Thought this was interesting

Thought this was interesting

[Darren New]

How to really get zero-defect code:
http://www.fastcompany.com/online/06/writestuff.html
-- 
Darren New 
San Diego, CA, USA (PST). Cryptokeys on demand.
   ** http://images.fbrtech.com/dnew/ **

Try our EbolaBurgers...
      So tender they melt in your mouth.

Re: Thought this was interesting

[Robert Dewar]

Darren New <dnew@san.rr.com> wrote in message news:<3D6688E1.3A1797CC@san.rr.com>...
> How to really get zero-defect code:
> http://www.fastcompany.com/online/06/writestuff.html

Well this article is a bit full of hype for my taste.

If anyone is interested in understanding how software that must be defect
free (e.g. safety-critical software) is written. I recommend reading the
HRG report.

By the way the shuttle software is not the greatest example here. The launch
of the very first shuttle was delayed because of a software bug :-)

Re: Thought this was interesting

[Jeffrey Carter]

Robert Dewar wrote:
> 
> Darren New <dnew@san.rr.com> wrote in message news:<3D6688E1.3A1797CC@san.rr.com>...
> > How to really get zero-defect code:
> > http://www.fastcompany.com/online/06/writestuff.html
> 
> Well this article is a bit full of hype for my taste.

What is really sad and scary is that the article is 6 years old, and we
all know how much the general state of software development has improved
since then. Such as my most recent contract, for a supposedly CMM Level
3 project, on which I was told, in effect, "You start coding and I'll go
find out what they want."

-- 
Jeff Carter
"Monsieur Arthur King, who has the brain of a duck, you know."
Monty Python & the Holy Grail

Re: Thought this was interesting

[Larry Kilgallen]

In article <5ee5b646.0208241501.71259703@posting.google.com>, dewar@gnat.com (Robert Dewar) writes:
> Darren New <dnew@san.rr.com> wrote in message news:<3D6688E1.3A1797CC@san.rr.com>...
>> How to really get zero-defect code:
>> http://www.fastcompany.com/online/06/writestuff.html
> 
> Well this article is a bit full of hype for my taste.
> 
> If anyone is interested in understanding how software that must be defect
> free (e.g. safety-critical software) is written. I recommend reading the
> HRG report.
> 
> By the way the shuttle software is not the greatest example here. The launch
> of the very first shuttle was delayed because of a software bug :-)

I think proof of the difficulty of eliminating defects is useful :-)

Re: Thought this was interesting

[tmoran]

I'm curious how you found that article.  It's six years old and I'd
never seen it mentioned anywhere.  The idea that software doesn't
have to be junk still seems foreign to most.

Re: Thought this was interesting (They Write the Right Stuff)

[Larry Kilgallen]

In article <Z8Y99.5719$OD5.514319399@newssvr13.news.prodigy.com>, tmoran@acm.org writes:
> I'm curious how you found that article.  It's six years old and I'd
> never seen it mentioned anywhere.  The idea that software doesn't
> have to be junk still seems foreign to most.

I have been referring people to this article for years.  I believe I
first read about it in the comp.lang.ada news.

Re: Thought this was interesting

[Darren New]

tmoran@acm.org wrote:
> I'm curious how you found that article.

Several indirect links from http://denbeste.nu/

-- 
Darren New 
San Diego, CA, USA (PST). Cryptokeys on demand.
   ** http://images.fbrtech.com/dnew/ **

Try our EbolaBurgers...
      So tender they melt in your mouth.

Re: Thought this was interesting

[Richard Riehle]

The really interesting question about this article is,
"How much of the code is written in Ada, if any?"

Richard Riehle

---------------------------------------------------------------------
Jeffrey Carter wrote:

> Robert Dewar wrote:
> >
> > Darren New <dnew@san.rr.com> wrote in message news:<3D6688E1.3A1797CC@san.rr.com>...
> > > How to really get zero-defect code:
> > > http://www.fastcompany.com/online/06/writestuff.html
> >
> > Well this article is a bit full of hype for my taste.
>
> What is really sad and scary is that the article is 6 years old, and we
> all know how much the general state of software development has improved
> since then. Such as my most recent contract, for a supposedly CMM Level
> 3 project, on which I was told, in effect, "You start coding and I'll go
> find out what they want."
>
> --
> Jeff Carter
> "Monsieur Arthur King, who has the brain of a duck, you know."
> Monty Python & the Holy Grail

Re: Thought this was interesting

[Darren New]

Richard Riehle wrote:
> The really interesting question about this article is,
> "How much of the code is written in Ada, if any?"

Last I heard, it was all FORTH. But then, last I heard, they were still on
core memory, too. 

-- 
Darren New 
San Diego, CA, USA (PST). Cryptokeys on demand.
   ** http://images.fbrtech.com/dnew/ **

Try our EbolaBurgers...
      So tender they melt in your mouth.

Re: Thought this was interesting

[Pat Rogers]

"Darren New" <dnew@san.rr.com> wrote in message
news:3D697F0E.7D4C1E8E@san.rr.com...
> Richard Riehle wrote:
> > The really interesting question about this article is,
> > "How much of the code is written in Ada, if any?"
>
> Last I heard, it was all FORTH. But then, last I heard, they were
still on
> core memory, too.

No, on-board shuttle flight software is mostly written in a very fine
language named HAL/S.  I wrote a few of the "HAL/S versus Ada" studies
for NASA (concerning the space station) in the early 80's.   Ada won,
but no matter who wrote them they always started with "Well, HAL could
do it, but..."

PS: Hands up if you have a copy of Mike Ryer's book.   Bonus points if
you know what HAL stands for. :-)

--
Patrick Rogers                       Consulting and Training in:
http://www.classwide.com          Real-Time/OO Languages
progers@classwide.com               Hard Deadline Schedulability
Analysis
(281)648-3165                                 Software Fault Tolerance

Re: Thought this was interesting

[Florian Weimer]

dewar@gnat.com (Robert Dewar) writes:

> If anyone is interested in understanding how software that must be defect
> free (e.g. safety-critical software) is written. I recommend reading the
> HRG report.

It seems that it has recently been decided that the report will be
made available for free download.  Is it already possible to download
a copy somwhere?

Re: Thought this was interesting

[Florian Weimer]

dewar@gnat.com (Robert Dewar) writes:

> If anyone is interested in understanding how software that must be defect
> free (e.g. safety-critical software) is written. I recommend reading the
> HRG report.

It seems that it has recently been decided that the report will be
made available for free download.  Is it already possible to download
a copy somewhere?

Re: Thought this was interesting

[Marin D. Condic]

I've seen and been involved in building software that needs to be "perfect".
(Not necessarily bug free, but without any bugs that would cause it to fail
to accomplish its mission.) The biggest problems with this kind of software
are time and money. You have to take the time to do it right and you have to
spend the money needed to be sure its right. Most software is driven by time
to market and low-cost provider issues. It operates against the
"Good/Fast/Cheap - pick two!" problem and "Good" is the loser. So far, most
markets just don't seem to want to demand better. At least not strong enough
to convince developers to shift their emphasis.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com

Enabling Digital.
Our Vision is to be the biggest supplier worldwide of digital gateway
technology.
www.pacemicro.com
<tmoran@acm.org> wrote in message
news:Z8Y99.5719$OD5.514319399@newssvr13.news.prodigy.com...
> I'm curious how you found that article.  It's six years old and I'd
> never seen it mentioned anywhere.  The idea that software doesn't
> have to be junk still seems foreign to most.

Re: Thought this was interesting

[Ted Dennison]

Jeffrey Carter <jrcarter@acm.org> wrote in message news:<3D682CFF.21CE8F23@acm.org>...
> What is really sad and scary is that the article is 6 years old, and we

As a matter of fact, I think I posted a link to it here 2 or 3 years ago...

Re: Thought this was interesting

[Ted Dennison]

tmoran@acm.org wrote in message news:<Z8Y99.5719$OD5.514319399@newssvr13.news.prodigy.com>...
> I'm curious how you found that article.  It's six years old and I'd
> never seen it mentioned anywhere.  The idea that software doesn't
> have to be junk still seems foreign to most.

Well, lets see...

There was a story on slashdot
http://slashdot.org/article.pl?sid=00/05/19/050258&mode=thread&tid=156
quite a while back.

The link was also posted here twice in 2000:
http://groups.google.com/groups?q=fastcompany+group:comp.lang.ada&hl=xx-bork&lr=&ie=UTF-8&oe=UTF-8&selm=3984AD1D.830B538%40below.for.email.address&rnum=2
and
http://groups.google.com/groups?q=fastcompany+group:comp.lang.ada&hl=xx-bork&lr=&ie=UTF-8&oe=UTF-8&selm=KASw4.1558%24dw3.93808%40news.wenet.net&rnum=3

Re: Thought this was interesting

[Jerry Petrey]

Pat Rogers wrote:

>
> PS: Hands up if you have a copy of Mike Ryer's book.   Bonus points if
> you know what HAL stands for. :-)
>
>

I believe it was named after Hal Laning, the author of MAC which it was
modeled after.
It was then named Hal/S for the shuttle version.

Jerry
--
---------------------------------------------------------------------------------

-- Jerry Petrey
-- Senior Principal Systems Engineer - Navigation (GPS/INS), Guidance, &
Control
-- Raytheon Missile Systems          - Member Team Ada & Team Forth
-- NOTE: please remove <NOSPAM> in email address to reply
---------------------------------------------------------------------------------

Re: Thought this was interesting

[Pat Rogers]

"Jerry Petrey @raytheon.com>" <"jdpetrey<NOSPAM> wrote in message
news:3D6A50DF.C9A56917@raytheon.com...
> > PS: Hands up if you have a copy of Mike Ryer's book.   Bonus
points if
> > you know what HAL stands for. :-)
>
> I believe it was named after Hal Laning, the author of MAC which it
was
> modeled after.

Maybe.  I was told a couple of definitions, including "High-level
Assembly Language", but never with a straight face.

> It was then named Hal/S for the shuttle version.

Right.

Re: Thought this was interesting

[Robert A Duff]

Jerry Petrey <"jdpetrey<NOSPAM>"@raytheon.com> writes:

> I believe it was named after Hal Laning, the author of MAC which it was
> modeled after.
> It was then named Hal/S for the shuttle version.

Interesting.  I always thought it showed a bit of sense of humor -- to
name a programming language for a spacecraft after the rogue computer in
"2001, A Space Odyssey".

- Bob

Re: Thought this was interesting

[Jeffrey Carter]

Ted Dennison wrote:
> 
> Jeffrey Carter <jrcarter@acm.org> wrote in message news:<3D682CFF.21CE8F23@acm.org>...
> > What is really sad and scary is that the article is 6 years old, and we
> 
> As a matter of fact, I think I posted a link to it here 2 or 3 years ago...

Quite possibly. I know I saw a similar article a few years ago.

-- 
Jeff Carter
"You couldn't catch clap in a brothel, silly English K...niggets."
Monty Python & the Holy Grail

Re: Thought this was interesting

[Jerry Petrey]

Robert A Duff wrote:

> Jerry Petrey <"jdpetrey<NOSPAM>"@raytheon.com> writes:
>
> > I believe it was named after Hal Laning, the author of MAC which it was
> > modeled after.
> > It was then named Hal/S for the shuttle version.
>
> Interesting.  I always thought it showed a bit of sense of humor -- to
> name a programming language for a spacecraft after the rogue computer in
> "2001, A Space Odyssey".
>
> - Bob

Yeah, I would like to believe that also since 2001 is my favorite of all
movies but,
alas they didn't have that much of a sense of humor.  It was also rumored that

Clark chose HAL by taking the previous letters in the IBM acronym but he has
said in interviews that he didn't do that - he just picked it at random.

Jerry

P.S.
for a little history of the language Hal/S and other such stuff, see:
http://www.unt.edu/UNT/departments/CC/Benchmarks/benchmarks_html/sepoct95/lunar.htm

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

-- Jerry Petrey
-- Senior Principal Systems Engineer - Navigation (GPS/INS), Guidance, &
Control
-- Raytheon Missile Systems          - Member Team Ada & Team Forth
-- NOTE: please remove <NOSPAM> in email address to reply
---------------------------------------------------------------------------------

Re: Thought this was interesting

[Hyman Rosen]

Marin D. Condic wrote:
 > Most software is driven by time to market and low-cost provider issues.
 > It operates against the "Good/Fast/Cheap - pick two!" problem and "Good"
 > is the loser.

There is also the problem of having to develop against
unstable underpinnings. Many programs run in cooperation
with system software which is itself underspecified,
unstable, and buggy. In such a world, even when you do
the best you can, it may still not be enough to produce
"good" software.

Apropos of this, if you have access to a Windows system,
try running the following simple program there. Make sure
to save all important work first! It just prints out a
string over and over again. It couldn't do anything bad,
could it? Especially not on WinNT or WIn2K, right?
Especially if you have no special privileges, right? :-)

with text_io; use text_io;
with ada.characters.latin_1; use ada.characters.latin_1;
procedure crash is
begin
   loop
     put("Hung up!" & HT & BS & BS & BS & BS & BS & BS);
   end loop;
end;

Re: Thought this was interesting

[tmoran]

The lack of public instances, and the language of EULAs, suggest that
you can't sue a vendor for buggy software.  In the rest of the world,
when people can't use the courts to resolve problems (eg, vendors of
illegal products) they develop other methods of conflict resolution.
Perhaps some mafia folks whose bookmaking operation lost money due
to a bug, will make the vendor an offer he can't refuse.

Re: Thought this was interesting (OT)

[Chad R. Meiners]

"Hyman Rosen" <hyrosen@mail.com> wrote in message
news:1030464758.505647@master.nyc.kbcfp.com...
> Apropos of this, if you have access to a Windows system,
> try running the following simple program there. Make sure
> to save all important work first! It just prints out a
> string over and over again. It couldn't do anything bad,
> could it? Especially not on WinNT or WIn2K, right?
> Especially if you have no special privileges, right? :-)
>
> with text_io; use text_io;
> with ada.characters.latin_1; use ada.characters.latin_1;
> procedure crash is
> begin
>    loop
>      put("Hung up!" & HT & BS & BS & BS & BS & BS & BS);
>    end loop;
> end;
>

hmm...   All I get is some garbage followed by a

raised ADA.IO_EXCEPTIONS.DEVICE_ERROR : s-fileio.adb:987

which although is a rather strange error, it doesn't crash my Win2K system.
:)

-CRM

Re: Thought this was interesting (OT)

[Frank J. Lhota]

My tests indicate that Mr. Rosen's program does seem to crash NT 4.0. I am
glad that MS fixed this in Win2K.

Re: Thought this was interesting (OT)

[Robert A Duff]

"Frank J. Lhota" <NOSPAM.lhota.adarose@verizon.net> writes:

> My tests indicate that Mr. Rosen's program does seem to crash NT 4.0.

With what symptom?  Any idea why?  (Just curious.)

>... I am
> glad that MS fixed this in Win2K.

- Bob

Re: Thought this was interesting (OT)

[Hyman Rosen]

Frank J. Lhota wrote:
> My tests indicate that Mr. Rosen's program does seem to crash NT 4.0. I am
> glad that MS fixed this in Win2K.

Really? I'm given to understand that this problem still
exists even in Windows XP. I haven't tried it there yet.

Re: Thought this was interesting (OT)

[Jeffrey Carter]

Hyman Rosen wrote:
> 
> Really? I'm given to understand that this problem still
> exists even in Windows XP. I haven't tried it there yet.

I had no trouble with the program on Win98 with GNAT 3.14p. I could
terminate the program with CTRL-C or by closing the DOS box.

-- 
Jeff Carter
"If you think you got a nasty taunting this time,
you ain't heard nothing yet!"
Monty Python and the Holy Grail

Re: Thought this was interesting (OT)

[SteveD]

"Frank J. Lhota" <NOSPAM.lhota.adarose@verizon.net> wrote in message
news:nbRa9.29920$3V5.25137@nwrddc02.gnilink.net...
> My tests indicate that Mr. Rosen's program does seem to crash NT 4.0. I am
> glad that MS fixed this in Win2K.
>
>

Not so fast.  I just tried this on my Win2K box and the next thing I saw was
the Compaq logo (reset).

I get the same result with ObjectAda 7.2 and GNAT 3.14p, but interestingly I
dont get the reset if I try the equivalent program in Microsoft C++.

SteveD

RE: Thought this was interesting

[Robert C. Leif]

From: Bob Leif
To: Tom Moran et al.
Unfortunately, no hungry trial lawyer has really tried this. The
readership of Comp.Lang.Ada would be an excellent source of witnesses
for the plaintiff(s). I doubt if the present EULAs would be considered
by the courts to be a contract negotiated by equals. I believe this is
why the industry wants state statutes to protect them from the
consequences of malpractice. A few large product defect verdicts would
have a very positive effect on software engineering including the use of
Ada.

-----Original Message-----
From: comp.lang.ada-admin@ada.eu.org
[mailto:comp.lang.ada-admin@ada.eu.org] On Behalf Of tmoran@acm.org
Sent: Tuesday, August 27, 2002 11:49 AM
To: comp.lang.ada@ada.eu.org
Subject: Re: Thought this was interesting

The lack of public instances, and the language of EULAs, suggest that
you can't sue a vendor for buggy software.  In the rest of the world,
when people can't use the courts to resolve problems (eg, vendors of
illegal products) they develop other methods of conflict resolution.
Perhaps some mafia folks whose bookmaking operation lost money due
to a bug, will make the vendor an offer he can't refuse.

Re: Thought this was interesting (OT)

[Pat Rogers]

"Hyman Rosen" <hyrosen@mail.com> wrote in message
news:1030486974.259451@master.nyc.kbcfp.com...
> Really? I'm given to understand that this problem still
> exists even in Windows XP. I haven't tried it there yet.

It does -- a hard reset of XP Home.

How?????????

Re: Thought this was interesting

[Ole-Hjalmar Kristensen]

Hyman Rosen <hyrosen@mail.com> writes:

<snip>

> Apropos of this, if you have access to a Windows system,
> try running the following simple program there. Make sure
> to save all important work first! It just prints out a
> string over and over again. It couldn't do anything bad,
> could it? Especially not on WinNT or WIn2K, right?
> Especially if you have no special privileges, right? :-)
> 
> with text_io; use text_io;
> with ada.characters.latin_1; use ada.characters.latin_1;
> procedure crash is
> begin
>    loop
>      put("Hung up!" & HT & BS & BS & BS & BS & BS & BS);
>    end loop;
> end;

Bizarre behaviour indeed. What puzzles me is that I attempted briefly
to write an equivalent C program, and it did not give the same
results. Any idea of what's so special about text_io?

Re: Thought this was interesting (OT)

[Ole-Hjalmar Kristensen]

Hyman Rosen <hyrosen@mail.com> writes:

> Frank J. Lhota wrote:
> > My tests indicate that Mr. Rosen's program does seem to crash NT 4.0. I am
> > glad that MS fixed this in Win2K.
> 
> Really? I'm given to understand that this problem still
> exists even in Windows XP. I haven't tried it there yet.

It crashes my Windows 2000, but then I manage to crash it myself a
couple of times a week without really trying...

Re: Thought this was interesting (OT)

[AG]

"Robert A Duff" <bobduff@shell01.TheWorld.com> wrote in message
news:wccznv8ynva.fsf@shell01.TheWorld.com...
> "Frank J. Lhota" <NOSPAM.lhota.adarose@verizon.net> writes:
>
> > My tests indicate that Mr. Rosen's program does seem to crash NT 4.0.
>
> With what symptom?  Any idea why?  (Just curious.)

Just tried it on W2Kpro (Athlon 1800+, 1Gb of memory
just in case it matters). Crashed it cold (Using latest free
Gnat version).

Symptoms: The DOS window displays a short string of text.
If you are really quick you can move the mouse around for
a fraction of a second but that's all. Than the whole thing
freezes and seconds later does a soft reboot.

>
> >... I am
> > glad that MS fixed this in Win2K.

Hmmm ...

Re: Thought this was interesting (OT)

[AG]

"AG" <ang@xtra.co.nz> wrote in message
news:lC0b9.432$Y3.49485@news.xtra.co.nz...
>
> "Robert A Duff" <bobduff@shell01.TheWorld.com> wrote in message
> news:wccznv8ynva.fsf@shell01.TheWorld.com...
> > "Frank J. Lhota" <NOSPAM.lhota.adarose@verizon.net> writes:
> >
> > > My tests indicate that Mr. Rosen's program does seem to crash NT 4.0.
> >
> > With what symptom?  Any idea why?  (Just curious.)
>
> Just tried it on W2Kpro (Athlon 1800+, 1Gb of memory
> just in case it matters). Crashed it cold (Using latest free
> Gnat version).

Having read the other postings to the group it seems like
it's a wide-spread phenomena. Now I'm really interested as
to why?

After all, someone did come up with that construct, any
comment about how come and where does it come from?

Re: Thought this was interesting

[AG]

"Ole-Hjalmar Kristensen" <oleh@vlinux.voxelvision.no> wrote in message
news:7vu1lfbejt.fsf@vlinux.voxelvision.no...
> Hyman Rosen <hyrosen@mail.com> writes:
>
> <snip>
>
> > Apropos of this, if you have access to a Windows system,
> > try running the following simple program there. Make sure
> > to save all important work first! It just prints out a
> > string over and over again. It couldn't do anything bad,
> > could it? Especially not on WinNT or WIn2K, right?
> > Especially if you have no special privileges, right? :-)
> >
> > with text_io; use text_io;
> > with ada.characters.latin_1; use ada.characters.latin_1;
> > procedure crash is
> > begin
> >    loop
> >      put("Hung up!" & HT & BS & BS & BS & BS & BS & BS);
> >    end loop;
> > end;
>
> Bizarre behaviour indeed. What puzzles me is that I attempted briefly
> to write an equivalent C program, and it did not give the same
> results. Any idea of what's so special about text_io?
>

My guess would be [guess - mind you] that it relies on some API
function which is supposed to handle it but does not. Well, of course,
it was not written in our favorite language :-)

Re: Thought this was interesting (OT)

[Robert Dewar]

"SteveD" <nospam_steved94@attbi.com> wrote in message news:<XVWa9.204365$983.396527@rwcrnsc53>...

> I get the same result with ObjectAda 7.2 and GNAT 3.14p, but interestingly I
> dont get the reset if I try the equivalent program in Microsoft C++.

Then it is not the equivalent program! You can always write a C program that
does exactly the same thing as any GNAT program, if you try and do not get
the same effect, you simply did not write an equivalent program, but rather
a similar but not identical program. Since this is a pretty bizarre bug in
the operating system, it is not so surprising that a slight change would
make it go away.

Re: Thought this was interesting (OT)

[Gautier]

Jeffrey Carter:

> I had no trouble with the program on Win98 with GNAT 3.14p. I could
> terminate the program with CTRL-C or by closing the DOS box.

On my Win98 machine too, there is no problem. But on the other hand
you can hang it only with Notepad - no need of an Ada compiler!

1) Save the following line under "crash.com"

h01X-00P_jXX)E0)E1)E25c_5r_5a_5s_5h_5__5_m5_e5_!RDW

2) Close the windows, save your hard work, wait for the delayed
   cache write.

3) Run "crash.com".

Funny, isn't it ?
________________________________________________________
Gautier  --  http://www.mysunrise.ch/users/gdm/gsoft.htm

NB: For a direct answer, address on the Web site!

Re: Thought this was interesting (OT)

[Frank J. Lhota]

I got the infamous BSOD.

Re: Thought this was interesting (OT)

[Dan Andreatta]

"AG" <ang@xtra.co.nz> wrote in message news:<_L0b9.434$Y3.52721@news.xtra.co.nz>...
> "AG" <ang@xtra.co.nz> wrote in message
> news:lC0b9.432$Y3.49485@news.xtra.co.nz...
> >
> > "Robert A Duff" <bobduff@shell01.TheWorld.com> wrote in message
> > news:wccznv8ynva.fsf@shell01.TheWorld.com...
> > > "Frank J. Lhota" <NOSPAM.lhota.adarose@verizon.net> writes:
> > >
> > > > My tests indicate that Mr. Rosen's program does seem to crash NT 4.0.
> > >
> > > With what symptom?  Any idea why?  (Just curious.)
> >
> > Just tried it on W2Kpro (Athlon 1800+, 1Gb of memory
> > just in case it matters). Crashed it cold (Using latest free
> > Gnat version).
> 
> Having read the other postings to the group it seems like
> it's a wide-spread phenomena. Now I'm really interested as
> to why?
> 
> After all, someone did come up with that construct, any
> comment about how come and where does it come from?

I do not know the details, but it was (is?) a well known bug in the
console subsystem of NT (or at least that is what I read about it).

Dan

Re: Thought this was interesting (OT)

[Dan Andreatta]

"AG" <ang@xtra.co.nz> wrote in message news:<_L0b9.434$Y3.52721@news.xtra.co.nz>...
> "AG" <ang@xtra.co.nz> wrote in message
> news:lC0b9.432$Y3.49485@news.xtra.co.nz...
> >
> > "Robert A Duff" <bobduff@shell01.TheWorld.com> wrote in message
> > news:wccznv8ynva.fsf@shell01.TheWorld.com...
> > > "Frank J. Lhota" <NOSPAM.lhota.adarose@verizon.net> writes:
> > >
> > > > My tests indicate that Mr. Rosen's program does seem to crash NT 4.0.
> > >
> > > With what symptom?  Any idea why?  (Just curious.)
> >
> > Just tried it on W2Kpro (Athlon 1800+, 1Gb of memory
> > just in case it matters). Crashed it cold (Using latest free
> > Gnat version).
> 
> Having read the other postings to the group it seems like
> it's a wide-spread phenomena. Now I'm really interested as
> to why?
> 

Found the reference

http://homepages.tesco.net/~J.deBoynePollard/FGA/csrss-backspace-bug.html

Dan
> After all, someone did come up with that construct, any
> comment about how come and where does it come from?

Re: Thought this was interesting (OT)

[Chad R. Meiners]

Just to note.  I have service pack 3 installed for win2k.

-CRM

Re: Thought this was interesting

[Jeffrey Creem]

"Ole-Hjalmar Kristensen" <oleh@vlinux.voxelvision.no> wrote in message
news:7vu1lfbejt.fsf@vlinux.voxelvision.no...
> Hyman Rosen <hyrosen@mail.com> writes:
> Bizarre behaviour indeed. What puzzles me is that I attempted briefly
> to write an equivalent C program, and it did not give the same
> results. Any idea of what's so special about text_io?
>

Hmm that is strange..The first time I saw this bug it was actually sent as a
3 or
4 line C program and I translated it to Ada (to prove that you can do
anything
in Ada that you do in C :).

In any case, rest assured that a very similar C program produces the same
results.

Re: Thought this was interesting

[SteveD]

"Jeffrey Creem" <jeff@thecreems.com> wrote in message
news:dIeb9.113450$aA.24929@sccrnsc02...
[snip]
>
> In any case, rest assured that a very similar C program produces the same
> results.
>

This one produces the same result when compiled with Visual C++ 6.0 and run
on W2k.  Using streaming operations on cout does not cause the problem.
Using puts instead of printf does not cause the problem.

#include <cstdio>

int main()
{
 for( ; ; )
 {
  printf( "Hung up!\t\b\b\b\b\b\b" );
 }
 return 0;
}

Re: Thought this was interesting

[Ole-Hjalmar Kristensen]

"SteveD" <nospam_steved94@attbi.com> writes:

> "Jeffrey Creem" <jeff@thecreems.com> wrote in message
> news:dIeb9.113450$aA.24929@sccrnsc02...
> [snip]
> >
> > In any case, rest assured that a very similar C program produces the same
> > results.
> >
> 
> This one produces the same result when compiled with Visual C++ 6.0 and run
> on W2k.  Using streaming operations on cout does not cause the problem.
> Using puts instead of printf does not cause the problem.
> 
> #include <cstdio>
> 
> int main()
> {
>  for( ; ; )
>  {
>   printf( "Hung up!\t\b\b\b\b\b\b" );
>  }
>  return 0;
> }

That may explain my problem in reproducing it, as I used puts.

Re: Thought this was interesting (Bounds Checking!)

[Warren W. Gay VE3WWG]

Dan Andreatta wrote:
...
> Found the reference
> 
> http://homepages.tesco.net/~J.deBoynePollard/FGA/csrss-backspace-bug.html

I find it amusing that this page states

<quote>

This bug is a combination of

     * A flaw in the design of Windows NT, and
     * A bounds-checking error in the code that handles high-level console I/O.

</quote>

Now, where are all of those non-Ada programmers that think
that bounds checking is unimportant? ;-)

-- 
Warren W. Gay VE3WWG
http://home.cogeco.ca/~ve3wwg

Re: Thought this was interesting (Bounds Checking!)

[Ryan]

Larry Kilgallen wrote:
> I don't see what is important about bounds checking in this case.
> It is only Windows that crashed :-)

Microsoft made two very bad decisions in writing Windows NT.  The more 
obvious was forcing the whole OS to go down when only the console I/O 
code fails.  The less obvious was writing that code in a language other 
than Ada :-)

Ryan

Re: Thought this was interesting (Bounds Checking!)

[Larry Kilgallen]

In article <3D6E561A.5080307@cogeco.ca>, "Warren W. Gay VE3WWG" <ve3wwg@cogeco.ca> writes:

> Now, where are all of those non-Ada programmers that think
> that bounds checking is unimportant? ;-)

I don't see what is important about bounds checking in this case.
It is only Windows that crashed :-)

"HAL" acronym (was Re: Thought this was interesting)

[Ben Brosgol]

> > > PS: Hands up if you have a copy of Mike Ryer's book.   Bonus
> points if
> > > you know what HAL stands for. :-)
> >
> > I believe it was named after Hal Laning, the author of MAC which it
> was
> > modeled after.
>
> Maybe.  I was told a couple of definitions, including "High-level
> Assembly Language", but never with a straight face.

When I was at Intermetrics in the '70s I recall hearing two explanations:
1) Named after Hal Laning, as was previously noted
2) Houston Algorithmic Language

Ben Brosgol
Ada Core Technologies

Re: "HAL" acronym (was Re: Thought this was interesting)

[Manny Apalisok]

Hi,

I read somewhere that HAL is a pun on the word IBM. Shift each letter
one position and you get IBM.

Manny Apalisok

Re: "HAL" acronym (was Re: Thought this was interesting)

[Frank J. Lhota]

"Manny Apalisok" <manuel@hkpc.org> wrote in message
news:caa2756a.0209022139.3d87dc7e@posting.google.com...
> I read somewhere that HAL is a pun on the word IBM. Shift each letter
> one position and you get IBM.

That was a popular story about why "HAL" was the name of the psychotic
computer in the film "2001: A Space Odyssey". Arthur C. Clark, however,
denies this.

Re: Thought this was interesting

[Hyman Rosen]

Ole-Hjalmar Kristensen wrote:
> That may explain my problem in reproducing it, as I used puts.

That's because puts appends a newline to the output after
printing its string. That makes the output of the program
much different, and doesn't trigger the bug.