Linux-Advocacy Digest #990, Volume #34 Tue, 5 Jun 01 19:13:07 EDT
Contents:
Re: Very interesting cracker article, and XP warning. ("Ayende Rahien")
Re: Justice Department LOVES Microsoft! (Rick)
Re: UI Importance ("Ayende Rahien")
Re: Argh - Ballmer ("Ayende Rahien")
Re: Kernel comparisions ("Ayende Rahien")
Re: Completely off-topic.
Re: What Microsoft's CEO should do (Tuomo Takkula)
Re: Back up in Linux ("Interconnect")
Re: UI Importance (GreyCloud)
Re: Justice Department LOVES Microsoft! (Greg Cox)
Re: UI Importance (GreyCloud)
Re: Windows advocate of the year. ("Weevil")
Re: What Microsoft's CEO should do (GreyCloud)
Re: What Microsoft's CEO should do (GreyCloud)
Re: Kernel comparisions (Tuomo Takkula)
Re: What Microsoft's CEO should do (GreyCloud)
Re: What Microsoft's CEO should do (GreyCloud)
Re: What Microsoft's CEO should do (GreyCloud)
Re: Chicken and egg problem (flatfish+++)
----------------------------------------------------------------------------
From: "Ayende Rahien" <don'[EMAIL PROTECTED]>
Subject: Re: Very interesting cracker article, and XP warning.
Date: Wed, 6 Jun 2001 00:26:27 +0200
"Form@C" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> "Ayende Rahien" <don'[EMAIL PROTECTED]> wrote in
> news:9fhc81$qne$[EMAIL PROTECTED]:
>
> > "Form@C" <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
> >> drsquare <[EMAIL PROTECTED]> wrote in
> >> news:[EMAIL PROTECTED]:
> >>
> >> <snip>
> >> > Yeah, but come on, all those scabs living in 3rd world places are
> >> > hardly going to be able to afford to get on the internet are they?
> >>
> >> Why worry about internet demand in the 3rd world? Just the US could
> >> easily swamp the existing address space - when every computer, phone,
> >> tv, radio, fridge, freezer, toaster, coke machine, stove, alarm
> >> system, cctv system and table lamp has its own IP address! And that's
> >> only for starters...
> >
> > No way in hell.
> > We are talking *128 bits* number.
>
> Eh? for *existing* address space? Nope.... I think you must have misread
my
> post or replied to the wrong one! :-) We arn't on IPv6 yet (well, most of
> us anyway!).
Okay, I thought that you talked about existing address space in IPv6, sorry.
------------------------------
From: Rick <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy,comp.os.ms-windows.advocacy
Subject: Re: Justice Department LOVES Microsoft!
Date: Tue, 05 Jun 2001 17:32:17 -0400
Daniel Johnson wrote:
>
> "Bob Hauck" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > On Sun, 03 Jun 2001 23:24:19 GMT, Daniel Johnson
> > <[EMAIL PROTECTED]> wrote:
> > > Well, Unix is easy to port to new hardware, but
> > > Unix apps are often not easy to port to other
> > > operating systems.
> > >
> > > What's so difficult to understand about that?
> >
> > Nothing, but I'm not clear on what's different about Unix in that
> > regard. Are you saying that Windows or Mac apps are easier to port to
> > another OS? I doubt that very much.
>
> No, I'm not saying that.
>
> I think you are trying to read a "Windows-uber-alles"
> meaning into it, and I didn't mean it that way.
>
> I'm saying that Unix (the OS) is more portable
> than NT (the OS) because of their attitudes on
> endianness.
>
> It's really not as hard as Rick is making it. :D
Im not making it hard, grinning idiot. Im trying to make sure you give
straight answers to questions. You implied the OS was portable, but he
apps that run onthe OS were less portable. Very interesting position.
--
Rick
------------------------------
From: "Ayende Rahien" <don'[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy
Subject: Re: UI Importance
Date: Wed, 6 Jun 2001 00:31:17 +0200
"Woofbert" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> In article <9fieej$snh$[EMAIL PROTECTED]>, "Ayende Rahien"
> <don'[EMAIL PROTECTED]> wrote:
>
> > No me, I tend to give assignments like:
> >
> > #include <stdio.h>
> > int main(){
> > unsigned int intArr[10],i;
> > for (i=0;i<10;i++) {
> > i[intArr] = !!i;
> > printf("%u ",i[intArr]);
> > }
> > return 0;
> > }
> >
> >
> > What is the the output of this program?
>
> I think there's an error in your example. As written, I expect that to
> cause a compilation error. 'i' is not an array, yet you access it as
> one. Did you mean
No, I meant it like it was written.
It's a little trick in C that will get you fired if you snick it into
production code.
When you use [], what happens is that the what is left to the brackets is
added with what is in the brakets, and then accessed.
There is *no* provisions on the location of the point & the location of the
integer.
> and the array is superfluous.
Why is it superflous?
It has 10 slots (can't remember the term in english right now)
And it is fulled from 0 to 9.
------------------------------
From: "Ayende Rahien" <don'[EMAIL PROTECTED]>
Subject: Re: Argh - Ballmer
Date: Wed, 6 Jun 2001 00:34:34 +0200
"Mart van de Wege" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> That's just it. If the COM object in question is generic, ie it offers
> services to more than one program, it is not inextricably tied to the one
> program, then it falls under the "separate works" clause.
> This is however (with my very limited knowledge of COM, ie almost none)
COM is a component that contain a piece of functionality.
Frex, you may get a COM object which will handle SMTP, another that will
handle POP3, etc.
The idea is that you get more reuse of the code this way.
You can call the COM object from another program, from a script, etc.
------------------------------
From: "Ayende Rahien" <don'[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Kernel comparisions
Date: Wed, 6 Jun 2001 00:38:56 +0200
"Bob Hauck" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> On Tue, 5 Jun 2001 13:33:28 +1200, Stuart Fox
> <[EMAIL PROTECTED]> wrote:
>
> > They can actually, and have done since 3.5 I believe. C: is a symbolic
link
> > to \\device\harddisk0 (usually). WinObj from
http://www.sysinternals.com
> > will show you the Symbolic links used by the system. Don't confuse -
"don't
> > use commonly" with "don't use at all", the functionality is there, just
not
> > widely used.
>
> It would get a lot more use if there were user-level tools provided to
> deal with them.
There is a tool which can do it.
I believe it's called "mountvol"
------------------------------
From: [EMAIL PROTECTED] ()
Subject: Re: Completely off-topic.
Date: Tue, 5 Jun 2001 20:04:42 +0200
In article <[EMAIL PROTECTED]>,
"Paul Dossett" <[EMAIL PROTECTED]> writes:
> "Reported in Audubon Society Magazine:
>
> A Mexican newspaper reports that bored Royal Air Force pilots stationed on
> the Falkland Islands have devised what they consider a marvelous new game.
>
> Noting that the local penguins are fascinated by airplanes, the pilots
> search out a beach where the birds are gathered and fly slowly along it at
> the water's edge. Perhaps ten thousand penguins turn their heads in unison
> watching the planes go by, and when the pilots turn around and fly back, the
> birds turn their heads in the opposite direction, like spectators at a
> slow-motion tennis match. Then, the paper reports, "The pilots fly out to
> sea and turn directly to the penguin colony and over-fly it.
>
> Heads go up, up, up, and ten thousand penguins fall over gently onto their
> backs."
This is quite an old story. The RAF funded research to see if their low
flying over penguin colonies did affect the penguins. The result was
that yes penguins do follow the aircraft but they don't fall over.
--
Over 100 security bugs in Microsoft SW last year. An infamous
record. The worst offending piece of SW, by far, IIS. 2001 isn't
looking any better.
------------------------------
From: Tuomo Takkula <[EMAIL PROTECTED]>
Subject: Re: What Microsoft's CEO should do
Date: 05 Jun 2001 23:58:30 +0200
"Erik Funkenbusch" <[EMAIL PROTECTED]> writes:
> "Ayende Rahien" <don'[EMAIL PROTECTED]> wrote in message
> news:9fi8iq$md7$[EMAIL PROTECTED]...
> > "Erik Funkenbusch" <[EMAIL PROTECTED]> wrote in message
> > news:oI%S6.7225$[EMAIL PROTECTED]...
..
> > Okay, why I don't like this?
> > Why would the kernel BSOD just because the GUI crash? It should restart
> it,
> > not stop.
>
> Why should it restart it? If the GUI crashes, that means something is
> seriously wrong, and will likely just crash again.
Utter nonsense. Of course there is something wrong if the GUI crashes,
but that doesn't mean it does crash again if you restart the GUI, or
that the rest of the operating system or the running tasks is in any
way compromised by that. Under Unix, restarting the window manager
(when Netscape graps the X token and dies) simply puts the running
applications into their old state. No reboot necessary.
By your reasoning it wouldn't make sense to start the computer again
as it 'would crash again', right?
A crashing GUI is a process that violated it's resource
boundaries. There is no reason whatsoever that the operating system
should be affected by this. The GUI is not essential to run an
operating system.
And if so? Let the GUI crash as often as it wants. The rest of the
system is not probably not much interested in that anyway (speaking of
daemons and services and such), and finally, in the situation that the
GUI crashes it should be ultimately the administrators decision to
take the system down and not that of the operating system. The OS is
not qualified for that.
Best regards
Tuomo Takkula
__
Chalmers University of Technology
Computing Science
Eklandagatan 86
S-41296 Göteborg, Sweden
[EMAIL PROTECTED]
+46-31-772 1052 phone
+46-31-165 655 fax
___
"Microsoft OS's are good because they encourage Intel to produce
faster CPUs for the rest of us to run Unix on."
George Dau
------------------------------
From: "Interconnect" <mark###@logichip.com.au>
Crossposted-To:
alt.os.linux,alt.os.linux.mandrake,comp.os.linux.hardware,comp.windows.x.kde,tw.bbs.comp.linux
Subject: Re: Back up in Linux
Date: Wed, 6 Jun 2001 08:14:49 +1000
Norman Levin <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
>
> I was about to respond to this append, when I saw my "groups" in Netscape
> messenger spin like a busted roulette machine. SIX groups? Give it a
break.
> If all 6 of these groups are proper groups for this question, then there
> are clearly 5 groups to many.
>
> Les Mikesell wrote:
> >
> > "Johan Kullstam" <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
> > > "Aaron R. Kulkis" <[EMAIL PROTECTED]> writes:
> > >
> > > > > Jerry Wong wrote:
> > > > >
> > > > > I used to backup the windows by Ghost software. In linux, can the
> > command "tar"
> > > > > be used like Ghost. I means to backup the whole Linux system by
tar it
> > and
> > > > > restore it when necessary. I have windows98 and two Linux system
in my
> > PC
> > > > > (Red Hat 7.0 and Mandrake 8.0), so I can tar one of them when
running
> > the other.
> > > > >
> > > > > Is it possible? Please give me some advice.
> > > >
> > > > tar is okay for groups of files, but NOT good for the entire system.
> > > >
> > > > learn to use dump, and/or BRU or Arkeia.
> > >
> > > dump is not a sufficient solution either. linus had a few comments
> > > about this on linux-kernel about a month ago. it seems that the only
> > > way to get a clean snapshot of the disk partition is to unmount it
> > > first and then save. this is especially true of the journalling
> > > filesystems.
> >
> > Tar is actually better than dump at dealing with an active filesystem
> > but neither will handle the case where multiple files must be saved
> > as a snapshot of a consistent state. What we need is kernel support
> > to freeze the 'real' filesystem while letting the system continue to
work
> > with changes staying in the buffers or paging out to swap if necessary -
and
> > of course, a way for the backup utility to read the frozen copy.
> >
> > Les Mikesell
> > [EMAIL PROTECTED]
>
> --
> Norman Levin
Yes Linux can easily back up your entire system. BASH plus tar plus cron =
problem solved. No need to get extra software.
------------------------------
From: GreyCloud <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy
Subject: Re: UI Importance
Date: Tue, 05 Jun 2001 15:06:44 -0700
Ayende Rahien wrote:
>
> "drsquare" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > On Tue, 5 Jun 2001 03:58:36 +0200, in comp.os.linux.advocacy,
> > ("Ayende Rahien" <don'[EMAIL PROTECTED]>) wrote:
> >
> > >> Compare that to a GUI, where you don't even get any options, you just
> > >> get what you're given, unless of course you want to go editing
> > >> shorcuts to put some arguments in, or using some shoddy built in
> > >> configuration tool. Personally, I like complete control over what I'm
> > >> doing with my command.
> > >
> > >What options do you get on the CLI that aren't given you?
> >
> > Arguments.
>
> Exist in GUI as well.
>
> > Pipes. Redirection.
>
> Are those the same thing? If not please explain the difference.
>
> > Command history.
>
> Not needed because it's not in the GUI nature.
Except of course in the drop down history panel of the URL in IE 5.5.
:-)
--
V
------------------------------
From: Greg Cox <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy,comp.os.ms-windows.advocacy
Subject: Re: Justice Department LOVES Microsoft!
Date: Tue, 05 Jun 2001 22:10:09 GMT
In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED] says...
> Daniel Johnson wrote:
> >
> > "Bob Hauck" <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
> > > On Sun, 03 Jun 2001 23:24:19 GMT, Daniel Johnson
> > > <[EMAIL PROTECTED]> wrote:
> > > > Well, Unix is easy to port to new hardware, but
> > > > Unix apps are often not easy to port to other
> > > > operating systems.
> > > >
> > > > What's so difficult to understand about that?
> > >
> > > Nothing, but I'm not clear on what's different about Unix in that
> > > regard. Are you saying that Windows or Mac apps are easier to port to
> > > another OS? I doubt that very much.
> >
> > No, I'm not saying that.
> >
> > I think you are trying to read a "Windows-uber-alles"
> > meaning into it, and I didn't mean it that way.
> >
> > I'm saying that Unix (the OS) is more portable
> > than NT (the OS) because of their attitudes on
> > endianness.
> >
> > It's really not as hard as Rick is making it. :D
>
> Im not making it hard, grinning idiot. Im trying to make sure you give
> straight answers to questions. You implied the OS was portable, but he
> apps that run onthe OS were less portable. Very interesting position.
>
Look at it this way: You can have NT with all of your NT apps running on
a big-endian CPU. It will run just fine. Where you get into trouble is
moving any files that contain binary data between a big-endian box and a
little-endian box. For instance, Microsoft Word writes binary data in
its .doc files and assumes little-endian format. A version of Microsoft
Word running on a big-endian CPU would either have to read/write its
binary files using a big-endian/little-endian translation layer for
every single binary data access to maintain compatability or completely
forgo compatability (probably unacceptable). This translation layer
creates a real performance problem on the big-endian box (since little-
endian is far more common, big-endian gets pimped). Since very little
of the OS gets moved between boxes, the apps are more sensitive to the
big-endian/little-endian interoperation problem.
--
[EMAIL PROTECTED]
------------------------------
From: GreyCloud <[EMAIL PROTECTED]>
Subject: Re: UI Importance
Date: Tue, 05 Jun 2001 15:11:40 -0700
Robert Morelli wrote:
>
> This is typical UNIX arrogance. The day has long since passed when
> UNIX could look down its technological nose at Windows. On the whole,
> UNIX software is nowhere near the quality of Windows software, and
> hasn't been for many years.
>
Baloney!
------------------------------
From: "Weevil" <[EMAIL PROTECTED]>
Subject: Re: Windows advocate of the year.
Date: Tue, 05 Jun 2001 22:16:27 GMT
Edward Rosten <[EMAIL PROTECTED]> wrote in message
news:9fja6g$37a$[EMAIL PROTECTED]...
> In article <HE9T6.8048$[EMAIL PROTECTED]>, "Erik Funkenbusch"
> <[EMAIL PROTECTED]> wrote:
>
> > "Donn Miller" <[EMAIL PROTECTED]> wrote in message
> > news:871yozutch.fsf@acs-24-154-37-
> >> I would say Erik F., but at times he makes excuses for Microsoft intead
> >> of just admitting "well, OK, MS really shouldn't do things this way".
> >
> > I don't "make excuses". You are simply judgemental, and unwilling to
> > accept that any viewpoint other than your own can be right. I don't see
> > things in black and white, but rather shades of grey.
>
> The rest of us see in colour ;-)
>
> -Ed
>
Go easy on Eric. He is, after all, the guy who argued that the sky isn't
really blue; it merely has a blue appearance. Or something like that.
--
Weevil
"The obvious mathematical breakthrough [for breaking encryption schemes]
would be development of an easy way to factor large prime numbers."
-- Bill Gates
------------------------------
From: GreyCloud <[EMAIL PROTECTED]>
Subject: Re: What Microsoft's CEO should do
Date: Tue, 05 Jun 2001 15:19:24 -0700
Stuart Fox wrote:
>
> "pip" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > daniel wrote:
> > >
> >
> > The core of the OS has not, and should not have anything to do with it's
> > interface utilities in terms of robustness. Here you may insert the
> > problems of GDI being in kernel space.
> >
>
> If you read "Inside Windows 2000", it thoroughly debunks the myth that it's
> bad for stability (from a guy with access to the source, and a guy with
> um... SoftICE)
>
> > > With respect to open standards and RFCs again the way Microsoft
> > > operates is a mess. With as much weight as they have why don't they
> > > realize that operating with a profit-first mentality in a state of
> > > paranoia they stunt their potential to work in a flourishing
> > > environment with the developer community of the world and, should they
> > > choose to contribute (not dictate) to the development of standards
> > > would gain greater acceptance and realize greater benefit to users
> > > around the world?
> >
> > Erm. Look here is a very simple rule :
> >
> > M$ cares about PROFITS not USERS. Of course it will listen to some
> > extent to it's users, but only to keep them from the alternatives.
> > Anything more is a waste of money.
> >
> Have you thought about where their profit comes from? That's right, their
> users. If they don't keep them happy, they'll lose them.
Well, they are losing this one user. With the damn licensing nonsense
MS has going for XP,... I'm giving that o/s about 20 ft. clearance from
me.
--
V
------------------------------
From: GreyCloud <[EMAIL PROTECTED]>
Subject: Re: What Microsoft's CEO should do
Date: Tue, 05 Jun 2001 15:21:34 -0700
Erik Funkenbusch wrote:
>
> "Ayende Rahien" <don'[EMAIL PROTECTED]> wrote in message
> news:9fhjt5$49d$[EMAIL PROTECTED]...
> >
> > "Stuart Fox" <[EMAIL PROTECTED]> wrote in message
> > news:9fhcis$bqn$[EMAIL PROTECTED]...
> > >
> > > "pip" <[EMAIL PROTECTED]> wrote in message
> > > news:[EMAIL PROTECTED]...
> > > > daniel wrote:
> > > > >
> > > >
> > > > The core of the OS has not, and should not have anything to do with
> it's
> > > > interface utilities in terms of robustness. Here you may insert the
> > > > problems of GDI being in kernel space.
> > > >
> > >
> > > If you read "Inside Windows 2000", it thoroughly debunks the myth that
> > it's
> > > bad for stability (from a guy with access to the source, and a guy with
> > > um... SoftICE)
> >
> > I don't have this book, and orderring it will take a month.
> > Can you give a list of the reasons?
>
> Basically, the reason is that the way NT is designed, if the GUI subsystem
> faults, then the OS blue screens anyways, whether or not it runs in kernel
> space. The OS's main thread drops to a blue screen when the GUI subsystem
> dies.
That was one of MS most dumbest decisions... bringing the GUI into ring
0. It should have been kept out in another ring.
--
V
------------------------------
From: Tuomo Takkula <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Kernel comparisions
Date: 06 Jun 2001 00:25:43 +0200
"Ayende Rahien" <don'[EMAIL PROTECTED]> writes:
> What is mmap, exactly?
It's great! In very simple terms, it maps a file (or some other
object) to memory. That means, you can access a file (or other object)
as you would access an array of, say, characters. This removes a lot
of headaches when you have to process a file or to deal with memory
shared between processes. You have to care for buffering, but the OS
will take care to give you only the parts of the file/object that you
actually want to access and by that it keeps things as fast as
possible.
Best regards
Tuomo Takkula
___
"Microsoft OS's are good because they encourage Intel to produce
faster CPUs for the rest of us to run Unix on."
George Dau
------------------------------
From: GreyCloud <[EMAIL PROTECTED]>
Subject: Re: What Microsoft's CEO should do
Date: Tue, 05 Jun 2001 15:24:08 -0700
Ayende Rahien wrote:
>
> "Erik Funkenbusch" <[EMAIL PROTECTED]> wrote in message
> news:oI%S6.7225$[EMAIL PROTECTED]...
> > "Ayende Rahien" <don'[EMAIL PROTECTED]> wrote in message
> > news:9fhjt5$49d$[EMAIL PROTECTED]...
>
> > > >
> > > > If you read "Inside Windows 2000", it thoroughly debunks the myth that
> > > it's
> > > > bad for stability (from a guy with access to the source, and a guy
> with
> > > > um... SoftICE)
> > >
> > > I don't have this book, and orderring it will take a month.
> > > Can you give a list of the reasons?
> >
> > Basically, the reason is that the way NT is designed, if the GUI subsystem
> > faults, then the OS blue screens anyways, whether or not it runs in kernel
> > space. The OS's main thread drops to a blue screen when the GUI subsystem
> > dies.
> >
>
> Okay, why I don't like this?
> Why would the kernel BSOD just because the GUI crash? It should restart it,
> not stop.
>
> This doesn't sound right, and it's certainly not an excuse.
> It would crash *anyway* ?
>
> Beside, I understand that Win2K2 can boot without GUI.
Your description makes a lot more sense. If I remember right David
Cutler designed NT. And he also was the main architect of vms... and vms
has 6 ring levels of code isolation. This would keep the kernel from
dropping when the GUI croaked.
--
V
------------------------------
From: GreyCloud <[EMAIL PROTECTED]>
Subject: Re: What Microsoft's CEO should do
Date: Tue, 05 Jun 2001 15:26:53 -0700
pip wrote:
>
> Ayende Rahien wrote:
> > > Basically, the reason is that the way NT is designed, if the GUI subsystem
> > > faults, then the OS blue screens anyways, whether or not it runs in kernel
> > > space. The OS's main thread drops to a blue screen when the GUI subsystem
> > > dies.
> > >
> >
> > Okay, why I don't like this?
> > Why would the kernel BSOD just because the GUI crash? It should restart it,
> > not stop.
> >
> > This doesn't sound right, and it's certainly not an excuse.
> > It would crash *anyway* ?
> >
> > Beside, I understand that Win2K2 can boot without GUI.
>
> Good point! It does not restart because it CAN'T. In windows I have had
> explorer faults which cause explorer to re-initialise, but a blue screen
> by definition means a dead system.
>
> BTW it should be noted by both sides that a graphics driver bug could
> cause either system to die a death no matter if the GUI "service" is run
> in user-space.
I've never seen that in Sun OS. I've seen it happen once when the
X-server died but the system stays up. I've seen Netscape on occasion
just disappear into thin air... the O/S got rid of an errant process.
If I were the CEO I'd re-instate the ring levels. This would increase
stability.
--
V
------------------------------
From: GreyCloud <[EMAIL PROTECTED]>
Subject: Re: What Microsoft's CEO should do
Date: Tue, 05 Jun 2001 15:29:05 -0700
Ayende Rahien wrote:
>
> "Erik Funkenbusch" <[EMAIL PROTECTED]> wrote in message
> news:My9T6.8045$[EMAIL PROTECTED]...
> > "Ayende Rahien" <don'[EMAIL PROTECTED]> wrote in message
> > news:9fi8iq$md7$[EMAIL PROTECTED]...
>
> > >
> > > Okay, why I don't like this?
> > > Why would the kernel BSOD just because the GUI crash? It should restart
> > it,
> > > not stop.
> >
> > Why should it restart it? If the GUI crashes, that means something is
> > seriously wrong, and will likely just crash again.
>
> Why should the GUI crashing cause a full system halt?
> NT is aimed at servers, not just desktops. This just doesn't makes sense.
> Other platforms don't crash if there is a crash in the GUI (well, not
> always.)
>
> > > This doesn't sound right, and it's certainly not an excuse.
> > > It would crash *anyway* ?
> >
> > Yes.
>
> Why? What is the reason for this decision?
Speed.
--
V
------------------------------
From: flatfish+++ <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Chicken and egg problem
Date: Tue, 05 Jun 2001 22:32:33 GMT
On Tue, 5 Jun 2001 23:57:51 +0200, "Ayende Rahien" <don'[EMAIL PROTECTED]>
wrote:
Part of the death knoll to OS/2 was when Wordperfect introduced WP for
OS/2 and it was actually a Windows program that was not as good as the
Windows version.
Best Wordpro at the time for OS/2 was Describe.
IBM made a tough decision and chose to maintain backward compatibility
and to support Windows applications and it turned out to be the wrong
decision because there was no reason to run Windows applications under
OS/2 when people could run them under Windows.
I have always believed that if IBM had cut the cord and went a
straight OS/2 path we might all be running OS/2 now instead of
Windows.
flatfish+++
"Why do they call it a flatfish?"
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list by posting to comp.os.linux.advocacy.
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Advocacy Digest
******************************
