A trip to Office Max
Joe User
local chess club meets on Sundays. Since it was right there, I decided to
drop in and see what I could find out about that chart they printed in last
Sunday's advertisement claiming that Win95/98/ME did not have pre-emptive
multitasking. This was one of the advantages of upgrading to XP (or 2K,
according to the chart).
It turns out that Microsoft sent them a big sign to put on their Windows XP
display shelf. The sign had the chart printed on it, and they simply copied
the chart for their Sunday ad. And for anyone wondering, Office Max did not
make a mistake -- the chart really did say that 95/98/ME did not have
pre-emptive multitasking. In fact, the assistant manager (thinking I was
looking for information) suggested that I go to their web site. He said
they probably have that chart on there somewhere, as well as a lot more
information about this exciting new operating system. :)
I haven't been to Microsoft's web site in quite a while. I'm not going to
bother now. It has been an unpleasant experience every time I've gone
there, and I'm not about to wade through their cockeyed spaghetti looking
for a chart just so I can prove a point on Usenet.
Erik, you probably know their site frontwards and backwards. *You* go look
for it.
--
Joe User
GreyCloud
It sounds like false advertising to me.
Adam Warner
Scroll down to "Preemptive Multitasking Architecture".
There in black and white Microsoft is claiming that Windows 9x and ME are
not preemptive multitasking operating systems. A white circle "indicates a
feature that is not available".
Regards,
Adam
Donn Miller
Adam Warner wrote:
> There in black and white Microsoft is claiming that Windows 9x and ME are
> not preemptive multitasking operating systems. A white circle "indicates a
> feature that is not available".
There's no way Win 9x could be pre-emptive -- they're way too clunky
with the way they multi-task to be truly pre-emptive.
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 80,000 Newsgroups - 16 Different Servers! =-----
GreyCloud
>
> Adam Warner wrote:
>
> > There in black and white Microsoft is claiming that Windows 9x and ME are
> > not preemptive multitasking operating systems. A white circle "indicates a
> > feature that is not available".
>
> There's no way Win 9x could be pre-emptive -- they're way too clunky
> with the way they multi-task to be truly pre-emptive.
That seems to be what Charles Petzold is saying as well.
"Yet the multitasking implemented in the early versions of windows was
not the traditional preemptive time-slicing found in multiuser operating
systems."
P. 1199 of the 5th edition of windows programming.
However, the ads of win95 did say that it "now has preemptive
multitasking".
And we know that win3.1 was called "cooperative multitasking".
Either the XP ad is a lie, or that win95 ad was a lie.
Donn Miller
> That seems to be what Charles Petzold is saying as well.
>
> "Yet the multitasking implemented in the early versions of windows was
> not the traditional preemptive time-slicing found in multiuser operating
> systems."
> P. 1199 of the 5th edition of windows programming.
Erik seems to think that SOME pre-emptiveness == complete and total
pre-emptiveness (as long as it's an MS product). Of course, if Linux
had the same shitty time-slicing as Win 98, he would've been saying that
Linux absolutely, positively is not pre-emptive. In fact, it would not
be a multi-tasking OS at all, but a co-operative process OS.
Erik Funkenbusch
news:9s57vv$10tjgq$1...@ID-105510.news.dfncis.de...
> Here's some proof for you Joe:
>
>
http://www.microsoft.com/technet/treeview/default.asp?url=/TechNet/prodtechn
ol/winxppro/evaluate/xphmvpro.asp
>
> Scroll down to "Preemptive Multitasking Architecture".
Interesting. However, the chart is not wrong. The key here is the word
"Architecture". As people here are so fond of mentioning, Windows 95 is
based on a DOS architecture, which is not preemptively multitasking. Yes,
Windows 95 does pre-emptively multitask, but the architecture was not
designed to do so.
I'm not sure if this is intentionally deceptive, or if they were really
trying to "simplify" the fact that the architectures of XP and Win95 are
radically different.
> There in black and white Microsoft is claiming that Windows 9x and ME are
> not preemptive multitasking operating systems. A white circle "indicates a
> feature that is not available".
Actually, Joe User stated that the ad said "Preemptive multitasking", which
would be incorrect. But that's not what the chart says.
As it turns out, Joe User *DID* misinterpret the chart, as I claimed. Yet
he said it was a simple chart, and presumably couldn't have misinterpreted
it.
Erik Funkenbusch
news:3BE66677...@cvzoom.net...
> GreyCloud wrote:
> > That seems to be what Charles Petzold is saying as well.
> >
> > "Yet the multitasking implemented in the early versions of windows was
> > not the traditional preemptive time-slicing found in multiuser operating
> > systems."
> > P. 1199 of the 5th edition of windows programming.
>
> Erik seems to think that SOME pre-emptiveness == complete and total
> pre-emptiveness (as long as it's an MS product). Of course, if Linux
> had the same shitty time-slicing as Win 98, he would've been saying that
> Linux absolutely, positively is not pre-emptive. In fact, it would not
> be a multi-tasking OS at all, but a co-operative process OS.
Windows 95 *ALWAYS* pre-empts applications. Where Win95 goes awry is that
because of the 16 bit mutex, it may not always be able to reschedule any
code that wants those same resources until the mutex is released. This
leads to somewhat clunky multitasking in some cases.
There is a difference between not pre-empting a process and not rescheduling
it. Preemptive refers only to the ability to preempt a thread.
Charlie Ebert
You are *ALWAYS* full of shit and don't know what your talking about.
Windows 95 is a dos based system emulation.
It's not even an operating system! Fool!
--
Charlie
The Ghost In The Machine
<er...@visi.com>
wrote
on Mon, 5 Nov 2001 06:32:41 -0600
<JFvF7.25$H7.3...@ruti.visi.com>:
No, sorry. Preemptive multitasking does in fact refer to processes
as well.
In older systems, such as pre-7 MacOS, for example, a process
had to explicitly or implicitly Yield(), allowing another process,
or the system, to take control of the micro. One of the Amiga's
innovations was that it implemented true preemptive multitasking.
(To be sure, it wasn't the first, although it may have been one
of the first user-level microcomputers.) However, the Amiga's
scheduler did not have dynamic reprioritization, which lead
to some unexpected behavior at times.
At the time, DOS didn't even understand the concept of processes,
although some interesting hacks were available to switch
between "windows", running more than one DOS app at a time.
Desqview/X is probably the best known.
On a system with true preemptive multitasking (such as Win95 -- yes,
they did have it AFAIK, at least for Win32 stuff), a process can
be interrupted at any time, and another process switched to. The
switching involves a TSS load, if I remember x86 correctly, and
the saving and reloading of the registers. (Yield() doesn't
require saving registers, as the system's in a known state,
or should be.)
I can't speak for the mutex, admittedly, and that may have gummed
up the works. I don't know if there was a separate mutex for
Win16 or what; the Win16 part, for compatibility reasons,
continued to be cooperative.
And Win95's memory management was problematical. It may have been
because of its overlay with DOS, which had some very weird ideas
on how to communicate with extended or expanded memory. On
at least one system, Windows operated apparently perfectly
while Linux was unhappy -- it turned out that the reason was
bad memory, and Windows eventually started to malfunction on that
system, too, as I recall. (It's been awhile.)
As for rescheduling -- the traditional management method, IIRC,
was to maintain a ready queue and a series of wait queues;
each wait queue was associated with some sort of resource.
Once a process relinquished a resource, it moved another process
from the wait queue to the ready queue (by using a kernel
routine), and went about its business. (This is admittedly deep
in the kernel somewhere. I'd have to look.)
If a process is preempted, the kernel, after saving everything,
selects a new process from the ready queue (if any), puts the
old process on the ready queue, and transfers control to
the process.
It is possible for no processes to be ready to run, except for a
special "null" process, which does nothing but halt. If the
null process is restarted, it halts again. This seemingly
silly behavior ensures that there's at least one process to
select, and halting a micro doesn't affect system behavior too
much as the quantum tick will interrupt it, as well as I/O. It
also reduces power consumption.
I don't know the specifics about Win95 reschedulng, beyond the
observation that it didn't halt and the x86 ran hot as a result.
One would think Win98 fixed that.
--
ew...@aimnet.com -- insert random misquote here
EAC code #191 116d:11h:24m actually running Linux.
We are all naked underneath our clothes.
Erik Funkenbusch
message news:slrn9udehg...@lexideb.athghost7038suus.net...
You misunderstand. Obviously it refers to processes because each process
must contain at least one thread.
Erik Funkenbusch
controller issues
"Adam Warner" <use...@consulting.net.nz> wrote in message
news:9s57vv$10tjgq$1...@ID-105510.news.dfncis.de...
> Here's some proof for you Joe:
>
>
http://www.microsoft.com/technet/treeview/default.asp?url=/TechNet/prodtechn
ol/winxppro/evaluate/xphmvpro.asp
>
> Scroll down to "Preemptive Multitasking Architecture".
Interesting. However, the chart is not wrong. The key here is the word
"Architecture". As people here are so fond of mentioning, Windows 95 is
based on a DOS architecture, which is not preemptively multitasking. Yes,
Windows 95 does pre-emptively multitask, but the architecture was not
designed to do so.
I'm not sure if this is intentionally deceptive, or if they were really
trying to "simplify" the fact that the architectures of XP and Win95 are
radically different.
> There in black and white Microsoft is claiming that Windows 9x and ME are
> not preemptive multitasking operating systems. A white circle "indicates a
> feature that is not available".
Actually, Joe User stated that the ad said "Preemptive multitasking", which
Donn Miller
Erik Funkenbusch wrote:
>
> Appologies if you see this twice, my news server has been having RAID
> controller issues
Must not be running a pre-emptive OS.
Donn Miller
drsquare wrote:
> In other words, it's a completely worthless piece of shit, yet many times
> more expensive than Linux, which is great.
Actually, it's *undefined* times more expensive than Linux.
Matthew Gardiner
"Adam Warner" <use...@consulting.net.nz> wrote in message
news:9s57vv$10tjgq$1...@ID-105510.news.dfncis.de...
I wonder if the New Zealand Commerce Commission would be interested in a bit
of detective work, to find out whether Microsoft have broken any laws in
regards to the Fair Trading Act and respective consumer protection acts.
Matthew Gardiner
Adam Warner
[cross-posted to nz.comp. Follow-up set to comp.os.linux.advocacy]
Well I know I paid for a preemptive multitasking operating system with
Windows 95 Matthew. It was a major selling point. Erik's response was
predicable.
Most importantly native 32-bit Windows applications DID let you work on
other applications while they were busy processing. It was a major
improvement over Windows 3.1.
By the way I've got my original retail box of Windows 95 upgrade sitting
in front of me right now.
I've scanned the front and back covers for you. I'm sure you'll all find
it interesting:
http://www.openphd.net/Windows_95_front_cover.jpg
http://www.openphd.net/Windows_95_back_cover_150DPI.jpg
Note the "preemptive multitasking" is listed in the "Performance Features"
in Windows 95. That Windows 95 is "32-bit, preemptive mulititasking" is
also described as "Programs can work simultaneously, so you can do more."
Regards,
Adam
GreyCloud
In the general sense of the term of multi-tasking and being premptive,
like VMS was, doesn't need a thread definition to qualify. Especially
since VMS 4.x never had threads but still had the preemptive
capabilities.
Joe User
Adam Warner <use...@consulting.net.nz> wrote in message
news:9s57vv$10tjgq$1...@ID-105510.news.dfncis.de...
A fine, and undoubtedly unpleasant, piece of detective work, Adam. Thanks.
That's pretty much the chart Microsoft sent to retailers, and Office Max
printed part of it in a Sunday newspaper insert. Erik implied that I was
either lying about the ad ("IF the ad exists...") or mistaken. He also
thought Office Max might have made an error, and he probably brought up the
possibility that the newspaper made a mistake. Not for a second did he
believe that Microsoft would claim such a thing, since he spent years on
Usenet arguing the opposite (and it's all documented at Google :).
I predict that Erik will now attempt to land on his feet and proclaim that a
"pre-emptive multitasking architecture" is not the same thing at all. He'll
say something like, "Yes, Windows 9x/ME used pre-emptive multitasking, but
they didn't have a pre-emptive multitasking ARCHITECTURE. If you're too
ignorant to understand that, I don't have time to give you a lesson in
operating system theory."
Oh wait, that was John Saunders' line.
Anyway, thanks, Adam.
--
Joe User
mmn...@mmnnoo.org
There's no reason to argue about or even question whether
win95 has pre-emptive multitasking. Just run this
simple infinite loop:
int main(int argc, char * argv[]) { while(1); return 0; };
This program never yields the processor, yet other programs can
run concurrently. So Win95 has preemptive multitasking, absolutely.
Which makes one wonder why this is being advertised as a new
feature of Win XP.
Erik Funkenbusch
news:l0FF7.2442$GD.6...@news1.rdc1.tn.home.com...
>
> Adam Warner <use...@consulting.net.nz> wrote in message
> news:9s57vv$10tjgq$1...@ID-105510.news.dfncis.de...
> > Here's some proof for you Joe:
> >
> >
>
http://www.microsoft.com/technet/treeview/default.asp?url=/TechNet/prodtechn
> ol/winxppro/evaluate/xphmvpro.asp
> >
> > Scroll down to "Preemptive Multitasking Architecture".
> >
> > There in black and white Microsoft is claiming that Windows 9x and ME
are
> > not preemptive multitasking operating systems. A white circle "indicates
a
> > feature that is not available".
> >
> > Regards,
> > Adam
>
> A fine, and undoubtedly unpleasant, piece of detective work, Adam.
Thanks.
So, you are admitting now that you mischaracterized what the chart said.
> That's pretty much the chart Microsoft sent to retailers, and Office Max
> printed part of it in a Sunday newspaper insert. Erik implied that I was
> either lying about the ad ("IF the ad exists...") or mistaken.
You are, one or the other. You said the ad claimed that Windows 95 didn't
have pre-emptive multitasking. The ad clearly says Pre-emptive multitasking
*ARCHITECTURE* (if you're up on your grammar rules, you realize that
Architecture is the noun here, and "pre-emptive multitasking" is an
adjective. You originally used Pre-emptive multitasking as a noun.)
> He also
> thought Office Max might have made an error, and he probably brought up
the
> possibility that the newspaper made a mistake. Not for a second did he
> believe that Microsoft would claim such a thing, since he spent years on
> Usenet arguing the opposite (and it's all documented at Google :).
Microsoft didn't claim such a thing.
> I predict that Erik will now attempt to land on his feet and proclaim that
a
> "pre-emptive multitasking architecture" is not the same thing at all.
He'll
> say something like, "Yes, Windows 9x/ME used pre-emptive multitasking, but
> they didn't have a pre-emptive multitasking ARCHITECTURE. If you're too
> ignorant to understand that, I don't have time to give you a lesson in
> operating system theory."
Wow, you're so psychic, nevermind the fact that I wrote just that many hours
before you "predicted" it. I wonder how you might have guessed? Could you
have read my response and then pretended not to have? Nah...
The point on the chart refers to the fact that XP is based on NT, an
achitecture designed from scratch to be pre-emptively multitasking. 9x is
based on a DOS architecture, which was not. The chart is not wrong.
You, however, were misleading when you made the original statement and did
in fact misunderstand it.
Erik Funkenbusch
news:PWJF7.55483$IR4.30...@news1.denver1.co.home.com...
It's not. It's just confusing. The feature is talking about the difference
between the NT and DOS based architectures (NT being designed to be
multitasking, DOS not designed that way).
blah
>> That's pretty much the chart Microsoft sent to retailers, and Office Max
>> printed part of it in a Sunday newspaper insert. Erik implied that I was
>> either lying about the ad ("IF the ad exists...") or mistaken.
> You are, one or the other. You said the ad claimed that Windows 95 didn't
> have pre-emptive multitasking. The ad clearly says Pre-emptive multitasking
> *ARCHITECTURE* (if you're up on your grammar rules, you realize that
> Architecture is the noun here, and "pre-emptive multitasking" is an
> adjective. You originally used Pre-emptive multitasking as a noun.)
I have the Office Max ad sitting on my desk at this moment, and the ad clearly
says "Pre-emptive multitasking", not "Pre-emptive multitasking architecture".
Donn Miller
> You are, one or the other. You said the ad claimed that Windows 95 didn't
> have pre-emptive multitasking. The ad clearly says Pre-emptive multitasking
> *ARCHITECTURE* (if you're up on your grammar rules, you realize that
> Architecture is the noun here, and "pre-emptive multitasking" is an
> adjective. You originally used Pre-emptive multitasking as a noun.)
> The point on the chart refers to the fact that XP is based on NT, an
> achitecture designed from scratch to be pre-emptively multitasking. 9x is
> based on a DOS architecture, which was not. The chart is not wrong.
That's just it. Windoze 98 is crap for exactly that reason: it's a
non-pre-emptive OS retrofitted with a hack that allows pre-emptive
multi-tasking, bolted onto a 16-bit kernel.
Erik Funkenbusch
news:483832766...@blah-example.net...
Joe User says differently, that the ad was the same chart.
So, which is correct now?
blah
No, Joe User said, "That's pretty much the chart Microsoft sent to retailers,
and Office Max printed part of it in a Sunday newspaper insert."
I posted this in the other thread a couple of days ago and directed your
attention to it, but here it is again. Save it as "table.tex" and
run `latex table; xdvi table` to view it. It is a verbatim copy
of the table in the Office Max ad.
\documentclass{article}
\begin{document}
\begin{tabular}{|l|c|c|c|c|c|}
\hline
\multicolumn{6}{|c|}{\rule[-3mm]{0mm}{8mm}
\bfseries \large Benefits of upgrading to Windows XP } \\
\hline
&Win&Win&{\bfseries NEW}&Win&{\bfseries NEW}\\
{\bfseries Features of \ldots } &
95/98 &
ME &
\bfseries{XP Home} &
2000 Pro &
\bfseries{XP Pro} \\
\hline
Dependable Windows 2000 engine &&&$\bullet$&$\bullet$&$\bullet$\\
\hline
Encrypting File System / multi-user &&&&$\bullet$&$\bullet$\\
\hline
Fast user switching &&&$\bullet$&&$\bullet$\\
\hline
File management &&&$\bullet$&&$\bullet$\\
\hline
Internet connection firewall &&&$\bullet$&&$\bullet$\\
\hline
Microsoft Windows Media Player &&&$\bullet$&&$\bullet$\\
\hline
Multilingual support &&&&$\bullet$&$\bullet$\\
\hline
Pre-emptive multitasking &&&$\bullet$&$\bullet$&$\bullet$\\
\hline
Remote assistance &&&$\bullet$&&$\bullet$\\
\hline
Remote Desktop &&&&&$\bullet$\\
\hline
System restore &&$\bullet$&$\bullet$&&$\bullet$\\
\hline
Windows Messenger &&&$\bullet$&&$\bullet$\\
\hline
Wireless networking support &&&$\bullet$&$\bullet$&$\bullet$\\
\hline
\end{tabular}
\end{document}
Charlie Ebert
HA HA HA HA!
Every day is a new adventure!
I swear this guy just cracks me up!
--
Charlie
Ian Pegel
What's all this about Win2k Pro being multi-user. Oh of course, that is
one at a time. A bit like my new multi-user pencil! MS certainly has a
way with words. "Sophistry" is one which springs to my mind.
--
"Measure not the days of the saints in onions"
Pegelism No. 82
Chris Ahlstrom
> Windows 95 *ALWAYS* pre-empts applications. Where Win95 goes awry is that
> because of the 16 bit mutex, it may not always be able to reschedule any
> code that wants those same resources until the mutex is released. This
> leads to somewhat clunky multitasking in some cases.
>
> There is a difference between not pre-empting a process and not rescheduling
> it. Preemptive refers only to the ability to preempt a thread.
Next you'll be saying that Win 95/98 have a very short peak interrupt
latency... only a couple hundred milliseconds.
Haw haw
Chris
--
Hack a Linux kernel today!
Sundial Services
rendering in Microsoft Windows is done by the individual threads
themselves, not by a separate window manager process. This is why we
have klunky software like PCAnywhere and Citrix in that world.
And _that is what puts nearly every thread in Windows at odds with every
other thread that the user ever _sees.
Chris Ahlstrom
> What Erik is talking about is the fact that all of the graphics
> rendering in Microsoft Windows is done by the individual threads
> themselves, not by a separate window manager process. This is why we
> have klunky software like PCAnywhere and Citrix in that world.
>
> And _that is what puts nearly every thread in Windows at odds with every
> other thread that the user ever _sees.
Ah, I see! Now I see why Win 2000 graphics are so clunky.
T. Max Devlin
>Donn Miller wrote:
>>
>> Adam Warner wrote:
>>
>> > There in black and white Microsoft is claiming that Windows 9x and ME are
>> > not preemptive multitasking operating systems. A white circle "indicates a
>> > feature that is not available".
>>
>> There's no way Win 9x could be pre-emptive -- they're way too clunky
>> with the way they multi-task to be truly pre-emptive.
>
>
>"Yet the multitasking implemented in the early versions of windows was
>not the traditional preemptive time-slicing found in multiuser operating
>systems."
>P. 1199 of the 5th edition of windows programming.
>
>multitasking".
>
>And we know that win3.1 was called "cooperative multitasking".
AFAIK, Win3.x was called 'task switching'. "Cooperative multitasking"
is the old MacOS.
>Either the XP ad is a lie, or that win95 ad was a lie.
Everything MS says is a lie. I don't see why seeing proof of any one of
them would be notable.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
GreyCloud
>
> Erik Funkenbusch wrote:
>
> > You are, one or the other. You said the ad claimed that Windows 95 didn't
> > have pre-emptive multitasking. The ad clearly says Pre-emptive multitasking
> > *ARCHITECTURE* (if you're up on your grammar rules, you realize that
> > Architecture is the noun here, and "pre-emptive multitasking" is an
> > adjective. You originally used Pre-emptive multitasking as a noun.)
>
> > The point on the chart refers to the fact that XP is based on NT, an
> > achitecture designed from scratch to be pre-emptively multitasking. 9x is
> > based on a DOS architecture, which was not. The chart is not wrong.
>
> That's just it. Windoze 98 is crap for exactly that reason: it's a
> non-pre-emptive OS retrofitted with a hack that allows pre-emptive
> multi-tasking, bolted onto a 16-bit kernel.
And pretty much what Charles Petzold said in his book "Programming
Windows Fifth Edition".
Erik Funkenbusch
> What's all this about Win2k Pro being multi-user. Oh of course, that is
> one at a time. A bit like my new multi-user pencil! MS certainly has a
> way with words. "Sophistry" is one which springs to my mind.
Win2k server includes Terminal Services, which allow multiple users to run
GUI programs at the same time. Win2k also includes a telnet server.
Peter Köhlmann
> Strolling through comp.os.linux.advocacy, I heard GreyCloud say to
>>Donn Miller wrote:
>>>
>>> Adam Warner wrote:
>>>
>>> > There in black and white Microsoft is claiming that Windows 9x and
>>> > ME are not preemptive multitasking operating systems. A white circle
>>> > "indicates a feature that is not available".
>>>
>>> There's no way Win 9x could be pre-emptive -- they're way too clunky
>>> with the way they multi-task to be truly pre-emptive.
>>
>>That seems to be what Charles Petzold is saying as well.
>>
>>"Yet the multitasking implemented in the early versions of windows was
>>not the traditional preemptive time-slicing found in multiuser operating
>>systems."
>>P. 1199 of the 5th edition of windows programming.
>>
>>However, the ads of win95 did say that it "now has preemptive
>>multitasking".
>>
>>And we know that win3.1 was called "cooperative multitasking".
>
> AFAIK, Win3.x was called 'task switching'. "Cooperative multitasking"
> is the old MacOS.
>
No, Win 3.x did *not* task-switch, it did cooperative multitasking.
That means the tasks running give up the processor volontarily or can
be switched away during the next system call, but *not* depending on
some timer running out. Thios means of course that any task running for
any lenght of time without calls into the system won't give up the
processor, until they do it actively on their own.
Peter
--
It's not about, 'Where do you want to go today?'
It's more like 'Where am I allowed to go today?'
and 'Where do I want my data to go today?'
T. Max Devlin
>T. Max Devlin wrote:
>> Strolling through comp.os.linux.advocacy, I heard GreyCloud say to
>>>Donn Miller wrote:
>>>>
>>>> Adam Warner wrote:
>>>>
>>>> > There in black and white Microsoft is claiming that Windows 9x and
>>>> > ME are not preemptive multitasking operating systems. A white circle
>>>> > "indicates a feature that is not available".
>>>>
>>>> There's no way Win 9x could be pre-emptive -- they're way too clunky
>>>> with the way they multi-task to be truly pre-emptive.
>>>
>>>That seems to be what Charles Petzold is saying as well.
>>>
>>>"Yet the multitasking implemented in the early versions of windows was
>>>not the traditional preemptive time-slicing found in multiuser operating
>>>systems."
>>>P. 1199 of the 5th edition of windows programming.
>>>
>>>However, the ads of win95 did say that it "now has preemptive
>>>multitasking".
>>>
>>>And we know that win3.1 was called "cooperative multitasking".
>>
>> AFAIK, Win3.x was called 'task switching'. "Cooperative multitasking"
>> is the old MacOS.
>
>No, Win 3.x did *not* task-switch, it did cooperative multitasking.
I'm not going to argue with you as to what it did or what these things
refer to. I am simply clarifying what they called it, and they most
definitely did not call it 'cooperative multitasking'.
I believe they simply called it 'multi-tasking' by Microsoft, which the
knowledgable segment of the industry recognized as meaning 'task
switching'. Win95 supposedly added 'pre-emptive' multi-tasking, but now
that is in doubt, as well.
Peter Köhlmann
> Strolling through comp.os.linux.advocacy, I heard Peter Köhlmann say to
>>
>>No, Win 3.x did *not* task-switch, it did cooperative multitasking.
>
> I'm not going to argue with you as to what it did or what these things
> refer to. I am simply clarifying what they called it, and they most
> definitely did not call it 'cooperative multitasking'.
>
> I believe they simply called it 'multi-tasking' by Microsoft, which the
> knowledgable segment of the industry recognized as meaning 'task
> switching'. Win95 supposedly added 'pre-emptive' multi-tasking, but now
> that is in doubt, as well.
>
Well, sure it did "multi-tasking" in a way. That is, it did allow to
execute several programs at the same time, and (mostly) it did seem they
were running side by side. But normally one would say "multi-tasking" and
mean "preemptive multi-tasking". So this was one of Microsofts word-games.
They weren't wrong (by the letter), but by the meaning.
In reality, it was just "cooperative multitasking", which meant just
that, the programs had to cooperate, otherwise....
Peter
--
This problem was sponsored by Microsoft
Joe User
Erik Funkenbusch <er...@visi.com> wrote in message
news:iiKF7.129$H7.1...@ruti.visi.com...
> "Joe User" <craft...@yahoo.com> wrote in message
> news:l0FF7.2442$GD.6...@news1.rdc1.tn.home.com...
> >
> > Adam Warner <use...@consulting.net.nz> wrote in message
> > news:9s57vv$10tjgq$1...@ID-105510.news.dfncis.de...
> > > Here's some proof for you Joe:
> > >
> > >
> >
>
http://www.microsoft.com/technet/treeview/default.asp?url=/TechNet/prodtechn
> > ol/winxppro/evaluate/xphmvpro.asp
> > >
> > > Scroll down to "Preemptive Multitasking Architecture".
> > >
> > > There in black and white Microsoft is claiming that Windows 9x and ME
> are
> > > not preemptive multitasking operating systems. A white circle
"indicates
> a
> > > feature that is not available".
> > >
> > > Regards,
> > > Adam
> >
> > A fine, and undoubtedly unpleasant, piece of detective work, Adam.
> Thanks.
>
> So, you are admitting now that you mischaracterized what the chart said.
No, Erik. The Office Max ad said exactly what I said it did.
> > That's pretty much the chart Microsoft sent to retailers, and Office Max
> > printed part of it in a Sunday newspaper insert. Erik implied that I
was
> > either lying about the ad ("IF the ad exists...") or mistaken.
>
> You are, one or the other.
A third possibility didn't occur to you, Erik. That would be the one
involving honesty and accuracy. No surprise there, though.
Again, the Office Max ad said exactly what I said it did. I still have it,
in fact. It says "pre-emptive multitasking," not "pre-emptive multitasking
architecture." The chart was even on the Office Max site earlier, exactly
as it appeared in the paper, but I can't seem to find it there, now.
Anyway, as you can see (unless you still think I'm lying), I didn't
misinterpret or "mischaracterize" anything.
> > I predict that Erik will now attempt to land on his feet and proclaim
that
> a
> > "pre-emptive multitasking architecture" is not the same thing at all.
> He'll
> > say something like, "Yes, Windows 9x/ME used pre-emptive multitasking,
but
> > they didn't have a pre-emptive multitasking ARCHITECTURE. If you're too
> > ignorant to understand that, I don't have time to give you a lesson in
> > operating system theory."
>
> Wow, you're so psychic, nevermind the fact that I wrote just that many
hours
> before you "predicted" it.
Usenet is not a perfectly linear, chronological system. Sometimes responses
appear on a server before the original article does.
> I wonder how you might have guessed? Could you
> have read my response and then pretended not to have? Nah...
No, Erik, your response was not on my system when I wrote my "prediction."
Wanna know what led to it?
While I was replying to Adam, I had his Microsoft link up in my browser. I
noticed that the MS chart used the phrase, "pre-emptive multitasking
architecture" rather than simply "pre-emptive multitasking." I was
imagining your reaction on seeing the chart right there on Microsoft's own
web site after you had implied I was lying about its very existence, and I
knew you'd feel a slight panic at being proven wrong. I knew you'd jump at
ANYthing to try to salvage something, and the "architecture" word seemed
like the immediate choice. I figured that, in your panic, you wouldn't
notice a much *better* argument, right there on the same page -- you'd be
in too much of a hurry to cling to the first bone you could find.
So, my "prediction" was really just an attempt to make it more difficult for
you to use the architecture argument. I assumed you'd see my prediction
before you responded to any of it.. I didn't know that you already HAD
responded and it just hadn't reached me. It never occurred to me that the
timing would be that perfect.
That better argument I mentioned...study that page carefully. You can find
it. It gives MS an out if anybody ever tries to make an issue of it.
It doesn't give *you* an out, though. You've already married yourself to
the architecture silliness.
> The point on the chart refers to the fact that XP is based on NT, an
> achitecture designed from scratch to be pre-emptively multitasking. 9x is
> based on a DOS architecture, which was not.
Just curious, Erik...have you ever argued on Usenet that Windows 95 was not
based on the DOS architecture, that it was instead an entirely new operating
system that happened to support legacy DOS apps as a convenience for
consumers?
Think carefully.
--
Joe User
Joe User
Actually, it IS being advertised as a new feature in 2k/XP, at least at
Office Max. There was a copy of this chart on their web site, in jpeg form,
and it was *exactly* like the one that appeared in the Sunday insert, colors
and all. This leads to the question of where they got the jpeg, since the
wording in that is slightly different from the wording on Microsoft's site.
Did they create the jpeg, basing it on material that MS sent them? Or did
MS send this little jpeg as a convenience to large retailers with web sites?
Interesting question, isn't it? And if this jpeg, the one that leaves out
the word "architecture," turns up on some other retailer's web site, we
could probably conclude that Microsoft created it, right?
--
Joe User
Erik Funkenbusch
> > The point on the chart refers to the fact that XP is based on NT, an
> > achitecture designed from scratch to be pre-emptively multitasking. 9x
is
> > based on a DOS architecture, which was not.
>
> Just curious, Erik...have you ever argued on Usenet that Windows 95 was
not
> based on the DOS architecture, that it was instead an entirely new
operating
> system that happened to support legacy DOS apps as a convenience for
> consumers?
>
> Think carefully.
No.
I have argued that Win95 is not a DOS application, and that it's not a DOS
"shell". I have not argued that it's "an entirely new operating system that
happened to support legacy DOS apps as a convenience for consumers".
Now, be careful about what YOU claim I said, because there are many long
standing arguments that I can prove via google.
Erik Funkenbusch
Again, you have yet to provide even a single other source using this chart.
So why do you keep saying "if..."?
T. Max Devlin
>T. Max Devlin wrote:
>
>> Strolling through comp.os.linux.advocacy, I heard Peter Köhlmann say to
>>>
>>>No, Win 3.x did *not* task-switch, it did cooperative multitasking.
>>
>> I'm not going to argue with you as to what it did or what these things
>> refer to. I am simply clarifying what they called it, and they most
>> definitely did not call it 'cooperative multitasking'.
>>
>> I believe they simply called it 'multi-tasking' by Microsoft, which the
>> knowledgable segment of the industry recognized as meaning 'task
>> switching'. Win95 supposedly added 'pre-emptive' multi-tasking, but now
>> that is in doubt, as well.
>
The issue is what it was called, not what it did.
T. Max Devlin
>"Joe User" <craft...@yahoo.com> wrote in message
>news:Ew_F7.6783$GD.18...@news1.rdc1.tn.home.com...
>> Interesting question, isn't it? And if this jpeg, the one that leaves out
>> the word "architecture," turns up on some other retailer's web site, we
>> could probably conclude that Microsoft created it, right?
>
>Again, you have yet to provide even a single other source using this chart.
>So why do you keep saying "if..."?
Because others have pointed out it does in fact come directly from
Microsoft. Why are you trying to pretend otherwise?
Peter Köhlmann
No, it did not "happen to support legacy DOS apps as a convenience for
consumers", it had to, to be sold at all. Otherwsie even those who would
have bought nearly anything with MS on it would have told them to stuff it
The Ghost In The Machine
<er...@visi.com>
wrote
on Tue, 6 Nov 2001 14:34:48 -0600
<FPXF7.255$H7.2...@ruti.visi.com>:
A capability that has existed on X -- as far as client programs
are concerned, anyway -- since its inception in the mid 80's,
if not earlier.
Telnet's even older.
Were we supposed to be impressed? :-)
--
ew...@aimnet.com -- insert random misquote here
EAC code #191 117d:02h:58m actually running Linux.
It's a conspiracy of one.
blah
> In comp.os.linux.advocacy, Erik Funkenbusch
> <er...@visi.com>
> wrote
> on Tue, 6 Nov 2001 14:34:48 -0600
> <FPXF7.255$H7.2...@ruti.visi.com>:
>>"Ian Pegel" <IanP...@netscape.net> wrote in message
>>> What's all this about Win2k Pro being multi-user. Oh of course, that is
>>> one at a time. A bit like my new multi-user pencil! MS certainly has a
>>> way with words. "Sophistry" is one which springs to my mind.
>>
>>Win2k server includes Terminal Services, which allow multiple users to run
>>GUI programs at the same time. Win2k also includes a telnet server.
> A capability that has existed on X -- as far as client programs
> are concerned, anyway -- since its inception in the mid 80's,
> if not earlier.
> Telnet's even older.
> Were we supposed to be impressed? :-)
I couldn't even get a Win2K telnet daemon to talk to me. The protocol had
been "embraced and extended" to the point of uselessness.
The Ghost In The Machine
<bl...@blah-example.net>
wrote
on Wed, 07 Nov 2001 18:11:22 GMT
<47382765...@blah-example.net>:
In that particular case, that might be a good thing. :-) Less chance
of the Win2k server being infected or rooted. :-)
Not that you'd be able to do an awful lot over a serial line anyway.
Not on Windows. (There are a few commands that don't require
a window, to be sure -- ipconfig comes to mind. And I don't know
how capable Terminal Services really is.)
But one very logical and relatively cheap system would have a bunch
of second-hand PCs being clustered around a higher-powered Sparc
single or multiCPU server running Linux or Solaris. The server would
do all the number crunching and the PCs would be basically X terminals,
displaying the results. An alternative would have the PCs fetch
the data from the number crunching and interpret the results
locally -- one could use gnuplot in a pinch. It's not as
pretty, perhaps -- but it's bog-standard and readily available.
Such a system would be highly scalable, as well -- replace the
Solaris machine with an S/390 and add more PCs, for example.
I really fail to see why Microsoft would be all that good here.
As a solution starting out, it's nearly perfect, of course -- my
understanding is that the costs of a MS server are very cheap
compared to an entry-level UNIX(R) server -- but it doesn't
scale well.
--
ew...@aimnet.com -- insert random misquote here
EAC code #191 118d:16h:37m actually running Linux.
This is a pithy statement. Please watch where you pith.
Erik Funkenbusch
message news:slrn9uisf1...@lexideb.athghost7038suus.net...
> In comp.os.linux.advocacy, Erik Funkenbusch
> <er...@visi.com>
> wrote
> on Tue, 6 Nov 2001 14:34:48 -0600
> <FPXF7.255$H7.2...@ruti.visi.com>:
> >"Ian Pegel" <IanP...@netscape.net> wrote in message
> >> What's all this about Win2k Pro being multi-user. Oh of course, that is
> >> one at a time. A bit like my new multi-user pencil! MS certainly has a
> >> way with words. "Sophistry" is one which springs to my mind.
> >
> >Win2k server includes Terminal Services, which allow multiple users to
run
> >GUI programs at the same time. Win2k also includes a telnet server.
>
> A capability that has existed on X -- as far as client programs
> are concerned, anyway -- since its inception in the mid 80's,
> if not earlier.
>
> Telnet's even older.
>
> Were we supposed to be impressed? :-)
No. I didn't ask you to be impressed, I just asked the original poster to
be correct.
charles lyttle
arguing that Windows 98 and NT were preemptive and had no DOS only to
have his patron pull the rug out from under him. He has to do something
to save face. Let's let him have his new definition of pre-emptive and
move on.
--
Russ Lyttle
<http://home.earthlink.net/~lyttlec>
lyt...@earthlink.net
Mark Kent
>
>Such a system would be highly scalable, as well -- replace the
>Solaris machine with an S/390 and add more PCs, for example.
>
>I really fail to see why Microsoft would be all that good here.
>As a solution starting out, it's nearly perfect, of course -- my
>understanding is that the costs of a MS server are very cheap
>compared to an entry-level UNIX(R) server -- but it doesn't
>scale well.
>
MS server would not be cheap compared to a linux one, though.
--
Mark Kent
Take out the ham to mail me.
The Ghost In The Machine
<ma...@NOHAM.otford.kent.btinternet.co.uk>
wrote
on Thu, 8 Nov 2001 06:39:19 +0000
<na9ds9...@192.168.1.1>:
I did say UNIX(R). Not Unix. One minor problem with Linux is
the credibility gap. Microsoft is very credible; Linux is not.
Why this is so, I can't precisely say, but "money talks", perhaps.
Microsoft has lots of that. :-)
We'll see. If nothing else, Microsoft's XP offering should be
the stablest Windows consumer OS on the planet.
(Too bad Linux could be the stablest *non*-Windows consumer OS on
the planet.)
[.sigsnip]
--
ew...@aimnet.com -- insert random misquote here
EAC code #191 119d:22h:20m actually running Linux.
This is the best part of the message.
The Ghost In The Machine
<56lG7.969$H7.4...@ruti.visi.com>:
It does appear that Win2k is multiuser, much like Win95 was multitasking.
Both are technically correct -- and both are good sales points.
And both are upstaged by Linux.
--
ew...@aimnet.com -- insert random multiuser OS here
EAC code #191 119d:23h:37m actually running Linux.
Use the source, Luke.
T. Max Devlin
Machine say:
>In comp.os.linux.advocacy, Mark Kent
><ma...@NOHAM.otford.kent.btinternet.co.uk>
> wrote
>on Thu, 8 Nov 2001 06:39:19 +0000
><na9ds9...@192.168.1.1>:
>>The Ghost In The Machine wrote:
>>>
>>>Such a system would be highly scalable, as well -- replace the
>>>Solaris machine with an S/390 and add more PCs, for example.
>>>
>>>I really fail to see why Microsoft would be all that good here.
>>>As a solution starting out, it's nearly perfect, of course -- my
>>>understanding is that the costs of a MS server are very cheap
>>>compared to an entry-level UNIX(R) server -- but it doesn't
>>>scale well.
>>>
>>
>>MS server would not be cheap compared to a linux one, though.
>
>I did say UNIX(R). Not Unix. One minor problem with Linux is
>the credibility gap. Microsoft is very credible; Linux is not.
Guffaw.
Mark Kent
>In comp.os.linux.advocacy, Mark Kent
><ma...@NOHAM.otford.kent.btinternet.co.uk>
> wrote
>on Thu, 8 Nov 2001 06:39:19 +0000
><na9ds9...@192.168.1.1>:
>>The Ghost In The Machine wrote:
>>>
>>>Such a system would be highly scalable, as well -- replace the
>>>Solaris machine with an S/390 and add more PCs, for example.
>>>
>>>I really fail to see why Microsoft would be all that good here.
>>>As a solution starting out, it's nearly perfect, of course -- my
>>>understanding is that the costs of a MS server are very cheap
>>>compared to an entry-level UNIX(R) server -- but it doesn't
>>>scale well.
>>>
>>
>>MS server would not be cheap compared to a linux one, though.
>
>I did say UNIX(R). Not Unix. One minor problem with Linux is
And I said Linux not Unix.
>the credibility gap. Microsoft is very credible; Linux is not.
Ha! Google, the city of largo, the central scotland police force,
hill house hammond, burlington coat factory, etc. etc. etc.
Plenty of high-profile organisations use linux on the desktop,
just about every large business seems to have linux somewhere.
And linux is secure.
>Why this is so, I can't precisely say,
Because it's not true, that's why.
I've removed the cobblers about stability, because it is amazingly
well established that linux is exceptionally stable.
Lee Sau Dan
The> In that particular case, that might be a good thing. :-) Less
The> chance of the Win2k server being infected or rooted. :-)
hahahahaha..... Cut the power cable, and it will NEVER be infected!
No maintenance overhead anymore! :D
The> I really fail to see why Microsoft would be all that good
The> here. As a solution starting out, it's nearly perfect, of
The> course -- my understanding is that the costs of a MS server
The> are very cheap compared to an entry-level UNIX(R) server --
The> but it doesn't scale well.
And the cost of *maintenance* is very very high. Don't forget the
cost of downtime, which is very high for situations where high
availability is required.
--
Lee Sau Dan 李守敦(Big5) ~{@nJX6X~}(HZ)
E-mail: dan...@informatik.uni-freiburg.de
Home page: http://www.informatik.uni-freiburg.de/~danlee
The Ghost In The Machine
<6ptfs9...@192.168.1.1>:
>The Ghost In The Machine wrote:
>>In comp.os.linux.advocacy, Mark Kent
>><ma...@NOHAM.otford.kent.btinternet.co.uk>
>> wrote
>>on Thu, 8 Nov 2001 06:39:19 +0000
>><na9ds9...@192.168.1.1>:
>>>The Ghost In The Machine wrote:
>>>>
>>>>Such a system would be highly scalable, as well -- replace the
>>>>Solaris machine with an S/390 and add more PCs, for example.
>>>>
>>>>I really fail to see why Microsoft would be all that good here.
>>>>As a solution starting out, it's nearly perfect, of course -- my
>>>>understanding is that the costs of a MS server are very cheap
>>>>compared to an entry-level UNIX(R) server -- but it doesn't
>>>>scale well.
>>>>
>>>
>>>MS server would not be cheap compared to a linux one, though.
>>
>>I did say UNIX(R). Not Unix. One minor problem with Linux is
>
>And I said Linux not Unix.
OK. :-)
>
>>the credibility gap. Microsoft is very credible; Linux is not.
>
>Ha! Google, the city of largo, the central scotland police force,
>hill house hammond, burlington coat factory, etc. etc. etc.
>Plenty of high-profile organisations use linux on the desktop,
>just about every large business seems to have linux somewhere.
I do wonder how long it takes to lose credibility. It's becoming
increasingly clear that Microsoft's days of dominance in the OS
world may be numbered. Linux is the joker in the pack -- a
very strong card to have -- and is supported by heavyweights
such as Dell, IBM, and HP.
(It doesn't help that Dell, last I looked, had entry-level server
hardware priced at about $1,500 (or less!) running an OS priced at
$799 or so.)
I'm now a bit curious as to how long IBM dominated the mainframe
market before being told to back off by the DOJ. I seem to
recall it was about a decade...and Windows has been dominant
for...oh, about a decade.... :-)
>
>And linux is secure.
I'm not as sure about Linux as about some other operating systems
(no, not Microsoft! :-) ), mostly because of the loadable modules,
but it's secure enough if one doesn't do anything horribly weird
like put a trojan module into /lib/modules*/... . :-)
Of course, the Linux Kernel is secure anyway; it's the other
things surrounding it that normally have problems, and those are
usually minor. (Although there is that glaring exception some
years back that lasted as a vulnerability for all of about 4
hours -- the IP fragmentation bug in the kernel. That's
arguably the most serious problem to date, although I'm not
that familiar with Ramen.)
>
>>Why this is so, I can't precisely say,
>
>Because it's not true, that's why.
>
>I've removed the cobblers about stability, because it is amazingly
>well established that linux is exceptionally stable.
That it is. 120 days and counting -- and I'm probably Tail End Charlie.
A friend of mine notified me via Email that he's had a box up for
a full year, 24/7.
And these are individual boxes, not webfarms. No IPMasq/NAT tricks here!
[.signsip]
--
ew...@aimnet.com -- insert random misquote here
EAC code #191 120d:12h:56m actually running Linux.
Hi. I'm a signature virus.
The Ghost In The Machine
<dan...@informatik.uni-freiburg.de>
wrote
on 09 Nov 2001 13:16:09 +0100
<m3hes49...@mika.informatik.uni-freiburg.de>:
>>>>>> "The" == The Ghost In The Machine <ew...@lexideb.athghost7038suus.net>
>>>>>> writes:
>
> The> In that particular case, that might be a good thing. :-) Less
> The> chance of the Win2k server being infected or rooted. :-)
>
>hahahahaha..... Cut the power cable, and it will NEVER be infected!
>No maintenance overhead anymore! :D
The best security measure for a computer is an air gap. :-)
I guess that qualifies. :-)
>
>
> The> I really fail to see why Microsoft would be all that good
> The> here. As a solution starting out, it's nearly perfect, of
> The> course -- my understanding is that the costs of a MS server
> The> are very cheap compared to an entry-level UNIX(R) server --
> The> but it doesn't scale well.
>
>And the cost of *maintenance* is very very high. Don't forget the
>cost of downtime, which is very high for situations where high
>availability is required.
I guess I should have added "small" in there somewhere. :-)
[.sigsnip]
--
ew...@aimnet.com -- insert random misquote here
EAC code #191 120d:12h:03m actually running Linux.
This is a .sig.
Ian Pegel
> In comp.os.linux.advocacy, blah
> <bl...@blah-example.net>
> wrote
> on Wed, 07 Nov 2001 18:11:22 GMT
> <47382765...@blah-example.net>:
>>The Ghost In The Machine <ew...@lexideb.athghost7038suus.net> wrote:
>>> In comp.os.linux.advocacy, Erik Funkenbusch <er...@visi.com>
>>> wrote
>>> on Tue, 6 Nov 2001 14:34:48 -0600
>>> <FPXF7.255$H7.2...@ruti.visi.com>:
>>>>"Ian Pegel" <IanP...@netscape.net> wrote in message
>>>>> What's all this about Win2k Pro being multi-user. Oh of course, that
>>>>> is one at a time. A bit like my new multi-user pencil! MS certainly
>>>>> has a way with words. "Sophistry" is one which springs to my mind.
>>>>
>>>>Win2k server includes Terminal Services, which allow multiple users to
>>>>run GUI programs at the same time. Win2k also includes a telnet
>>>>server.
No, you don't get away with that one! The ad specifically mentions Win2k
*PRO* as being multi user. It's one user at a time. Just like my pencil,
only not as reliable.
We all know about Terminal Services. What a wonderfully appropriate name.
Sort of undertakers for the OS world ;-) That's on Win2K server, not PRO.
Now do promise me you'll try harder. As far as I can see, you are
completely outclassed and have resorted to refuting points that have not
been made. A little sad, really.
--
Ian
"He knows his onions"
Erik Funkenbusch
> "The Ghost In The Machine" <ew...@lexideb.athghost7038suus.net> typed:
>
> > In comp.os.linux.advocacy, blah
> > <bl...@blah-example.net>
> > wrote
> > on Wed, 07 Nov 2001 18:11:22 GMT
> > <47382765...@blah-example.net>:
> >>The Ghost In The Machine <ew...@lexideb.athghost7038suus.net> wrote:
> >>> In comp.os.linux.advocacy, Erik Funkenbusch <er...@visi.com>
> >>> wrote
> >>> on Tue, 6 Nov 2001 14:34:48 -0600
> >>> <FPXF7.255$H7.2...@ruti.visi.com>:
> >>>>"Ian Pegel" <IanP...@netscape.net> wrote in message
> >>>>> What's all this about Win2k Pro being multi-user. Oh of course, that
> >>>>> is one at a time. A bit like my new multi-user pencil! MS certainly
> >>>>> has a way with words. "Sophistry" is one which springs to my mind.
> >>>>
> >>>>Win2k server includes Terminal Services, which allow multiple users to
> >>>>run GUI programs at the same time. Win2k also includes a telnet
> >>>>server.
>
> No, you don't get away with that one! The ad specifically mentions Win2k
> *PRO* as being multi user. It's one user at a time. Just like my pencil,
> only not as reliable.
Win2k pro includes telnet. I have no idea where this concept that if the
GUI isn't remoted it's not multiuser comes from. Many machines are
multiuser without any GUI whatsoever.
> We all know about Terminal Services. What a wonderfully appropriate name.
> Sort of undertakers for the OS world ;-) That's on Win2K server, not PRO.
>
> Now do promise me you'll try harder. As far as I can see, you are
> completely outclassed and have resorted to refuting points that have not
> been made. A little sad, really.
You should try harder. I stated that Win2k server had terminal services,
but stated that win2k had a telnet server.
T. Max Devlin
Oh, lordy lordy lordy. Erik has no idea where the concept that if the
GUI isn't remoted it's not multiuser comes from. "Many machines" (none
of them having Microsoft OSes, apparently) are multiuser without any
GUI.
Sometimes, you just have to laugh. Bwah-ha-ha-ha-ha-ha-ha!!!
Donn Miller
> Oh, lordy lordy lordy. Erik has no idea where the concept that if the
> GUI isn't remoted it's not multiuser comes from. "Many machines" (none
> of them having Microsoft OSes, apparently) are multiuser without any
> GUI.
>
> Sometimes, you just have to laugh. Bwah-ha-ha-ha-ha-ha-ha!!!
Funny how my machine has all of the above features, and I didn't have to
pay out the nose to get them. In fact, I just downloaded and installed
my OS. How much IS Windoze 2000 pro, anyways?
T. Max Devlin
>T. Max Devlin wrote:
>
>> Oh, lordy lordy lordy. Erik has no idea where the concept that if the
>> GUI isn't remoted it's not multiuser comes from. "Many machines" (none
>> of them having Microsoft OSes, apparently) are multiuser without any
>> GUI.
>>
>> Sometimes, you just have to laugh. Bwah-ha-ha-ha-ha-ha-ha!!!
>
>Funny how my machine has all of the above features, and I didn't have to
>pay out the nose to get them. In fact, I just downloaded and installed
>my OS. How much IS Windoze 2000 pro, anyways?
About 50% of the price of the hardware needed to run it, from what I've
heard.
charles lyttle
>
> On Sat, 10 Nov 2001 17:06:56 GMT, in comp.os.linux.advocacy,
> (T. Max Devlin <tm...@commercelinks.net>) wrote:
>
> >Strolling through comp.os.linux.advocacy, I heard Donn Miller say:
> >>T. Max Devlin wrote:
> >>
> >>> Oh, lordy lordy lordy. Erik has no idea where the concept that if the
> >>> GUI isn't remoted it's not multiuser comes from. "Many machines" (none
> >>> of them having Microsoft OSes, apparently) are multiuser without any
> >>> GUI.
> >>>
> >>> Sometimes, you just have to laugh. Bwah-ha-ha-ha-ha-ha-ha!!!
> >>
> >>Funny how my machine has all of the above features, and I didn't have to
> >>pay out the nose to get them. In fact, I just downloaded and installed
> >>my OS. How much IS Windoze 2000 pro, anyways?
> >
> >About 50% of the price of the hardware needed to run it, from what I've
> >heard.
>
> supply of beer!
With Linux, I saved enough on the two computers to get a third computer
and a months supply of beer
charles lyttle
>
> On Sun, 11 Nov 2001 17:11:55 GMT, in comp.os.linux.advocacy,
> (charles lyttle <lyt...@earthlink.net>) wrote:
>
> >drsquare wrote:
> >
> >> >>> Sometimes, you just have to laugh. Bwah-ha-ha-ha-ha-ha-ha!!!
> >> >>
> >> >>Funny how my machine has all of the above features, and I didn't have to
> >> >>pay out the nose to get them. In fact, I just downloaded and installed
> >> >>my OS. How much IS Windoze 2000 pro, anyways?
> >> >
> >> >About 50% of the price of the hardware needed to run it, from what I've
> >> >heard.
> >>
> >> Ouch! Use Linux and get two computers! Or get one computer and a week's
> >> supply of beer!
> >
> >With Linux, I saved enough on the two computers to get a third computer
> >and a months supply of beer
>
> beer!
That would have been a good idea, but the object was to set up a
computer lab and teach some classes. The lab needs 6-10 computers. I had
the students build their own computers as part of the class. When we get
some better workstations, these old machines are going into a Beowulf
cluster or server or some such.
The Ghost In The Machine
<tm...@commercelinks.net>
wrote
on Sat, 10 Nov 2001 17:06:56 GMT
<bnnqutooohbbm2rgh...@4ax.com>:
>Strolling through comp.os.linux.advocacy, I heard Donn Miller say:
>>T. Max Devlin wrote:
>>
>>> Oh, lordy lordy lordy. Erik has no idea where the concept that if the
>>> GUI isn't remoted it's not multiuser comes from. "Many machines" (none
>>> of them having Microsoft OSes, apparently) are multiuser without any
>>> GUI.
>>>
>>> Sometimes, you just have to laugh. Bwah-ha-ha-ha-ha-ha-ha!!!
>>
>>Funny how my machine has all of the above features, and I didn't have to
>>pay out the nose to get them. In fact, I just downloaded and installed
>>my OS. How much IS Windoze 2000 pro, anyways?
>
>About 50% of the price of the hardware needed to run it, from what I've
>heard.
It depends on the hardware. I will go with what I know, which
happens to be Dell. Selecting a PowerEdge tower server, one is
presented with a number of options, from the modest 500SC to the
pricey 6400.
The default 500SC configuration is $678.00 (saving $100.00) as of today.
It has one Celeron 900Mhz, 256MB, 1 20 GB IDE 7.2KRPM drive,
an onboard IDE and NIC, no OS, no software support, no tape backup,
a 48X IDE CDROM, keyboard, no monitor, Logitech mouse,
E-documentation, no modem, no UPS, 3 year bronze-level hardware support,
no installation, no printers.
There is no RedHat OS option with this selection. Windows 2000
server (5 clients) is +$799. They also offer NT4 and Netware 5.1.
One could, however, easily throw money at it by builking up on RAM
(2GB SDRAM = +$1694) or adding hard drives or an Ultra160 PCI SCSI
Adapter Card.
In this case, the OS costs more than the machine. To be sure,
it's not that beefy a machine -- but it does make me wonder.
The default 1400SC isn't much better; the base configuration is
$1534.00 (saving $100), which includes a single P3 1.0GhZ (it's
dual-capable; an additional 1.0Ghz P3 is +$599), 256MB,
SmartUPS 700 (+$299), onboard IDE, onboard SCSI (no PERC3-DCL RAID),
one 18GB 10KRPM U160SCSI, onboard NIC, a PV100T Travan-5 tape backup
(+$199), no tape backup software, 48x IDE CD-ROM, standard Windows
keyboard, no monitor, Logitec mouse, E-documentation, no installation
service, no modem, no printers.
Again, a Win2k 5-client server is $799.
A Win2k 5-client server for the quad-capable 6400 is still $799.
(Who'd want it?) Another option is a Win2k Advanced Server with
25 client licenses for $3295 or a Win NT 4.0 Enterprise 25-client
version for $3799.
The configured price for a 256MB 1 Xeon-900 MHZ (1MB Cache),
onboard SCSI, one non-RAID SCSI drive, onboard NIC, no OS, silver support
no tape, no firewalls -- $5,330 (saving $600.00). But this is
bare-bones; the configuration page mentions such things as 2 GB RAM
(+$1429), quad Xeon 900 Mhz 2MB cache (+$16799), split backplane
RAID5/5 (which requires a RAID card), etc.
Perhaps we need to be more careful here when discussing price. :-)
[.sigsnip]
--
ew...@aimnet.com -- insert random machine config here
EAC code #191 1d:19h:05m actually running Linux.
The US gov't spends about $54,000/second. I wish I could.