Linux-Advocacy Digest #55, Volume #29 Mon, 11 Sep 00 15:13:09 EDT
Contents:
Re: Criteria in Evaluating Distributions:
Re: Computer and memory
Re: The internet was built on WIndow 95? (was Re: How low can they go...?)
([EMAIL PROTECTED])
Re: ZDNet reviews W2K server; I think you'll be surprised....
Linux through the ages (Nathaniel Jay Lee)
Re: [Q] linux on mac? (Jim Richardson)
Re: [Q] linux on mac? (Jim Richardson)
Re: Vs: ZDNet reviews W2K server; I think you'll be surprised....
([EMAIL PROTECTED])
Re: Sherman Act vaguery [was: Would a M$ Voluntary Split Save It?] ("2 + 2")
Re: ZDNet reviews W2K server; I think you'll be surprised.... ("Keith T. Williams")
Re: End-User Alternative to Windows (Wowie! It's Howie Zowie!)
Re: How low can they go...?
Re: How low can they go...?
Re: Enemies of Linux are MS Lovers (Damien)
----------------------------------------------------------------------------
From: <[EMAIL PROTECTED]>
Subject: Re: Criteria in Evaluating Distributions:
Date: Mon, 11 Sep 2000 10:44:29 -0700
Reply-To: <[EMAIL PROTECTED]>
Lee Reynolds <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Before anyone even thinks about getting a linux system online they'd
> better be prepared to lay down about a hundred bucks for an external
> modem.
Yes, before anyone thinks about getting a Linux system online they should
aquire a good external serial modem, but that this is true for *any* OS.
External serial modem are more portable across hardware platforms and
therefore maintain that vaule of the investment into them all the better
than internal modem. That said, if someone perfers to go the internal modem
route and thereby locking their modem into their current hardare platform,
there are plenty of real hardware modems available. Linux and use these as
can any other OS available for that particular hardware platform and though
it were an external model, because it does provide its own UART and appears
as a serial port to the hardware and the OS so viable,
Then come the software modem (winmodems). There are those that have had
Linux drivers created for them either by reverse engineering or from key
informations provided by the manufacterers. There are are a few
manufacturers who are now beginning to supply Linux drivers for their
software modems. For more information of Linux compatible software modems
called linmodems check out: http://www.linmodems.org or the on-line
information available from the manufacturers. If the manufacturer of your
"favorite" software modem does not support Linux contact them and and
encourage them to do so.
Software modems do require considerable processing power from their host
computers as well as the special drivers. That is why you often see the
claim that you need a Pentium class processor to handle certain modem
speeds. This is foolishness since a 386DX can run perfectly well at top
speed with a v.90 or other 56K modem protocol, providing the host has a good
highspeed UART that has a working multibyte buffer recieve buffer. But then
this is true for a 1 Ghz processor as well, if it drives a 8250 or a 16450
with a one byte recieve buffer, there will be lost bytes.
Software modems are devices with limited useful lifespans, they will only be
usable in newer versions of the supporting OS as long as the drivers for
that new version are made available for that new version. One the hardware
manufacturer nolonger activly support the unit the purchaser will be
dependent on the luck that the future version of the preferred OS will
include drivers for that unit.
I have never been foolish enough to aquire a software modem, but I have help
others to try to salvage their investment in software modem when the
manufacturer has abandoned them. In every case so far, the modems that I
have so encountered were manufactured from four to six years ago, the
efforts to have them work with current releases of Windows has been futile.
Your statement that an external modem will cost $100.00 is an exageration of
at least 20%. A decade ago you estimate would have been on the low side;
however, it does not reflect today's market place, where $80.00 for such a
device is on the high side and $50.00 to $60.00 is common. Unless you are
looking for some special unit designed for a mainframe or some exotic
situation.
------------------------------
From: <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Computer and memory
Date: Mon, 11 Sep 2000 10:57:16 -0700
Reply-To: <[EMAIL PROTECTED]>
Donovan Rebbechi <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> On Mon, 11 Sep 2000 16:23:36 +0200, Matthias Warkus wrote:
> >It was the Fri, 08 Sep 2000 13:50:14 GMT...
> >...and Nathaniel Jay Lee <[EMAIL PROTECTED]> wrote:
> >> Funny how he says the Europeans are 'tainted' by ignorant
> >> and oblivious words. Jesus, I wonder if that guy is
> >> reading a different USENET than the rest of us? I seem to
> >> see more blathering and totally idiotic statements out of
> >> US connected people than out of the European connected
> >> people.
> >
> >Correction: The US posters do not only produce more arrogant and
> >idiotic postings that all of the European posters. They produce more
> >of then that all other posters together, and I estimate that, in
> >international groups like this, the amount of US bullshit is at least
> >thrice the amount of rest-of-the-world bullshit.
>
> Well the fact that the Americans are allowed to post in their first
> language but the Europeans are not lowers the bar somewhat for the
> AMericans. The insular Europeans rarely get to post in the first place.
There are also newgroups such as de.* fr.* fj.* it.* in which the standard
language is not English. That levels the field a bit and it would be
nessary to examine the contents of those newgroups as well before such a
blanket statement as "more blathering and totally idiotic statements out of
US connected people than out of the European connected
people." can be made.
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To:
comp.lang.java.advocacy,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: The internet was built on WIndow 95? (was Re: How low can they go...?)
Date: 11 Sep 2000 14:05:16 -0400
R.E.Ballard ( Rex Ballard ) <[EMAIL PROTECTED]> writes:
> The first graphical Web
> Browser - Viola, was written exclusively for UNIX.
You might say "the first web browser that could display inline images."
The first point-and-click web browser was written for the NeXT. Then
came a command-line Unix browser that looked pretty clunky. This was
all Tim Berners-Lee had to show when he came to MIT in 1991 talking as
if this WWW project was going to take over the Internet. The lead
TechInfo developers didn't believe him, and neither did I when they
showed me what he had showed them.
http://browsers.evolt.org/worldwideweb/NeXT/WorldWideWeb.html
--
Bruce R. Lewis http://brl.sourceforge.net/
------------------------------
From: <[EMAIL PROTECTED]>
Crossposted-To: alt.destroy.microsoft
Subject: Re: ZDNet reviews W2K server; I think you'll be surprised....
Date: 11 Sep 2000 18:01:41 GMT
In alt.destroy.microsoft Donovan Rebbechi <[EMAIL PROTECTED]> wrote:
> It's absurd to compare aggressive marketting and vendor
> lock to wholesale destruction of the environment. Your post
> is diversionary nonsense.
What do you think happens to all the RAMs, Hard-disk, casing, diskette's,
CPUs that get thrown out when the next offering from the Dark Lords
require an upgrade?
You think wholesale waste of the earth's resource feeding the Redmond
behemouth is good for the earth's environment?
------------------------------
From: [EMAIL PROTECTED] (Nathaniel Jay Lee)
Subject: Linux through the ages
Date: Mon, 11 Sep 2000 18:09:54 GMT
Reply-To: [EMAIL PROTECTED]
This weekend my wife and I attended a Renaissance Faire in
Minnesota. I've been to a few of these before, and was
looking forward to it just for the sake of going. But
while there, I was pleasantly suprised at something.
I wore a Linux t-shirt to the faire (and no, that's not a
misspelling). It has a penguin on the front, and a quote
on the back from Linus, "The Linux philosophy is 'laugh in
the face of danger'. Oops, wrong one. It's 'do it
yourself'. Yeah, that's it."
Anyway, I had several people comment on the shirt (most of
them dressed in Renaissance garb and speaking in 'old
English'). The first was a woman in a traditional English
gown and cape. She just said that she liked the shirt,
'it meks me lawff'.
The comments that caught my attention however was when I
was looking through the swords in a shop (hobby of mine,
sword collecting). One of the 'shop-keepers' came up and
started asking about my shirt. Pretty soon we were having
a big conversation about Linux (kind of interesting
considering he was wearing Elizabethan era clothing and
speaking in old English). While we were talking we were
joined by a group of 'Robin-Hood' people (they were a part
of one of the shows) and three of them started talking
about Linux with us.
The funny thing to me is the fact that here we were in a
Renaissance atmosphere, many of the people wearing the
ancient clothing, standing in a sword shop, surrounded by
nothing but dust and 'old technology' and discussing
Linux. I just thought it was an amusing story. (And my
wife laughed because 'all of you geeks find eachother,
even in a different time period'.)
--
[EMAIL PROTECTED]
Nathaniel Jay Lee
------------------------------
From: [EMAIL PROTECTED] (Jim Richardson)
Subject: Re: [Q] linux on mac?
Date: Mon, 11 Sep 2000 11:19:50 -0700
Reply-To: [EMAIL PROTECTED]
On 11 Sep 2000 14:22:18 GMT,
Brian Langenberger, in the persona of <[EMAIL PROTECTED]>,
brought forth the following words...:
>Rich C <[EMAIL PROTECTED]> wrote:
>: "." <[EMAIL PROTECTED]> wrote in message
>: news:8pgu4t$1jt2$[EMAIL PROTECTED]...
>:> Anon Y. Mous <[EMAIL PROTECTED]> wrote:
>:> > if i install linux on a mac, how difficult is the conversion?
>:>
>:> It depends on what you mean by "conversion". LinuxPPC has become
>:> (after quite a long time) incredibly easy to install. Same for
>:> Yellowdog. Theyre both redhat based.
>:>
>:> You of course wont be able to run any of your mac software, but
>:> it could be argued that that is a Good Thing (TM).
>:>
>: Perhaps the hardest thing would be converting that cyclops mouse to
>: something with 2 or 3 buttons. You need to middle click to paste in many
>: Linux apps. You can simulate a third button with 2 buttons, but how do you
>: simulate a third button with only one?
>
>But since all the new Macs are USB-based, it shouldn't be
>tough to plug in a nice 3-button mouse to one. I might make my
>next box a Mac (for quietness and cool-running chips) when I'm
>sure the MacOS can be exorcised completely.
>
I used MkLinux for a year or two on a PPC, it worked nice. Until I got the 3
button trackball I had to use the option and apple keys as modifiers for the
other 2 buttons, a pain, but doable.
Never did get entirely rid of Macos thanks to the open firmware, but I hear
LinuxPPC does, although I have never used it.
(Jim, who sometimes wonders about buying an Imac and Linuxing it...)
--
Jim Richardson
Anarchist, pagan and proud of it
WWW.eskimo.com/~warlock
Linux, because life's too short for a buggy OS.
------------------------------
From: [EMAIL PROTECTED] (Jim Richardson)
Subject: Re: [Q] linux on mac?
Date: Mon, 11 Sep 2000 11:21:33 -0700
Reply-To: [EMAIL PROTECTED]
On Mon, 11 Sep 2000 02:55:47 -0400,
Anon Y. Mous, in the persona of <[EMAIL PROTECTED]>,
brought forth the following words...:
>apparently, in the boot record of the system disk on the macintosh, there is
>a pointer to a second operating system. there once was a package that
>allow the system to restart into the other OS, etc. it is my guess that
>this was a feature that apple employed for debugging their developments.
>
>what i am wondering is if the linux OS for the macs use this. that's the
>conversion that is of interest...
>--
>*please post to NG only*
>
>
>----------
>In article <8pgu4t$1jt2$[EMAIL PROTECTED]>, [EMAIL PROTECTED] (.) wrote:
>
>
>> Anon Y. Mous <[EMAIL PROTECTED]> wrote:
>>> if i install linux on a mac, how difficult is the conversion?
>>
>> It depends on what you mean by "conversion". LinuxPPC has become
>> (after quite a long time) incredibly easy to install. Same for
>> Yellowdog. Theyre both redhat based.
>>
>> You of course wont be able to run any of your mac software, but
>> it could be argued that that is a Good Thing (TM).
>>
>>
There is a Mac emulator that ships with LinuxPPC I think, but I don't know
how well it works. (I would dearly love to be able to play Escape velocity
again.)
--
Jim Richardson
Anarchist, pagan and proud of it
WWW.eskimo.com/~warlock
Linux, because life's too short for a buggy OS.
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: alt.destroy.microsoft
Subject: Re: Vs: ZDNet reviews W2K server; I think you'll be surprised....
Date: Mon, 11 Sep 2000 18:08:55 GMT
In article <d74v5.185$[EMAIL PROTECTED]>,
"Ville Niemi" <[EMAIL PROTECTED]> wrote:
> The keyword is not 'some' its 'unnecessary'. Any product that requires
> changes, however trivial, on existing installations is bad design.
Sometimes
> you need to do design compromises, and settle for less than best
design.
> Specifically if you make a change that affects interoperability you
should
> have a very good reason for it. As in 'had to do it this way'.
Question
> really is 'Had M$ no other way of doing this?'
>
> The point many in this NG are trying to make is that M$ deliberately
CHOSE
> to create this problem in order to monopolize the market. From the
users
> viewpoint this would seem as 'unnecessary' as it can be, and
consequently
> the product is BAD design, and to reiterate, 'crap'.
>
> Of course, if you have a different definition of bad design...
>
> Ville
>
>
I have the same feeling when I consider 'Had M$ no other way of doing
this?'. I can use Word97 to exchange files with Word95 in Word95
format; I can run DOS and Win31 programs without LFN conflict; I use
IE5 which supports DHTML and includes some MS extensions, yet it still
can properly render a HTML2 document. All these do not require me to
write a single line of code. I expect an implementation with new (or
proprietory) extensions can be cooperate with existing implementations
which do not require these new features. Therefore, I will regard it as
AD with partial DNS compatibility. After all, it is a bold step to
switch established DNS service to a new released platform.
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: "2 + 2" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.os.os2.advocacy,comp.sys.mac.advocacy
Subject: Re: Sherman Act vaguery [was: Would a M$ Voluntary Split Save It?]
Date: Mon, 11 Sep 2000 14:33:22 -0400
T. Max Devlin wrote in message ...
>Said 2 + 2 in comp.os.linux.advocacy;
>>T. Max Devlin wrote in message ...
> [...]
>>>Yes, I'm aware of how free markets work. What makes you think that
>>>markets where prices are controlled and competition excluded by a
>>>monopoly follow these rules?
>>
>>It may or may not.
>>
>>You're letting definitions rather than substance rule.
>
>It does not, and there is no question about that. I'm letting the 'rule
>of reason' and a clear understanding of anti-trust law rule. You're
>just engaging in idle speculation in defense of a monopolist.
>
>>It makes no difference whether you call it a "monopoly" or a market
leader.
>>That's just playing with definitions.
>
>Tell it to the judge. He knew it was bullshit when Microsoft's lawyers
>tried that, but maybe he'll listen to you, huh?
>
>>What is required is a further analysis of whether
>
>
>Apparently a premature transportation. Would you like to try again, or
>do you see my point, now?
Finally, we are agreed on something. :)
Let me say some things about remedies.
This is the area I have focused on.
Again, I think there is a good chance of a monopoly finding on the desktop
OS.
The tech tying issue is a big "if" in my mind. I don't have the famous Max
crystal ball where he knows the outcome. In the law, there's no crystal
ball, and only something more like a football bouncing erradically on the
turf.
But let's assume a remedy is needed.
It's clear that in the existing order, by how it was marked up by the sides,
that Microsoft will divest the OS part. Let's call the divested part " New
Windows" or just Windows.
I would require divestment of the desktop OS defined as the line from DOS to
whereever the present code base is in that line.
Note: competive = competitive
That's the desktop OS. It would get the rights to the Windows trademark.
The rest of Microsoft would include the server OS and the workstation OS
(not on an exclusive basis if the code bases had been integrated).
I would set up immediate competition in the desktop OS market for the
benefit of consumers. Microsoft would be expected and encourage to field a
competiting consumer OS product on the basis of its workstation OS.
And this product would be required to run all apps of the new Windows
entitiy for a short number of years. At which point, there would be a
market driven necessity to run mutual apps.
This new Windows, which would have the dominent PC market, would have cross
licensing rights with the rest of the Microsoft OS platform to ensure its
technological basis or equivalence.
This company would have multi-billion dollar profits at its onset. It would
be expected to enter the server OS market immediately, since it would have
the cross-license to the Windows 2000 Server code. App compatibility would
be required for a short number of years.
Each of the two platforms would have cross licensing rights to all
middleware, including browser, transaction processing, etc.
Turning from split remedies to conduct remedies, the new Windows would be
required to support all important middleware. A panel of experts would be
set up to determine what exactly would be included.
All new middleware included, such as .NET middleware would be required to
reciprocate such that "Windows" apps ran on those "coming into the fold" of
middleware inclusion. This would extend interoperability greatly for the
benefit of consumers.
No living off the desktop consumer without paying. This would end all this
two-faced "kill the desktop" talk, without evaluating its impact on the
desktop consumer. Let the market kill the desktop if it wants, but don't
have the DoJ doing it, as an unintended consequence. Or perhaps as a
consequence of all those antitrust lawyers having conflicts of interest from
the dollar signs in their eyes from the new Gilded Age.
Is the highjacking of a split that does not end the monopoly anything more
that the desire to feed off that monopoly? End the monopoly even though it
hurts the feeing of lawyers.
The Java Platform would probably be grandfathered in. It's probably unfair
to make any kind of reciprocity requirements retroactive.
Basically, instead of dictating technology, this solution attempts to ensure
its availability.
Instead of stripping the middleware out, it focuses on stripping the
monopoly out, as defined by the case findings of the relevant market.
Once the monopoly is divested, then the DoJ really has no interest, except
enforcing the inclusion of middleware in "Windows," via a panel of experts.
The inclusion of middleware would go to having access to competing browsers
as icons for the user. Or OEMs could have no icons presented, but for
fairness, the underlying code would be there for any browser or other form
of middleware that the panel requires included.
All kinds of middleware, like AOL's messenger product would be included, as
the panel ruled.
Now you can get into details about whether OEMs would have to include all of
this code. Certainly the basic plumbing would have to be there to make
competing products work.
Also, the new Windows company would have a good argument that it has a
competitive need to offer a uniform product. So, as long as competing
middleware was included, with access on a "fair" basis, this argument makes
sense.
Basically, the panel would determine what is fair competiton. If it is
included and it works without interference from Windows where the Windows
has a competing product, then fine.
2 + 2
That would
>
>--
>T. Max Devlin
> -- Such is my recollection of my reconstruction
> of events at the time, as I recall. Consider it.
> Research assistance gladly accepted. --
>
>
>-----= 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! =-----
------------------------------
From: "Keith T. Williams" <[EMAIL PROTECTED]>
Crossposted-To: alt.destroy.microsoft
Subject: Re: ZDNet reviews W2K server; I think you'll be surprised....
Date: Mon, 11 Sep 2000 14:53:35 -0400
<[EMAIL PROTECTED]> wrote in message
news:8pj6m5$l6j$[EMAIL PROTECTED]...
> In alt.destroy.microsoft Donovan Rebbechi <[EMAIL PROTECTED]> wrote:
>
> > It's absurd to compare aggressive marketting and vendor
> > lock to wholesale destruction of the environment. Your post
> > is diversionary nonsense.
>
> What do you think happens to all the RAMs, Hard-disk, casing, diskette's,
> CPUs that get thrown out when the next offering from the Dark Lords
> require an upgrade?
>
> You think wholesale waste of the earth's resource feeding the Redmond
> behemouth is good for the earth's environment?
>
Nobody is forced to UPGRADE, if what you have works for you, keep it.
------------------------------
From: Wowie! It's Howie Zowie! <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,comp.os.linux.misc
Subject: Re: End-User Alternative to Windows
Date: Mon, 11 Sep 2000 18:39:32 GMT
In article <[EMAIL PROTECTED]>,
sinister-catsup <[EMAIL PROTECTED]> wrote:
> robert w hall <[EMAIL PROTECTED]> wrote:
> > In article <[EMAIL PROTECTED]>, The Ghost
> > In The Machine <[EMAIL PROTECTED]> writes
> > >All in all, a nice little system, doing things which OS/2 had
> > >problems in < 4 meg of memory at the time (and DOS couldn't do
> > >at all).
> > >
> > >But Commodore dies, and Windows wins out over all. Go fig.
> > >
> > >[snip for brevity]
> > >
> > But, long before, there was OS9 level 2 on the 6809 - a system not to b=
> e
> > surpassed by Intel until the '386 (and Linux) ...
> >=20
> > and coming up to date (well fairly) - what's wrong with keeping the
> > horrible M$ stuff but running it under win4lin under linux - thus
> > keeping windows 9x in its right place as an ordinary task in linux
> > user-mode - see enthusiastic thread on alt.os.linux
>
> thats what I do, but I really only use windoze for games, someone needs t=
> o get
> off their buttocks and make an emulated workable version of directx, when=
> that
> happens Redmond truly can bite me. I've tried playing with installing dir=
> ectx
> 7.0 on win4lin and I have gotten some curious results. Directdraw kinda w=
> orks,
> direct3d, forget about it :(
>
>
You could always stick to opengl. Heck, on my win98 machine at home, I only
play one game that uses any directx libs. But, when the directx stuff is
emulated properly, you can bet yer bottom dollar that I'll use my linux box
for games full time!
--
Howie Zowie (Wowie)
Year 2k Neo-Antiprophet
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: <[EMAIL PROTECTED]>
Crossposted-To:
comp.lang.java.advocacy,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: How low can they go...?
Date: Mon, 11 Sep 2000 11:10:01 -0700
Reply-To: <[EMAIL PROTECTED]>
Simon Cooke <[EMAIL PROTECTED]> wrote in message
news:dk2v5.69542$[EMAIL PROTECTED]...
>
> <[EMAIL PROTECTED]> wrote in message
> news:8ph3mm$n3c$[EMAIL PROTECTED]...
> > For software like the newer versions of Quicken the "programmers" have
> > surrendered the control of the appearence, quality, performance, and
> > behavior of their software to the whims of those who developed the
> rendering
> > software AKA browser. This is not the action of rational, experienced,
> > competent, professional programmers. HTML is useful and valid for its
> > intended purpose but this is not it.
>
> Not even if:
>
> (1) it provides an interface that users are comfortable with (the web)
> (2) it allows for the inclusion of lots of text on the screen to explain
the
> UI
> (3) the interface was originally forms based, and as such could be easily
> turned into a single page with hyperlinks?
First:
How do you address the issue of the loss of control of the appearance,
quality, performance, and behavior of the software to the whims of thoses
who developed the rendering software AKA browser. How do you also address
the issue of the program's user interface going haywire should a minor
upgrade via a service pack cause the renderer to no longer render the
program's user interface the way that the programmers of the program had
intended?
------------------------------
From: <[EMAIL PROTECTED]>
Crossposted-To:
comp.lang.java.advocacy,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: How low can they go...?
Date: Mon, 11 Sep 2000 11:26:23 -0700
Reply-To: <[EMAIL PROTECTED]>
Simon Cooke <[EMAIL PROTECTED]> wrote in message
news:ZB2v5.69549$[EMAIL PROTECTED]...
> Many many many developers do -- but not usually in shrink-wrap products
that
> you can buy in stores (exceptions being PDA synch software for the palm).
> Outlook is useful to bind to for *INTRANET* work.
>
> Besides... how do you think that the ILoveYou virus worked? :)
Is that a benefit of programming that way? I can see the promotional slogan
now, "Program using these Microsoft scantioned methods and help Mellisa say
I love you to your customers too.
------------------------------
From: [EMAIL PROTECTED] (Damien)
Crossposted-To: alt.microsoft.sucks,alt.destroy.microsoft
Subject: Re: Enemies of Linux are MS Lovers
Reply-To: [EMAIL PROTECTED]
Date: 11 Sep 2000 18:59:47 GMT
On Mon, 11 Sep 2000 00:15:21 -0700, in alt.destroy.microsoft
[EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
|
| Damien <[EMAIL PROTECTED]> wrote in message
| news:[EMAIL PROTECTED]...
| > On Sun, 10 Sep 2000 20:32:46 -0500, in alt.destroy.microsoft
| > David Sidlinger <[EMAIL PROTECTED]> wrote:
| > | "Damien" <[EMAIL PROTECTED]> wrote in message
| > | news:[EMAIL PROTECTED]...
| > | > On Sun, 10 Sep 2000 13:19:47 -0400, in alt.destroy.microsoft
| > | > Gary Hallock <[EMAIL PROTECTED]> wrote:
| > | > | David Sidlinger wrote:
| > | > |
| > | > | > Unix can be broken just as easily as Windows. It's just
| > | > | > that a lot of unskilled developers write apps for Windows that
| don't
| > | behave.
| > | > | > I can crash a Unix machine with C++ just as easily as I can crash
| > | Windows.
| > | > | >
| > | > |
| > | > | Really? Please explain how a C++ app can crash Unix.
| > | >
| > | > Two rules. It can't run as root, and it has to run under reasonable
| > | ulimits.
| > | >
| > | > Also, I expect you to include the source code so I can run it on a few
| > | > of the machines I have access to.
| >
| > | Just declare a pointer and try to write to it's location in memory w/o
| > | initializing it. Won't work every time, but, eventually, you're going
| to
| > | cause some freaky stuff to go on.
| >
| > No freaky stuff. Just a segfault. Everytime. No cascade failure.
| > No crash. If I had more time I would write the program you described
| > and run it a few thousand times on a Linux and Sun box. If some one
| > would be so kind as to write the C code I'll still do it.
|
| Ok, here is two versions of the C source that was requested. The first one
| performs as requested. There a chance that the unitialized automatic
| variable "addr", just may some times point valid address causing it the exit
| without doing causing a segment fault. So I am provided the second one as
| well will do so every time. I runs in an endless loop so the program will
| not stop until it is killed or causes the segment fault by incrementing the
| address and writing again. Make sure you have the diskspace to hold the
| core dumps that these will generate. Other than that no harm should be done
| by running these. But don't blame me if they find a real bug. ;-) Also,
| don't run these on an OS that does not provide the kind of protection that
| unix has to prevent that damage these can cause.
[snip]
Thanks for the code. Here some interm results.
pointer.cc
--
void main( void )
{
int datum;
int *addr;
while( 1 ) {
*(addr++) = datum;
}
}
--
pointer.sh
--
#!/bin/sh
while (true)
do ./a.out
done
--
I am currently using two Redhat 6.2 machines. A laptop with 266Mhx
Pentium, 80MB RAM with the 2.2.14-5.0 kernel I compiled myself. The
second is a desktop machine, used mostly as a mail/file/web server.
It has a 200Mhz AMD K6 with 96MB of RAM. The laptop is the machine
I'm posting from right now.
Compiled with
gcc pointer.cc
then I ran
./pointer.sh
On both machines top shows pointer.sh using about 15% CPU. Nothing
else has crashed yet. I'll keep it running on the desktop for about
24 hours at least. I had planned on finding a Solaris machine to try
it on, but it seems pretty silly at this point. Maybe I'll try it on
an NT or NT5 machine.
------------------------------
** 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 (and comp.os.linux.advocacy) via:
Internet: [EMAIL PROTECTED]
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
******************************
