes:

>
>Web servers? Hey, check out the recent reviews on, say, Microsoft IIS on
>'NT vs. some of the web servers available for UNIX. While folks will
>argue performance vs. features endlessly, we win the ease of installation
>category every time.

Microsoft has a long, long way to go in the internet business. I was at
Microsoft's own web site the other day, looking for their ``NT for Unix Geeks''
white paper. Available only in, you guessed it, Word 6.0 format. _Real_ wired.
But don't worry, you could get a free viewer -- for W31 or W95 only. Pretty
astonishing, for a document ostensibly aimed at ``those who have at least
seen a Windows machine running''.

So I went to comments page, a little CGI form. Unfortunately, they appear
to be trying to use BackOffice, because after I clicked submit, I got a
nice page full of error messages about failure to convert blah blah blah.

Of course, I can't submit a bug report, cuz that's the form for commenting.
No wonder Bill Gates can claim that MS software is bug-free -- ``Haven't
heard any complaints''.

CGI is not exactly rocket science. If Microsoft can't get their own site
working, it's hard to see how they can in good conscience be trying to sell
this to customers. If it doesn't have to work, I can make anything easy to
install.

May I gently suggest that Microsoft use any of the standard, well tested,
widely used Web servers available. A web site for a major corporation
is too important to screw up. I think Apache is currently the market leader,
you can try it for free, and if you like it, you can keep it for free.
You get source, so you can even patch it to make it claim that it's BackOffice
or whatever...

[William Edward Webber]

Hoi Yan Lydia Lam (lyd...@gatekeeper.digital.com) opined:
:
: On 1 Jun 1996, Ariel Mazzarelli wrote:
: >So what will microsoft port on linux?

:
: Nothing. M$ cannot make any money selling their technologically-inferior
: hash to us. They do not want to play a game they will definitely lose.

:
Hmm, I've seen *a lot* of people on Linux groups wishing there were
a Linux version of MIE. Netscape didn't become bearable on my
machine until I upgraded to 40MB RAM :(.
--
William Webber Postgrad. Dip. in CS, RMIT, Australia
w...@yallara.cs.rmit.edu.au http://www.ozemail.com.au/~wew/
Interests: linux, chess, hermeneutics
"Well, I'll say this for you, the quality of your stupidity is rising"
- Lucy, from Peanuts.

[A Shelton]

a...@netcom.com (Anthony D. Tribelli) writes:

>A Shelton (ashe...@yallara.cs.rmit.EDU.AU) wrote:
>: The particular example I was thinking of however was the fact that visual
>: C++ (I think, I don't use these products) for 32bit only gets sold as
>: an NT version.

>Visual C++ has supported Win95 for quite a while, older versions were
>WinNT specific.

Hm.. That's not what the developer I spoke with said.. but then I recall
he doesn't use win95.

>Just out of curiousity, why is it that you keep commenting on topics (DOS
>extenders, game development, Visual C++, ...) you have no knowledge of or
>no experience with? I think this is the third time in recent history I've
>seen you use qualifications like: I don't have, I don't use, leads me to
>believe, ... You could do a little research before commenting, or at least
>let us know when your guessing in the original posts. :-)

And you tend to argue about the argument rather than facts.

I don't invent opinions, and I will indicate where the information
is second hand so that knowledgeable people can correct me. It's
called discussion.

As to arguing with windows people I certainly am not going to run it
(even assuming I had 4-5000 (Dos4GW) and 2000 (VC++) just for
`research'.

>: Yes, It is not quite as broken. However most people accept the fact that
>: microsoft uses an emulation layer to avoid the requirement of actually
>: writing macintosh code.

>Actually, it's not too terrible an approach. A big chunk of software is
>not system dependent (the core application code) or not performance
>critical (dialog box waiting for a user to type or click). Where
>Microsoft may have screwed up with this approach is that they may have
>skipped the performance tuning step where you identify the performance
>critical areas and add conditional compilation. For fun I recompiled some
>Windows based molecular visualization code for the Macintosh and after I
>replaced Win GDI polygon and circle calls with Apple QuickDraw calls the
>program ran twice as fast as the untuned original recompile, and as far
>as my stopwatch relexes permit I believe the tuned code runs about as
>fast as our native Mac version compiled with MetroWerks.

I run linux remember, I realize what /usr/src/linux/arch means..I
also know that there should be minimal cost to performance for this
process.

You seem to be suggesting that MS is incapable of managing
something so simple as a software port and shipped a major product
without sufficient testing... I agree..

Perhaps we can assume they actually emulated the windows API on the
macintosh so they would have to do no code modification at all?

--
Linux, because raw power can be....addictive.
Andrew Shelton s940...@yallara.cs.rmit.edu.au
GCS(2.1)-d+H+sw+v-C++UL+>L+++E-N++WV--R++tv-b+D++e+fr*y?

[Joe Sloan]

In article <01bb523a.3fd9a6b0$ed8d389d@toddn40>,

Todd Needham [Microsoft] <to...@microsoft.com> wrote:

>So tell me: why hasn't UNIX taken over the world? Why do so many people
>run Windows at home? At work? And anywhere UNIX is running?

Gee Todd, I thought you would know about this for sure!

First of all, how do you define the "world"?

If we're speaking of the internet, then you ought to know as well
as I that it is powered almost exclusively by non-microsoft machines.

If we're talking about the average home PeeCee, I,m sure you are well
aware of the stranglehold that microsoft has on that market, which came
about through a bit of good luck and a lot of hard-nosed business tactics.
Any new OS in that market atthis point would have to be 100 times better
than the ms products to have half a chance of succeeding, so deeply
entrenched is the giant from redmond at this point.

>Yes, there are good paying programming jobs available in UNIX. But what
>programming market is growing more quickly?

Good question.

Let's see here:
Do I want to be a VB/OLE plumber for a rather modest paycheck?

Nah, don't think so.

--
Joe Sloan - I have seen the future and it is Linux -

[bme...@bruce.cs.monash.edu.au]

a...@netcom.com (Anthony D. Tribelli) writes:

>Incredibly naive and very wrong with respect to commercial software.

>There are more linux user's, but they don't constitute a "market" since
>most will not spend any money.

Uhm, are you sure? Regarding me personally, you are right wrt software ---
I haven't touched a spreadsheet in all my life, and I haven't done more than
about 20 pages on a word processor since discovering TeX. That's because
I either don't need to do this stuff or have superior tools which I do
understand.

However, I _am_ a major customer at my local hardware shop (they started
giving me 10% discounts without me even asking :-), and they know that the
phrase "yes, it comes with Windows drivers" will earn them a pitying smile
and nothing else from me, while "The Sony drive is mentioned in the kernel
source as being able to read raw audio data" will make me shell out
considerable money for a Sony76E. They know they can't sell a Matrox
Millenium to me, but could sell an ELSA card without much problems (well,
if I was currently in the market for a graphics card).

I have also bought a copy of the Caldera network desktop for its Novell
support, as I have to interface with a Novell network. If someone writes
a really good WYSIWYG interface to LaTeX, I will gladly pay for it. If
someone brought out SimCity 2000, Civilization or Railroad Tycoon for linux,
my order would be guaranteed.

Bernie

--
==============================================================================
I just switched back yo my good old home system. I might have lost articles
in the process, and there might be stuff that is not set up properly yet.
Please bear with me...... ;-)

[Todd Needham [Microsoft]]

> >Should have been more clear, sorry. This thread was following
development
> >tools. "Windows only rules the desktop" is several orders of magnitude
> >larger installed base than MVS mainframes
>
> No it's not. The number of users served by MVS is roughly equal to the
> number of users served by Windows. That's because a single MVS machine
can
> serve several hundred users whereas a Windows desktop can only serve one
> user.

It's clear the thrust of this thread got lost: Development tools. The
market for development tools (second sentence, first paragraph above). I
was trying to say that "Windows only rules the desktop" is several orders
of magnitude larger development tools market than the MVS development
tools market. The dominant software development platform today is
Windows.

And by the way, a large percentage of those users being served by MVS are
running, you guessed it, Windows.

> Schools are supposed to teach computer science, not Windows. The

Yesssss, and if you teach computer science to some extent using Windows,
you're accomplishing two objectives at once: teaching computer science
and exposing students to the dominant OS.

> programmers that they graduate are equally capable of programming for
any
> OS. If they only knew Windows they would have a hard time finding a job.

Never in any of my posts, have I suggested they teach only Windows. And
no, they're not equally capable of programming for any OS. UNIX being the
primary OS in higher education means students are well prepared and
experienced in developing for that platform. One of the major problems
faced by the PC software development industry today is the 3-6 month
training period we put new graduates through teaching them to develop
software in a PC environment. This could be significantly reduced by
earlier exposure to Windows at the University level. Obviously
universities are seeing this same demand or we wouldn't see Windows being
used in research at MIT, Cornell, Berkeley and other top CS schools.

> The majority of programmers in the world still program in COBOL. But CS
> grads tend to want to do more interesting stuff than COBOL or Windows
> office stuff. Maybe that's why they use UNIX so much.

Excuse me, but while there is still a huge installed base of COBOL code,
the majority of programmers entering the market today do not program in
COBOL. The single largest market for new programmers today is PC's (and
they're not using COBOL).

> >Really bad analogy. "PC's are to mainframes what gas guzzlers are to
> >Hondas/Toyotas/etc.".
>
> Nope. Honda/Toyota wiped out the gas guzzler. But the PC hasb't wiped
out
> the mainframe. Mainframe sales are rising at between 50 to 60 percent
per
> year right now. Like TV didn't wipe out radio either.

Your point?

> The analogy I was getting at was you guys up in Seattle really don't
have
> a clue what is really going on in the world of Information Systems
because
> you have too much "Windows Everywhere". But in the real world, Windows
is
> just a glorified calculator/typewriter/terminal. It's an office machine.

> If people want to do computing they run UNIX or OS/2 or MVS.

Right. I personally tend to measure the real world by the market.
Market=real world, thank you very much.

Next I suppose you're going to tell me that DEC, IBM, HP and Compaq don't
have a clue what's going on in the world of Information Systems as well?
These are the companies SELLING 'NT into IS.

> One more question. How on earth does anything you have to say here
> relate to Linux. You seem to be talking about Windows, about development
> tools, about the education system. None of this is relevant to Linux.
> Maybe you should be over in a Windows newsgroup or hanging out in the
> college newsgroups...

Actually, if you trace this thread back to the start, I was merely
correcting mis-perceptions posted concerning Windows NT. Everything else
has grown out of answering questions and correcting further
mis-perceptions. And given that this is one of the biggest threads
currently running in this newsgroups, there seems to be significant
interest...

-Todd

"No conversions being attempted to the Church of Microsoft."

[Todd Needham [Microsoft]]

> If we're speaking of the internet, then you ought to know as well
> as I that it is powered almost exclusively by non-microsoft machines.

If you're saying the majority of servers on the Internet today are not
PC's, I'd agree. If you're saying the majority of client machines
accessing the Internet today are not PC's, I'd disagree. Is there a point
here?

> If we're talking about the average home PeeCee, I,m sure you are well
> aware of the stranglehold that microsoft has on that market, which came
> about through a bit of good luck and a lot of hard-nosed business
tactics.
> Any new OS in that market atthis point would have to be 100 times better

> than the ms products to have half a chance of succeeding, so deeply

> entrenched is the giant from redmond at this point.

I'll take that as a compliment to Microsoft, IBM, HP, DEC, Compaq, Dell,
Gateway and all the others who've worked so hard to make the PC a
standard.

> Let's see here:
> Do I want to be a VB/OLE plumber for a rather modest paycheck?

Nothing wrong with that. You can be a high paying C++ programmer instead.
Or you could be a UNIX plumber for a rather modest paycheck. Windows
programmers and UNIX programmers both run the gamut of the pay scale
('course, I'd wager there are more millionaire Windows programmers out
there than UNIX).

-Todd

[Todd Needham [Microsoft]]

> I don't invent opinions, and I will indicate where the information
> is second hand so that knowledgeable people can correct me. It's
> called discussion.

Sure, you indicate where it's second hand after you're called on it
(re-read your original post on VC++ only being available on 'NT).

-Todd

[Darin Johnson]

> Right. I personally tend to measure the real world by the market.
> Market=real world, thank you very much.

Yeah right. It's this attitude that says "if the majority of the
market is 70%, we can ignore the other 30%". Get real, the market is
the people willing to spend money on a product, and even if they're
the minority they're still the market.

The real world, by your logic, is composed of 386's. Because that's
the market. Pentium systems, or macs, or risc machines, or
mainframes, are used only by a smaller segment, and thus less
real-world. The "real world" by this logic has no use for Win95 or
NT, because the market is people that have Win3.1 on 4 meg machines.
By this attitude, nothing new would ever happen! Are 30% or more of
the people living in a fantasy realm? I don't think so.

You get a distorted view of what the real world is if you only look at
a slice of the total market. And when people say "the market" in the
software industry, they do not mean the total market.

Luckily, there are companies out there that do not use the same logic,
and who make good money selling to the minority. In some industries,
it's considered outstanding to even get 5% of the market or less.

--
Darin Johnson
djoh...@ucsd.edu O-

"You used to be big."
"I am big. It's the pictures that got small."

[Todd Needham [Microsoft]]

> Microsoft has a long, long way to go in the internet business. I was at
> Microsoft's own web site the other day, looking for their ``NT for Unix
Geeks''
> white paper. Available only in, you guessed it, Word 6.0 format. _Real_
wired.
> But don't worry, you could get a free viewer -- for W31 or W95 only.
Pretty
> astonishing, for a document ostensibly aimed at ``those who have at
least
> seen a Windows machine running''.

Whoa. Major faux pas on our webmasters part!

I've done a quick HTML conversion of the document in question. It's
available as http://internet.microsoft.com/tools/tmp/nt4unix.htm

The graphics are now crap, but everything else seems to have come across
well. Email me your shipping address if you want a printed copy or give
me another file format you can accept.

> So I went to comments page, a little CGI form. Unfortunately, they
appear
> to be trying to use BackOffice, because after I clicked submit, I got a
> nice page full of error messages about failure to convert blah blah
blah.

Strange. I just tested it with both Netscape Navigator and our Internet
Explorer 3.0 beta 1 and it worked fine under both. Give it another shot.

Also, at the bottom of the feeback page is an email alias to use in case
your browser doesn't support forms (and no, I'm not saying that was your
particular problem).

Our web server has been incredibly stable which is commendable given that
it's been our biggest beta test site for new releases and that according
the CERN folks, we're the 8th most heavily trafficked site.

-Todd

[A Shelton]

"Todd Needham [Microsoft]" <to...@microsoft.com> writes:

oops, wasn't using a question mark to the MS standard....

blah blah blah (VC++ only on NT?) blah blah

would surely be taken as indicating a question being asked. If you
have further problems on how this works feel free to take it to
e-mail.

>-Todd

[A Shelton]

"Todd Needham [Microsoft]" <to...@microsoft.com> writes:

>> Schools are supposed to teach computer science, not Windows. The

>Yesssss, and if you teach computer science to some extent using Windows,
>you're accomplishing two objectives at once: teaching computer science
>and exposing students to the dominant OS.

So what? It's just the particular flavor of topping on the core OS
services.

Besides, Unix is mostly used because it's the cheapest means to support
hundreds of users at once, the fact that it can be used in teaching
comes second.

>> programmers that they graduate are equally capable of programming for
>any
>> OS. If they only knew Windows they would have a hard time finding a job.

>Never in any of my posts, have I suggested they teach only Windows. And
>no, they're not equally capable of programming for any OS. UNIX being the
>primary OS in higher education means students are well prepared and
>experienced in developing for that platform. One of the major problems
>faced by the PC software development industry today is the 3-6 month
>training period we put new graduates through teaching them to develop
>software in a PC environment. This could be significantly reduced by
>earlier exposure to Windows at the University level. Obviously
>universities are seeing this same demand or we wouldn't see Windows being
>used in research at MIT, Cornell, Berkeley and other top CS schools.

Universities using it for research (any examples?) is not the same
as using it for teaching.

This argument could be extended to Cobol, Novell, Oracle and many
other products...CS courses are long enough thanks. And these products,
including winXX.lib are liable to dramatic change at the vendors
convenience.

>> The analogy I was getting at was you guys up in Seattle really don't
>have
>> a clue what is really going on in the world of Information Systems
>because
>> you have too much "Windows Everywhere". But in the real world, Windows
>is
>> just a glorified calculator/typewriter/terminal. It's an office machine.

>> If people want to do computing they run UNIX or OS/2 or MVS.

>Right. I personally tend to measure the real world by the market.
>Market=real world, thank you very much.

Weak. Market==real world only at this moment. Uni's not only have to
train for the market at 3-4 years plus (graduation) but for X years
after that.

>Next I suppose you're going to tell me that DEC, IBM, HP and Compaq don't
>have a clue what's going on in the world of Information Systems as well?
>These are the companies SELLING 'NT into IS.

Compaq doesn't have an OS, All the other manufacturers have other OS's of
at least equal importance to them. I'm typing on an alpha hosted system,
but it ain't running NT.

>> One more question. How on earth does anything you have to say here
>> relate to Linux. You seem to be talking about Windows, about development
>> tools, about the education system. None of this is relevant to Linux.
>> Maybe you should be over in a Windows newsgroup or hanging out in the
>> college newsgroups...

>Actually, if you trace this thread back to the start, I was merely
>correcting mis-perceptions posted concerning Windows NT. Everything else
>has grown out of answering questions and correcting further
>mis-perceptions. And given that this is one of the biggest threads
>currently running in this newsgroups, there seems to be significant
>interest...

you haven't done any such thing. You've re-iterated the Borg Mantra
a couple of times though. Which mis-perceptions do you consider to
have answered ?

>-Todd

>"No conversions being attempted to the Church of Microsoft."

conversions are irrelevant, you will be assimilated.

[Harrison]

In article <01bb5235.ef9cb6f0$ed8d389d@toddn40>,

Todd Needham [Microsoft] <to...@microsoft.com> wrote:

>Can you (or anyone else reading this message) tell me roughly how many
>Linux users there are out there?
>
>-Todd

No, and for very good reasons.

1) Linux does not require you to register, sign a non-disclosure agreement,
or any other such nonsense. In fact, it is *encouraged* to share your
copy of the Linux OS with as many people as you like - without the
the threat of being sued (or causing international trade conflicts).

2) the UNIX paradigm is different than of NT. How many concurrent users does
NT support on a peecee box? (and I mean shell users, not *just* file/
printer serving) Sorry for the lack of specific Linux numbers, but if I
remember correctly ftp.cdrom.com uses a P6-150 and supports +1,000
concurrent users running FreeBSD (a sister OS to Linux - yet another
beautiful, free UNIX). Can NT support +1,000 concurrent shell users on
the PC platform in a 24/7 production environment? I think not. Following
this line of thought, each UNIX box (or Novell box for that matter) can
support more users than each NT box - fewer UNIX boxes are needed to
support the same user load. (and how well does NT scale compared to UNIX? :-)
BTW, I'm sending this message on a 486dx100 Linux box that has 350 accounts,
and is over a thousand miles away from where I am typing it...

You may object to this line of thought by arguing that Linux is used
predominately as a desktop (i.e. single user) OS. In the quotes of the
NT installed base, how many of those machines are NT workstation? (and
how many concurrent users does the workstation version support?)

Is there a difference between the installed base of an OS and the
number of users? I think so, and apparently so does microsoft since
they require a license for each connect to an NT server, even though
it is only one server.... just as wordperfect requires a license for
each user on our Novell lan even though there is physically only one
copy of wordperfect....

Just one more observation:

You have "microsoft" stamped all over your headers, but have no
disclaimers on your posts - I must assume that you are "on the clock" ...
I must commend you for your "spin doctor" skills. I hope bill is dishing
out a reasonable amount of his billions for your services.

Microsoft must really be worried (as it should be) about the
phenomenal growth of Linux to devote resources to posting on c.o.l.a
(and NOT crossposting, may I add...)

--

- Eric
_______________________alias harr...@econ.CSUChico.edu='Eric Harrison'

[A Shelton]

ashe...@yallara.cs.rmit.EDU.AU (A Shelton) writes:

replying to your own post is pretty lame, but I F*'d it.

>"Todd Needham [Microsoft]" <to...@microsoft.com> writes:

>>> Schools are supposed to teach computer science, not Windows. The

>>Yesssss, and if you teach computer science to some extent using Windows,
>>you're accomplishing two objectives at once: teaching computer science
>>and exposing students to the dominant OS.

>So what? It's just the particular flavor of topping on the core OS
>services.

>Besides, Unix is mostly used because it's the cheapest means to support
>hundreds of users at once, the fact that it can be used in teaching
>comes second.

The above is true, but not really fair to unix... Further thinking
about it suggested lots of reasons to teach unix.

Unix has an extremely mature toolset, an elegant API with good
tools for IPC and other core OS tasks with a minimum of fluff.
It is well documented, stable and well understood. These docs
are available from a wide number of sources other than the
vendor and are sufficiently detailed you could replicate the
OS from them.

It exemplifies open standards such as POSIX, does not have the
desktop bias of NT and doesn't force you to mess with GUI's when
you're doing hello world.

It has a philosophy of writing small tools which are as capable of
being used by a program as a user and allow powerful programs to be
easily constructed.

A great number of development tools are available free, as are
entire unix systems. Source code can be used as a teaching tool
if needed.

It can do cool things like program the sound card via
ls -lR / > /dev/audio :)

Given these, what advantages does NT offer to OS courses?

(also, you'll have to convinve uni lecturers, who only vaguely
remember the `real' world that NT is better... The quote I heard
today was "Never make me depend on a program that uses OLE, it's
designed to fail.")

[Todd Needham [Microsoft]]

> 1) Linux does not require you to register, sign a non-disclosure
agreement,
> or any other such nonsense. In fact, it is *encouraged* to share your
> copy of the Linux OS with as many people as you like - without the
> the threat of being sued (or causing international trade conflicts).

I must have missed your point. Windows NT users aren't required to
register or sign a non-disclosure either.

> 2) the UNIX paradigm is different than of NT. How many concurrent users
does
> NT support on a peecee box?

Depends on what you're doing, of course. I.e., Oracle server on 'NT and
SQL Server on 'NT have different architectures and different overheads.
Netscape server for NT and our Internet Information Server as well. I
sometimes feel like determining scaling is a black art given all the
different hardware and server applications out there.

> beautiful, free UNIX). Can NT support +1,000 concurrent shell users
on
> the PC platform in a 24/7 production environment? I think not.
Following

I'm not sure what you're getting at, but yes, we run 1000+ concurrent web
users on our web site. We've run 1000+ concurrent database users on SQL
Server. And folks other than Microsoft have 'NT running in 24/7
production environments (pop over to www.microsoft.com for some case
studies).

> this line of thought, each UNIX box (or Novell box for that matter)
can
> support more users than each NT box - fewer UNIX boxes are needed to
> support the same user load.

You've lost me here. Can you explain how you arrive at this????

> (and how well does NT scale compared to UNIX? :-)

Some UNIX variants currently scale better.

> You may object to this line of thought by arguing that Linux is used
> predominately as a desktop (i.e. single user) OS. In the quotes of
the
> NT installed base, how many of those machines are NT workstation?
(and
> how many concurrent users does the workstation version support?)

Don't object to it at all. I myself am an 'NT desktop user (though I run
'NT Server, not Workstation on my notebook). I don't argue with anything
you've said about UNIX in general or Linux in particular (though I'd
preferred you stuck to one or the other for arguments sake - it makes it
kind of hard flip flopping back and forth). On the other hand, it's
obvious you're not as well informed about 'NT.

> Is there a difference between the installed base of an OS and the
> number of users? I think so, and apparently so does microsoft since
> they require a license for each connect to an NT server, even though
> it is only one server.... just as wordperfect requires a license for
> each user on our Novell lan even though there is physically only one
> copy of wordperfect....

Two different issues: installed base and how a company chooses to license
their product.

> You have "microsoft" stamped all over your headers, but have no
> disclaimers on your posts - I must assume that you are "on the clock"
..

> I must commend you for your "spin doctor" skills. I hope bill is dishing

> out a reasonable amount of his billions for your services.

Uhm, thanx, I think. I haven't really thought of anything I've posted
here as spin doctoring, but I'll take it as a compliment anyways! This is
more curiosity and interest in the competition than a serious career
objective. And the only disclaimers I feel a need to put in are the
places it says "I personally feel...".

> Microsoft must really be worried (as it should be) about the
> phenomenal growth of Linux to devote resources to posting on c.o.l.a
> (and NOT crossposting, may I add...)

Uhm, no, again, not really. As far as I know, I'm the only MS person
posting here and it's only from curiosity. I manage MS relations with
university research groups and that's about the only place we run into
Linux. My presence here is really curiosity (I don't really expect to
convert anyone, but I like to keep the bashers honest...).

-Todd

[Todd Needham [Microsoft]]

> Yeah right. It's this attitude that says "if the majority of the
> market is 70%, we can ignore the other 30%". Get real, the market is
> the people willing to spend money on a product, and even if they're
> the minority they're still the market.

True, but the percentages are a little off. If you look at, say, people
_buying_ development tools or people _buying_ desktop productivity
applications, then we have offerings covering the majority of the market
(composed of Windows and the Mac today).

> The real world, by your logic, is composed of 386's. Because that's
> the market. Pentium systems, or macs, or risc machines, or
> mainframes, are used only by a smaller segment, and thus less
> real-world. The "real world" by this logic has no use for Win95 or
> NT, because the market is people that have Win3.1 on 4 meg machines.
> By this attitude, nothing new would ever happen! Are 30% or more of
> the people living in a fantasy realm? I don't think so.

What? We're the world's largest Mac software developer. Why would you
think we're ignoring the Mac? And those new machines that OEM's are
shipping with '95 or 'NT? Those are Pentium and Pentium PRO for the most
part (and the percentage is growing all the time). I also don't hear
anyone at Microsoft complaining about our upgrade rate on Win3.1 systems
(did we think it was going to happen over night? No. Do we believe
everyone can or will upgrade? No. Do we still sell 16-bit applications
for them? Yes.). Growing the market means there are transitional periods
and we certainly take that into account. The trick is properly
forecasting the transitions (to date, we've done pretty well).

> You get a distorted view of what the real world is if you only look at
> a slice of the total market. And when people say "the market" in the
> software industry, they do not mean the total market.

No offense, but I think Microsoft is as aware of the total market as
anyone. We are, however, careful how we deploy our finite development and
marketing resources (and as a shareholder, I salute Microsoft for the job
it's done to date). If I'm misunderstanding you, please correct me.

> Luckily, there are companies out there that do not use the same logic,
> and who make good money selling to the minority. In some industries,
> it's considered outstanding to even get 5% of the market or less.

And more power to 'em! (Though I'd argue they use the same logic we do,
just targeting different markets). Did you think Microsoft was going to
own everything?

-Todd

[Todd Needham [Microsoft]]

> Universities using it for research (any examples?) is not the same
> as using it for teaching.

It's not the same as using it for teaching, but then again, I'm involved
in research, not curriculum (I'm sure others at Microsoft could help you
with the later if you're interested).

Examples:

Thorsten van Eicken, Werner Vogels, Stephen Vavasis, Dean Krafft at
Cornell
Neil Gershenfeld and Joshua Smith at the MIT Media Lab ("Things that
Think project")
Carl Hewitt, Carl Manning, Michael Oren, Wissam Y Ali-Ahmad, Steven
Lerman, Sepehr Kiani, Stephen Garland at MIT (mostly LCS)
Stefan Nilsson, Mike Stonebreaker, Tom Anderson, Robert Wilensky (Chair),
Scott Hernandez, Ferenc Kovac, Brian Barsky (and several dozen others
there).

Check out www.research.microsoft.com in the coming weeks for a more
comprehensive (though by no means complete) set of pointers to Microsoft
technology in university research.

> This argument could be extended to Cobol, Novell, Oracle and many
> other products...CS courses are long enough thanks. And these products,
> including winXX.lib are liable to dramatic change at the vendors
> convenience.

True. This is part of the reason we're working with schools on electronic
distribution of materials to help them keep up-to-date.

> Weak. Market==real world only at this moment. Uni's not only have to
> train for the market at 3-4 years plus (graduation) but for X years
> after that.

And we're trying to shorten that cycle.

> >Next I suppose you're going to tell me that DEC, IBM, HP and Compaq
don't
> >have a clue what's going on in the world of Information Systems as
well?
> >These are the companies SELLING 'NT into IS.
>

> Compaq doesn't have an OS, All the other manufacturers have other OS's
of
> at least equal importance to them. I'm typing on an alpha hosted system,
> but it ain't running NT.

What's your point? These are still companies selling Windows NT into the
enterprise.

> >Actually, if you trace this thread back to the start, I was merely
> >correcting mis-perceptions posted concerning Windows NT. Everything
else
> >has grown out of answering questions and correcting further
> >mis-perceptions. And given that this is one of the biggest threads
> >currently running in this newsgroups, there seems to be significant
> >interest...
>

> you haven't done any such thing. You've re-iterated the Borg Mantra
> a couple of times though. Which mis-perceptions do you consider to
> have answered ?

If you bothered to actually check before unleashing blind emotional
rhetoric, you'd find that (a) all of this can be traced back to my
correcting mis-perceptions from posts by Chris Petit and Josh Stern
regarding 'NT in response to a question by Julian Regel and (b) this is
the only thead in which I've participated.

Take a deep breath and repeat after me: it's just a discussion alias,
it's just a discussion alias, it's just a discussion alias...

-Todd

[Todd Needham [Microsoft]]

> Unix has an extremely mature toolset, an elegant API with good
> tools for IPC and other core OS tasks with a minimum of fluff.
> It is well documented, stable and well understood. These docs
> are available from a wide number of sources other than the
> vendor and are sufficiently detailed you could replicate the
> OS from them.
>
> It exemplifies open standards such as POSIX, does not have the
> desktop bias of NT and doesn't force you to mess with GUI's when
> you're doing hello world.

Jeez, Andrew. Can't you be happy to stick with espousing "facts" on a
topic you're knowledgeable about?

'NT passed Government POSIX certification and is being sold into the gov't
on POSIX compliant contracts; 'NT doesn't have a desktop bias and is
available in a superset specifically servicing the server market and
lastly, 'NT doesn't require you to use the GUI to do a "hello world".

-Todd

[Colin Smith]

drsoran (drs...@ni.cba.csuohio.edu) wrote:
: Todd Needham [Microsoft] (to...@microsoft.com) wrote:
: : 3. Out of the box, NT (like Linux) contains a full networking environment
: : (including TCP/IP with sockets and Novell compatible IPX/SPX). Nothing
: : you need to add for networking unless you need NFS (available from a
: : number of vendors).

: How are NT's out of the box DNS and http servers? Do you guys

You missed the network transparent Windowing system - ala X which is free under
Linux. How much is NFS for NT? How much is an X server?

: have any discounts where a student can obtain NT on a CD for $7? How
: about $50? I was flipping through an Egghead catalog and they had a
: really hilarious typo.. they had NT Server listed for $700.. sheesh..
: they should proof read those things before they print them. Oh well... I
: guess I'll have to settle for Linux until Egghead gets that price fixed.. ;)

: : 5. Microsoft and other tools vendors offer significantly discounted
: : educational pricing on their development tools. We also offer the same on

Excuse me, but how much is the C compiler & editors & libs under NT? How
much under Linux?

i: : our OS's. Documentation for the OS and for our development tools is
: : available in electronic (read: full text searchable) form through the
: : Microsoft Developer Network.

Electronic documentation? Doesn't that mean you don't get any manuals?
What an innovation!! Check out

http://sunsite.unc.edu/mdw/linux.html

And how much does it cost to join the developers network?

: If the source is freely available to NT, why don't more
: universities have it? Do they have to indenture 10% of their grad
: students to become slaves at Microsoft for 10 years or something? Why
: not put the source code on an ftp site? Oh.. I forgot.. you guys are
: into licenses..

: : You and the others who responded are correct that today the lion's share
: : of university research occurs on UNIX variants. However, the commercial
: : world runs on Windows. It's a disconnect that I hope to see resolved....

I'm all for it! Everyone should run Linux! Unless you like FreeBSD of course.

: I do as well.. Keep up the good work.. I'm sure within the next
: 10 years you'll have the commercial world running back to UNIX..
--
Linux; 64bit, multi-tasking, multi-user and Free.
-----------------------------------------------------------------
As Will Rogers would have said, "There is no such things as a free
variable."
-----------------------------------------------------------------
Linux mellifluous 1.2.13 #4 Fri May 31 08:32:30 BST 1996 i486
2:57pm up 6 days, 6:07, 2 users, load average: 0.38, 0.13, 0.10
-----------------------------------------------------------------
Colin Smith (co...@mellifluous.europe.dg.com)
-----------------------------------------------------------------
My opinions are mine and not necessarily anyone else's.

[Matt Kennel]

Todd Needham [Microsoft] (to...@microsoft.com) wrote:

: > What I meant was that the temptation of utilising programming tools
: > for market manipulation. Such as prematurely retiring old tools to
: > force upgrades, making premier tools available only on the desired
: > platforms (VC++ 32 is only on NT?), upgrading tools primarily to
: > react to marketing needs and needless incompatibility to existing
: > standards.

: While a number of competitors have claimed MS has a monopoly on OS (and it
: seems no one in this newsgroup thinks so), I've don't believe I've ever
: heard anyone say we have a monopoly on development tools.

You do have a monopoly on churning the API in such a way to make
competitor's development tools potentially unstable or obsolete,
you must admit.

You do have a monopoly on knowing which API parts are about to be
upgraded and so you can direct your development group to start
working in direction X vs direction Y.

Historically microsoft has been good at development tools
(e.g. applesoft basic) and general purpose application programs
(first version of Mac Excel).

The problem is when they try to define and propagandize
overall architecture, which I don't believe they make very
well. This is we get the *screw*.

It's too bad, because it would be nice to have an open
and free Microsoft creating development tools and applications
in a fully competitive market for the brave new world of Java
and a fantastic and amazing fluid world of distributed objects
and fundamentally network centric operating systems.

But noo!! Everybody at microsoft must pay homage at the
Edifice of Windows and OLE, for better or worse. I think
it's "for worse". Look at Ballmer, he's sounds like a
a Goebbels for the Windows Party! Screw that.

Microsoft should break up into OS and Development.

Development
can move ahead and stay up, and OS can hang around servicing its
legacy customers like IBM does for its obsolete mainframe operating
systems.

This would be good for microsoft, free markets, and the industry
in general.

: -Todd

--
Matthew B. Kennel/m...@caffeine.engr.utk.edu/I do not speak for ORNL, DOE or UT
Oak Ridge National Laboratory/University of Tennessee, Knoxville, TN USA/
*NO MASS EMAIL SPAM* It's an abuse of Federal Government computer resources
and an affront to common civility.

[Bob Nelson]

Harrison (harr...@rocko.lab.csuchico.edu) wrote:

> Microsoft must really be worried (as it should be) about the
> phenomenal growth of Linux to devote resources to posting on c.o.l.a
> (and NOT crossposting, may I add...)

Microsoft isn't concerned one iota about linux. As Todd has pointed
out, it has yet to register as even a blip on the radar screen of that
large entity. OTOH, linux is very well known to Dick Matthew Stallman
and his cronies at the "free" software foundation as evidenced by their
recent "re-education" campaign. linux has little to fear from what Bill
does in Redmond. What's going down with Dick in Cambridge, though, is a
matter of great concern.

--
=============================================================================
Bob Nelson: Dallas, Texas, U.S.A. - bne...@netcom.com
Linux for fun, M$ for $$$...and the NFL for what really counts!
=============================================================================

[Darin Johnson]

"Todd Needham [Microsoft]" <to...@microsoft.com> writes:

> > If we're speaking of the internet, then you ought to know as well
> > as I that it is powered almost exclusively by non-microsoft machines.
>
> If you're saying the majority of servers on the Internet today are not
> PC's, I'd agree. If you're saying the majority of client machines
> accessing the Internet today are not PC's, I'd disagree.

Just a nit, but I'd expect someone from "microsoft.com" to realize
that PC's can be non-microsoft machines, so to start talking about
"PC's" in your reply, especially on a Linux newsgroup, is a bit
strange.

> > Any new OS in that market atthis point would have to be 100 times better
>
> > than the ms products to have half a chance of succeeding, so deeply
> > entrenched is the giant from redmond at this point.
>
> I'll take that as a compliment to Microsoft, IBM, HP, DEC, Compaq, Dell,
> Gateway and all the others who've worked so hard to make the PC a
> standard.

You can take it that way, but that's not the way I read it.
(people who worked hard to make a collection of hardware bugs into a
standard don't deserve compliments)

--
Darin Johnson
djoh...@ucsd.edu O-

The trouble with conspiracy theories are that they assume
the government is organized.

[Colin Smith]

bme...@bruce.cs.monash.edu.au wrote:
: a...@netcom.com (Anthony D. Tribelli) writes:

: I have also bought a copy of the Caldera network desktop for its Novell

: support, as I have to interface with a Novell network. If someone writes
: a really good WYSIWYG interface to LaTeX, I will gladly pay for it. If

It seems that LyX is only free.

[Anthony D. Tribelli]

Joe Sloan (j...@digital.netvoyage.net) wrote:
: Anthony D. Tribelli <a...@netcom.com> wrote:

: >Incredibly naive and very wrong with respect to commercial software.

: >There are more linux user's, but they don't constitute a "market" since

: >most will not spend any money. Most linux users will get along quite well

: >with the freely available software. In this context, WinNT does have a
: >larger "market" in spite of fewer users.

: Tell me, Tony, how did you arrive at this conclusion? I mean where you
: proclaim that "most linux users" will not spend any money.

I used "will not" in the context of "not needing to", not in the context
of "having a philosophical objection to".

: Could you share your research with me? I'd really like to examine the
: statistics you collected.

Many linux users, possibly including yourself :-), have in the past been
very happy to provide references to the various free word processing,
spreadsheet, etc... programs available for linux.

Linux advocates claim millions of users, how do WordPerfect for linux
sales compare to WordPerfect for Windows sales?

: Don't kid yourself Tony - Linux users will pay good money for software
: if they think the quality is up to snuff...

You are ignoring an important point, the user needs to be dissatisfied
with the freely available software. I suspect this occurs more often with
DOS/Win freeware than with linux freeware.

Tony
--
------------------
Tony Tribelli
adtri...@acm.org

[Darin Johnson]

> > Luckily, there are companies out there that do not use the same logic,
> > and who make good money selling to the minority. In some industries,
> > it's considered outstanding to even get 5% of the market or less.
>
> And more power to 'em! (Though I'd argue they use the same logic we do,
> just targeting different markets). Did you think Microsoft was going to
> own everything?

I was not talking about Microsoft here, I was talking about the
industry in general. Most companies make one decision when they start
"will we support Win3.1 or Macs", and don't even bother thinking about
unix or os/2 or gem or whatever. Yes, it's a pain in the butt to make
things portable, and the MS ability to release a new standard every
month that no one else uses doesn't help; but if you get good
programmers (don't get me started on the quality of C++ programmers),
portability can be made much simpler (heck, just get programmers that
have used more than one machine or system in their lives and require
win32s and use a portable gui tool).

Although most people out there have heard of or seen a Mac, you can
usually find a max of 1 shelf at most major stores for them, certainly
giving the prospective buyer the idea that they're not well supported,
and many will even have the notion that popularity implies quality.
And I've only once seens OS/2 software sold in a store (although OS/2
itself is common enough, especially the version w/o Windows). And
then reading the magazines and weekly rags, you'd be even more hard
pressed to see that anything bug Win existed (Apple gets reported on
whenever the stock takes a dip though). And all of this is justified
under the heading of "the majority of the people use windows" (the
majority certainly does not use NT, but NT gets plenty of media
writeups, I guess they make exceptions if it's Windows). I've
actually seen people tell customers (real customers, those ready to
write 6 figure checks) be told "oh, that system's on the way out".

And I am sick of being told that I am not a "real user" or that I
don't count. Tyrany of the majority it seems.

--
Darin Johnson
djoh...@ucsd.edu O-

Support your right to own gnus.

[Anthony D. Tribelli]

Evan Leibovitch (ev...@telly.telly.org) wrote:
: Anthony D. Tribelli <a...@netcom.com> wrote:

: >Michael Dillon (mic...@memra.com) wrote:

: >: There are *MORE* installed Linux systems
: >: than there are installed NT systems. The Linux market is actually larger.

: >There are more linux user's, but they don't constitute a "market" since
: >most will not spend any money.

: We shall see. If there was no (potential) market then why are a growing
: number of companies such as Corel, Delrina and McAfee creating Linux
: ports? This market is indeed not very mature, it's hardly been tapped.

I never claimed there is no linux market. My argument is that the
statement claiming linux has a larger market is incorrect.

: >Most linux users will get along quite well

: >with the freely available software. In this context, WinNT does have a
: >larger "market" in spite of fewer users.

: I don't follow the logic behind this. If there are more Linux
: installations than NT, a smaller proportion of Linux sites need
: to require commercial software XYZ to have it sell as well as an NT
: version ...

True, but my argument is that linux user's are much less likely to need to
buy commercial software than WinNT users and that their raw numbers do
not make up for this.

[Josh Stern]

Todd Needham [Microsoft] <to...@microsoft.com> wrote:

>> >Actually, if you trace this thread back to the start, I was merely
>> >correcting mis-perceptions posted concerning Windows NT. Everything
>else
>> >has grown out of answering questions and correcting further
>> >mis-perceptions. And given that this is one of the biggest threads
>> >currently running in this newsgroups, there seems to be significant
>> >interest...

> (a) all of this can be traced back to my

>correcting mis-perceptions from posts by Chris Petit and Josh Stern
>regarding 'NT in response to a question by Julian Regel

I think its unfortunate that some people have used the occasion
of your postings to unlease a torrent of wild and breathless agitation
against Microsoft, but I object to your claim that my original
and only posting to this thread, advocating Linux for a CS-related
university environment, contained any "mis-perceptions" regarding
Windows NT or any other Microsoft product. I continue to believe
that the advantages that I cited for Linux, including source
code availability for ~99% of everything that runs on Linux,
low cost (particularly when we consider that students will want
to have OS, development tools, documentation, and networking, both
in their labs and on their home machines), superior networking, and
compatibility with the standard platform of Computer Science related
research in academia are very significant for the application context
that was being discussed.

You followed my posting with a set of remarks, attempting to
mitigate these advantages for Linux in comparison to NT and to point
out areas where you thought NT held some advantage. I didn't
agree with some of your arguments - in particular, I would dispute
your claims of equivalent out of the box networking and superior
development tools - but I didn't follow up on it. This was
partly because I was quite busy last week, partly because I happen
to agree with the idea that students should be exposed to both
platforms, and partly because I thought that your remarks could
basically have been paraphrased as "Yes, Linux has some advantages
in all of the mentioned areas, but NT is okay too, and besides we are
taking over the World so it behooves you to deal with it,"
and I personally disapprove of the strong tendency in contemporary
discourse to automatically translate every debate that is prima facie a
discussion of merit to a debate about either historical
or prognosticated future economic performance of some variety.
Nevermind though, I feel like making an exception today.

In my previous posting I restricted myself to responding to
the specifics of Julian Regal's request for pro-Linux advocacy
information - btw, your response to said request was therefore,
ipso facto, off-topic :). Anyway, since the ensuing conversation
drifted, as these conversations so often do, to a general
comparison of OS and prognosis for growth, I thought
I would post again to point out that despite
the fact that Linux and Windows NT are very different,
there are also many close parallels between them that
portend strong future growth. These include the
following features.

1) Low Cost - Both benefit from low cost wrt other OS's.
Of course Linux is free (hence cheaper), but NT benefits
from low cost compared to some other commercial products with
comparable workstation and server capabilities

2) Modelled after good prototypes - Both systems are
very young, but both were able to build on knowledge of
existing solutions: BSD, SysV Unix, and Minix in the case
of Linux, and VMS + Windows 3 in the case of NT.

3) High Volume Distribution - Microsoft has perfected
one model of distribution. Namely, get most all of the places
where non-specialists might go to purchase your type
of product to exclusively stock your brand!!!
People got DOS and WIndows 3.1 because it came with
low-cost computers, whether they asked for it or not.
Now they are getting WIndows 95 preloaded, and from what I
read in the magazines, it seems to already be a
pre-ordained conclusion that Windows NT will
accompany most Pentium Pro machines out the door.
You can claim that the acceptance of your products
is propelled by consumer demand, but the 'beauty' of
this distribution model is that the majority of consumers,
who are ignorant and uncaring of most distinctions
between OS platforms, don't have to demand anything
or even have a preference in order to wind up buying your stuff.
Rather, they would have to incovenience themselves not to
purchase a Microsoft OS. Thus, we can
observe the commercially polarized track record of Windows95,
which has, relatively speaking, flopped as a retail product (it
was no longer even advertised in the last two PC software mail
order catalogs that I received, including one from Egghead), but
nevertheless it has been an awesome success for MS, because it
maintained your stranglehold on the preload market
and served as an effective Trojan Horse for the
advance of WIN32 software (that Gates guy is f!&!in brilliant!).

Linux however, also has a highly effective model for software
distribution that can achieve high volume - that is, people can
freely download it from the internet, 24 hours a day,
7 days a week, 365 days a year, they can buy it for a
few dollars in popular bookstores, they can order the OS *and*
a gazillion programs for $25, and they can give it
away to their friends. That is not just the OS, but also
editors, browsers, comm packages, games, paint programs,
utilities, compilers, interpreters, databases, etc. etc.
I'm sure people who don't know Linux yet think "Ppff, shareware,"
because they simply have no idea of the order of magnitude
difference in quality between the GPL'd Linux works and
the sort of ratty DOS VGA-mode nagware they once got of
some godforsaken BBS back in `89.

Right now the growth rate of Linux in the US and Europe is
remarkable, but this distribution model will start to be
compararively more powerful as computers become widely available
in places like India, China, South America, and the Soviet Union.
The people who form the vanguard - who have computers in
these locations now - are definitely technical types (and they
like Linux). In these societies, $75 US for a commercial OS
(not to mention US $300 for an office suite or compiler)
is a very significant chunk of capital.
Not only that, but anyone there can modify most all of the Linux
software to fit local conditions, and in
so doing, feel that they are part owners of the
system and not merely a statistic for a foreign
capitalist (In fact they can even sell their modified versions).

Also, as Linux becomes more widely known
in non-technical circles here, I expect that some computer
system retailers will investigate pre-loading it
alongside their standard offering as a value-added
package to help distinguish their offerings from those of
their nearly identical competitors. I think they would be
particularly inclined to do this if they could rely on a company
like RedHat to handle the massive tech support effort
that would be engendered by such a program.

4) Primarily Software - Linux and NT both exist,
primarily for their own sake as software. I take this to
contrast with the case of software supplied by some
of the traditional Unix vendors who seem to have thought of their
system software primarily as a condiment to the sale of their hardware
and service contracts.

5) Massive Development Resources + Persistence - MS can afford to keep
a lot of talented developers working at full throttle on
their products. MS has a firm belief in their products and will
stick with the good ones until they turn into financial winners
(Sorry about 'Bob', you can't win them all).
Linux has some talented developers working at full throttle and a huge
and ever growing number of talented developers/true_believers/fanatics
working in most of their spare time. Take if from me, they/we
(collectively) won't quit. Unlike the case with some MS competitors,
there is no possibility that a corporate higher-up to look at the bottom
and line and think about pulling the plug on Linux development or
scaling back, as there is no corporation and no bottom line.

6) High Performance - both have good virtual memory architectures,
though Linux is measurably better and is more thrifty with resources.

Well, that's actually a lot of parallels. I don't need to
guess that Microsoft representatives would choose rather to
stress differences between the two platforms, such as the
fact that popular Microsoft applications only run on MS's
chosen platforms, and that people who are completely unfamiliar
with computers and lack basic typing skills can achieve a
minimal level of functionality more quickly in an
environment such as MS-WIN where nearly every operation
(in a limited set) is a mouse-based multiple choice.
Though important to some, these advantages
strike me as, historically speaking, much less causally significant t
than the traditional MS strengths of low cost, persistence, and control
of the high capacity distribution channels.
I hope that I have done a good job of laying out the case for why
Linux is potentially a strong contender in these areas as well.

- Josh

--
-------------------------------------------------------------------------------
jstern
jst...@primenet.com
-------------------------------------------------------------------------------

[Joe Sloan]

In article <01bb53c6$c10e9910$ed8d389d@toddn40>,

Todd Needham [Microsoft] <to...@microsoft.com> wrote:

>I must have missed your point. Windows NT users aren't required to
>register or sign a non-disclosure either.

My, you are a slippery one!
The man was talking about source code - if no nondisclosure agreement
is needed, why bother the dean with all that paperwork?

>> beautiful, free UNIX). Can NT support +1,000 concurrent shell users
>> on the PC platform in a 24/7 production environment? I think not.

>I'm not sure what you're getting at, but yes, we run 1000+ concurrent web

>users on our web site.

I should think it fairly obvious what he was "getting at" - He asked
a simple question, which you neatly sidestepped, in true microsoft
fashion. So I'll repeat his question: can windows nt support 1000+
concurrent shell users on intel hardware?

>And folks other than Microsoft have 'NT running in 24/7 production
>environments

yawn....
This sort of thing is remarkable only to those from a microsoft
background. Unix, VMS and mainframe users take this stability for granted.

>(pop over to www.microsoft.com for some case studies).

"Case studies"? yeah, right, they couldn't possibly be biased....

>> (and how well does NT scale compared to UNIX? :-)
>
>Some UNIX variants currently scale better.

What do you mean, "currently"? do you really imagine that the commercial
Unix vendors and the Linux/FreeBSD community will stand still and patiently
wait for windows nt to pass them up? Talk about your rose-colored glasses!

>Uhm, thanx, I think. I haven't really thought of anything I've posted
>here as spin doctoring, but I'll take it as a compliment anyways!

Oh please! all your postings have added up to nothing more than one
monotonous microsoft pitch - is there nothing sacred anymore? Isn't
it enough that we have to endure the blaring, incessant microsoft hype
on television and in the print media - must you constantly harass and
torment those who visit this newsgroup to discuss and learn about Linux?

Please stop this senseless waste of bandwidth. If we want to hear
microsoft propoganda, we can all find it quite easily on our own!

--
Joe Sloan
j...@netvoyage.net

[Todd Needham [Microsoft]]

Colin,

> You missed the network transparent Windowing system - ala X which is
free under
> Linux. How much is NFS for NT? How much is an X server?

Yeah, mentioned that in another thread. As for price, you'd have to check
with the multiple vendors offerings in these areas. Some pointers are
available from http://internet.microsoft.com/tools/tools.htm

> Excuse me, but how much is the C compiler & editors & libs under NT? How
> much under Linux?

I don't have a clue as to any (if there are any) _commercial_ compilers
available for Linux, so I can't give you prices. Perhaps others in this
newsgroup can point you to some.

Our Visual C++ Standard Edition has an academic price of less than $50.
If a schools installing a lab and needs licenses only, these are less than
$25.

> i: : our OS's. Documentation for the OS and for our development tools
is
> : : available in electronic (read: full text searchable) form through
the
> : : Microsoft Developer Network.
>
> Electronic documentation? Doesn't that mean you don't get any manuals?

Of course not, silly!

> What an innovation!! Check out
>
> http://sunsite.unc.edu/mdw/linux.html

Very cool site! Is there a search engine somewhere there that I missed?

-Todd

[Todd Needham [Microsoft]]

Matt,

> You do have a monopoly on churning the API in such a way to make
> competitor's development tools potentially unstable or obsolete,
> you must admit.

I'm not sure I understand. By "monopoly on churning the API" do you mean
we own the API? Then yes. I agree. As for churning the API in such a

way to make competitor's development tools potentially unstable or

obsolete" are you crazy? If we didn't provide great backward
compatibility with each new release, our customers would shoot us. And if
we didn't get early information and beta's to tools vendors you wouldn't
have folks like Borland releasing beta compilers concurrent with our beta
OS (Win95 in this instance). And you'd certainly hear development tools
vendors bitching about it on the front page of very PC rag.

> You do have a monopoly on knowing which API parts are about to be
> upgraded and so you can direct your development group to start
> working in direction X vs direction Y.

Well, if the development group doesn't have the spec, they can hardly do
the work. As for tools vendors, we have an early release program and work
very closely with them.

> Historically microsoft has been good at development tools
> (e.g. applesoft basic) and general purpose application programs
> (first version of Mac Excel).
>
> The problem is when they try to define and propagandize
> overall architecture, which I don't believe they make very
> well. This is we get the *screw*.

Great rhetoric. Any examples?

> It's too bad, because it would be nice to have an open
> and free Microsoft creating development tools and applications
> in a fully competitive market for the brave new world of Java
> and a fantastic and amazing fluid world of distributed objects
> and fundamentally network centric operating systems.

Uhm, that's what we're doing (trying to do). Certainly you've heard about
our forthcoming Java support by now...? If not, jump over to
www.microsoft.com and do a search on "java"

> But noo!! Everybody at microsoft must pay homage at the
> Edifice of Windows and OLE, for better or worse. I think

We all have our own view. Microsoft doesn't view these as antithetical to
your vision in the previous paragraph.

> it's "for worse". Look at Ballmer, he's sounds like a
> a Goebbels for the Windows Party! Screw that.

So he's enthusiastic (and let's leave the name calling out - nobody
appreciates it and it doesn't add anything to the discussion).

> Microsoft should break up into OS and Development.
>
> Development
> can move ahead and stay up, and OS can hang around servicing its
> legacy customers like IBM does for its obsolete mainframe operating
> systems.

Yeah, right. Now THERE'S a model for success.

> This would be good for microsoft, free markets, and the industry
> in general.

No it wouldn't be.