* A scary story from the real world.
@ 2007-11-09 17:33 Per Sandberg
2007-11-09 20:46 ` Jerry
` (2 more replies)
0 siblings, 3 replies; 16+ messages in thread
From: Per Sandberg @ 2007-11-09 17:33 UTC (permalink / raw)
I was on a conference this week and that opened my eyes on the complete
ignorance of good compilers and languages in some the Reliable SW
communities.
http://www.issre2007.hv.se/extra/pod/
There was lots of talks about on how to detect values out of bounds and
other problems that reasonable languages with type-system would find
probably at compile time.
One Microsoft person said "we had lots of crashes in the system and the
cause of that was that the driver vendors did not look on the return
code from functions"
Then the blamed the poor programmer for not reading the secret "users
manual".
/Per
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: A scary story from the real world.
2007-11-09 17:33 A scary story from the real world Per Sandberg
@ 2007-11-09 20:46 ` Jerry
2007-11-09 21:15 ` Per Sandberg
2007-11-09 22:09 ` anon
2007-11-10 0:06 ` Brian May
2 siblings, 1 reply; 16+ messages in thread
From: Jerry @ 2007-11-09 20:46 UTC (permalink / raw)
On Nov 9, 10:33 am, Per Sandberg <per.sandb...@bredband.net> wrote:
> I was on a conference this week and that opened my eyes on the complete
> ignorance of good compilers and languages in some the Reliable SW
> communities.
> http://www.issre2007.hv.se/extra/pod/
> There was lots of talks about on how to detect values out of bounds and
> other problems that reasonable languages with type-system would find
> probably at compile time.
>
> One Microsoft person said "we had lots of crashes in the system and the
> cause of that was that the driver vendors did not look on the return
> code from functions"
> Then the blamed the poor programmer for not reading the secret "users
> manual".
> /Per
Was Ada even mentioned?
Jerry
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: A scary story from the real world.
2007-11-09 20:46 ` Jerry
@ 2007-11-09 21:15 ` Per Sandberg
2007-11-09 21:25 ` Larry Kilgallen
` (2 more replies)
0 siblings, 3 replies; 16+ messages in thread
From: Per Sandberg @ 2007-11-09 21:15 UTC (permalink / raw)
Jerry wrote:
> On Nov 9, 10:33 am, Per Sandberg <per.sandb...@bredband.net> wrote:
>> I was on a conference this week and that opened my eyes on the complete
>> ignorance of good compilers and languages in some the Reliable SW
>> communities.
>> http://www.issre2007.hv.se/extra/pod/
>> There was lots of talks about on how to detect values out of bounds and
>> other problems that reasonable languages with type-system would find
>> probably at compile time.
>>
>> One Microsoft person said "we had lots of crashes in the system and the
>> cause of that was that the driver vendors did not look on the return
>> code from functions"
>> Then the blamed the poor programmer for not reading the secret "users
>> manual".
>> /Per
>
> Was Ada even mentioned?
> Jerry
>
>
No !
And that was what i found scary.
From my perspective the whole conference was about how to we make the
best out after we have crashed instead of how do we avoid to crash.
I mentioned Ada in connection to the Microsoft talk and the comment from
the Microsoft person was something along the lines
Ada is never going to make it and I and don't believe in it.
I wont quite since i don't remember the exact worthing.
/Per
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: A scary story from the real world.
2007-11-09 21:15 ` Per Sandberg
@ 2007-11-09 21:25 ` Larry Kilgallen
2007-11-10 23:03 ` Jerry
2007-11-11 14:24 ` Brian Drummond
2 siblings, 0 replies; 16+ messages in thread
From: Larry Kilgallen @ 2007-11-09 21:25 UTC (permalink / raw)
In article <K34Zi.15908$CT3.3476@newsfet01.ams>, Per Sandberg <per.sandberg@bredband.net> writes:
> Jerry wrote:
>> On Nov 9, 10:33 am, Per Sandberg <per.sandb...@bredband.net> wrote:
>>> I was on a conference this week and that opened my eyes on the complete
>>> ignorance of good compilers and languages in some the Reliable SW
>>> communities.
>>> http://www.issre2007.hv.se/extra/pod/
>>> There was lots of talks about on how to detect values out of bounds and
>>> other problems that reasonable languages with type-system would find
>>> probably at compile time.
>>>
>>> One Microsoft person said "we had lots of crashes in the system and the
>>> cause of that was that the driver vendors did not look on the return
>>> code from functions"
>>> Then the blamed the poor programmer for not reading the secret "users
>>> manual".
>>> /Per
>>
>> Was Ada even mentioned?
>> Jerry
>>
>>
>
> No !
> And that was what i found scary.
> From my perspective the whole conference was about how to we make the
> best out after we have crashed instead of how do we avoid to crash.
That seems to be the attitude on the Secure Coding mailing list:
http://www.securecoding.org/list/charter.php
http://krvw.com/mailman/listinfo/sc-l
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: A scary story from the real world.
2007-11-09 17:33 A scary story from the real world Per Sandberg
2007-11-09 20:46 ` Jerry
@ 2007-11-09 22:09 ` anon
2007-11-10 0:06 ` Brian May
2 siblings, 0 replies; 16+ messages in thread
From: anon @ 2007-11-09 22:09 UTC (permalink / raw)
Microsoft may have a lot of weigh but they are more wrong in what
they say then they are right! And they are great at passing the
blame. What they forget to say is that from the 1980s they have
enforce bad programming techniques. An example is the use of Basic
and allowing one line programs even though that one line was
multi-statement taking 16k to 64k in length.
Another is their views on multicore and parallel processing.
I was talking about multicore and parallel processing back in September
I came across an article about Microsoft and multicore processors.
The article was something like Microsoft saying back in the 1980s
that we would never need more than 640K for a ram size.
Title: Microsoft sees shift to parallel in 10yrs
Posted: 03 Sep 2007
Multicore processors are driving a historic shift to a new parallel
architecture for mainstream computers. But a parallel programming
model to serve those machines will not emerge for five to 10 years,
according to experts from Microsoft Corp.
...
by Rick Merritt
EE Times
In talking to another person they said:
> This is kinda funny, mostly because some serious multi-core machines
> are available today. And with AMD entering the race with a quad
> Opteron, and 8-core systems from Intel on the immediate horizon, I
> expect to see prices for 4/8-way systems fall to a reasonable range
> within a year or two. Now, granted, my definition of "reasonable" may
> be slightly higher than some people's, but still.. they are coming,
> and they'll be here before the decade mark for sure.
And it is funny because there are basically four groups that are
powering parallel processing. Group 0, or the original parallel group
which is the scientific community. With government and tax payers
paying the bill they have the money for multicore and parallel
processing power, now.
Then there is Group 1 which is business. Mostly at the movement is
controlled by the internet and internet servers, they not only need
the bandwidth but they need the processing power was well. Even though
there is a small shift away from porn, the internet is still being paid
by the biggest online money maker services aka porn. They have the money
for multicore processing power and the need to use it, now.
Count Movies makers as Group 2. Movie makers want to use special
effects that are sometime dangerous or impossible to do. So, the
count on special effects software to do the movie magic. The more
power they have the more they can do in record time. And with the
movie going public able and welling to pay for that special effects
movie, the movie makers have the cash for that power. And their need
is now.
And Group 3, is the newest group but has deep pockets aka the Gamer,
they want more power and are willing to achieve the power at any cost.
And the Gamer is not going to wait for Microsoft's idea of 10 years.
They want it, Yesterday. And with system like the "Powerstation 3" a
8/7 (1 processor disabled) they are not waiting. The "Powerstation 3"
uses an IBM RISC multicore processor that IBM has stated that they are
coming out with a line of business and server models using that same
multicore processor.
Besides multicore and parallel processing groups. There is the masses
of the world. They are not waiting for Microsoft, they want more power
now, like the functions in an IPHONE or the new IPOD. And these
embedded system are less likely to crash. So, those programmers who
wrote the code have it going on. And some of those system may have
been written in Ada.
So, the power brokers groups and the technology are not waiting for
Microsoft to play once again catch up on technology. They are slowly
moving away from listening to a Dinosaur called Microsoft.
May Be Microsoft should do a background check of the programmers they
are blaming. They probability would find these programmer have been
mind wipe to believe only in Microsoft dogma. With no true vision
of the current status or the future of the computer world.
But the scary part is that Microsoft may have to play hard ball aka
the software patent or licenses just to stay in the game. Oops, sorry,
they are already playing the licensing game, once again.
Plus, to Microsoft "Ada" is permanently dead! And C/C++, JAVA are on
their death bed as well. Which leaves "Visual Basic" and the .net for the
future of programming.
In <0Q0Zi.409$CT3.318@newsfet01.ams>, Per Sandberg <per.sandberg@bredband.net> writes:
>I was on a conference this week and that opened my eyes on the complete
>ignorance of good compilers and languages in some the Reliable SW
>communities.
> http://www.issre2007.hv.se/extra/pod/
>There was lots of talks about on how to detect values out of bounds and
>other problems that reasonable languages with type-system would find
>probably at compile time.
>
>One Microsoft person said "we had lots of crashes in the system and the
>cause of that was that the driver vendors did not look on the return
>code from functions"
>Then the blamed the poor programmer for not reading the secret "users
>manual".
>/Per
>
>
>
>
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: A scary story from the real world.
2007-11-09 17:33 A scary story from the real world Per Sandberg
2007-11-09 20:46 ` Jerry
2007-11-09 22:09 ` anon
@ 2007-11-10 0:06 ` Brian May
2007-11-10 5:44 ` Larry Kilgallen
2007-11-10 21:40 ` Harald Korneliussen
2 siblings, 2 replies; 16+ messages in thread
From: Brian May @ 2007-11-10 0:06 UTC (permalink / raw)
>>>>> "Per" == Per Sandberg <per.sandberg@bredband.net> writes:
Per> One Microsoft person said "we had lots of crashes in the system and
Per> the cause of that was that the driver vendors did not look on the
Per> return code from functions"
Per> Then the blamed the poor programmer for not reading the secret "users
Per> manual".
I think some driver vendors could write buggy code even if a good
language was used. The general attitude in the Windows world is to try
and work around the problem as opposed to finding out why a driver
installation repeatedly crashes at a given point on a given computer.
I see only two solutions:
1. Microsoft write or review drivers themselves. Unlikely to
happen. Even with an open source model like Linux, some drivers that
end up in the kernel are horrible (or so I have heard); there are
simply too many drivers to review and possibly rewrite every one.
2. Move driver to a separate user space process somehow, so only that
one driver crashes instead of the whole computer. Also it should be
immediately obvious which driver crashed, so the complaints go to the
right place. (See "The Hurd" and "Minix" for examples).
--
Brian May <bam@snoopy.apana.org.au>
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: A scary story from the real world.
2007-11-10 0:06 ` Brian May
@ 2007-11-10 5:44 ` Larry Kilgallen
2007-11-10 21:40 ` Harald Korneliussen
1 sibling, 0 replies; 16+ messages in thread
From: Larry Kilgallen @ 2007-11-10 5:44 UTC (permalink / raw)
In article <sa4pryjx8tf.fsf@snoopy.microcomaustralia.com.au>, Brian May <bam@snoopy.apana.org.au> writes:
> I see only two solutions:
>
> 1. Microsoft write or review drivers themselves. Unlikely to
> happen. Even with an open source model like Linux, some drivers that
> end up in the kernel are horrible (or so I have heard); there are
> simply too many drivers to review and possibly rewrite every one.
Some hardware vendors are unwilling to share the details of how their
hardware works with other companies.
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: A scary story from the real world.
2007-11-10 0:06 ` Brian May
2007-11-10 5:44 ` Larry Kilgallen
@ 2007-11-10 21:40 ` Harald Korneliussen
1 sibling, 0 replies; 16+ messages in thread
From: Harald Korneliussen @ 2007-11-10 21:40 UTC (permalink / raw)
On Nov 10, 1:06 am, Brian May <b...@snoopy.apana.org.au> wrote:
> I see only two solutions:
>
> 1. Microsoft write or review drivers themselves. Unlikely to
> happen. Even with an open source model like Linux, some drivers that
> end up in the kernel are horrible (or so I have heard); there are
> simply too many drivers to review and possibly rewrite every one.
They do have a driver signing procedure, now, and some sort of
conformance testing. I don't know the details, but it has obviously
helped a lot compared to what we had before.
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: A scary story from the real world.
2007-11-09 21:15 ` Per Sandberg
2007-11-09 21:25 ` Larry Kilgallen
@ 2007-11-10 23:03 ` Jerry
2007-11-11 0:11 ` tmoran
` (2 more replies)
2007-11-11 14:24 ` Brian Drummond
2 siblings, 3 replies; 16+ messages in thread
From: Jerry @ 2007-11-10 23:03 UTC (permalink / raw)
> I mentioned Ada in connection to the Microsoft talk and the comment from
> the Microsoft person was something along the lines
> Ada is never going to make it and I and don't believe in it.
> I wont quite since i don't remember the exact worthing.
>
> /Per
Assuming your paraphrasing is mostly accurate, the Microsoft response
is _partly_ nonsensical.
"Ada is never going to make it:" All you (Microsoft) have to do is to
start using it. "Making it" is a non-issue.
"...I don't believe in it." Aha--now we're getting somewhere. Language
choice is apparently like religion, based on a belief system. People's
choice of religion is almost always the religion in which they were
born into. It must be the same for programming languages. But religion
is inherently belief-based whereas programmers take pride in being
logical and rational. So why _do_ programmers act as though language
choice is a belief system?
I wish some smart psychologist would do research on the propagation
over decades of crappy languages as the first choice of most
programmers.
Jerry
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: A scary story from the real world.
2007-11-10 23:03 ` Jerry
@ 2007-11-11 0:11 ` tmoran
2007-11-11 8:46 ` Dmitry A. Kazakov
2007-11-12 7:36 ` Harald Korneliussen
2 siblings, 0 replies; 16+ messages in thread
From: tmoran @ 2007-11-11 0:11 UTC (permalink / raw)
> whereas programmers take pride in being logical and rational.
IMHO, homo sapiens does rational thought like flying squirrels fly.
Yes, sort of, some of the time, but not like an eagle or a fly.
There are many things people do better than computers, but logic
isn't one of them.
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: A scary story from the real world.
2007-11-10 23:03 ` Jerry
2007-11-11 0:11 ` tmoran
@ 2007-11-11 8:46 ` Dmitry A. Kazakov
2007-11-12 7:36 ` Harald Korneliussen
2 siblings, 0 replies; 16+ messages in thread
From: Dmitry A. Kazakov @ 2007-11-11 8:46 UTC (permalink / raw)
On Sat, 10 Nov 2007 15:03:13 -0800, Jerry wrote:
> "...I don't believe in it." Aha--now we're getting somewhere. Language
> choice is apparently like religion, based on a belief system. People's
> choice of religion is almost always the religion in which they were
> born into. It must be the same for programming languages. But religion
> is inherently belief-based whereas programmers take pride in being
> logical and rational. So why _do_ programmers act as though language
> choice is a belief system?
Because programming language is a thought system. It is much
self-sufficient. Once one gets into such system he loses most of his
ability to judge rationally about it while staying within it.
> I wish some smart psychologist would do research on the propagation
> over decades of crappy languages as the first choice of most
> programmers.
add here crappy OSes, crappy CPU architectures, crappy software... There is
a natural law of Growing Crappiness. (:-))
--
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: A scary story from the real world.
2007-11-09 21:15 ` Per Sandberg
2007-11-09 21:25 ` Larry Kilgallen
2007-11-10 23:03 ` Jerry
@ 2007-11-11 14:24 ` Brian Drummond
2 siblings, 0 replies; 16+ messages in thread
From: Brian Drummond @ 2007-11-11 14:24 UTC (permalink / raw)
On Fri, 09 Nov 2007 22:15:10 +0100, Per Sandberg
<per.sandberg@bredband.net> wrote:
>Jerry wrote:
>> On Nov 9, 10:33 am, Per Sandberg <per.sandb...@bredband.net> wrote:
>>> I was on a conference this week and that opened my eyes on the complete
>>> ignorance of good compilers and languages in some the Reliable SW
>>> communities.
>>> http://www.issre2007.hv.se/extra/pod/
>> Was Ada even mentioned?
>> Jerry
>>
>>
>
>No !
>And that was what i found scary.
> From my perspective the whole conference was about how to we make the
>best out after we have crashed instead of how do we avoid to crash.
>
Apropos of which, are we supposed to laugh or cry on reading this?
http://www.techreview.com/Infotech/17831/page1/?a=f
"I think the real problem is that "we" (that is, we software developers)
are in a permanent state of emergency, grasping at straws to get our
work done. We perform many minor miracles through trial and error,
excessive use of brute force, and lots and lots of testing, but--so
often--it's not enough.
Software developers have become adept at the difficult art of building
reasonably reliable systems out of unreliable parts. The snag is that
often we do not know exactly how we did it: a system just "sort of
evolved" into something minimally acceptable. Personally, I prefer to
know when a system will work, and why it will."
... considering the source...
- Brian
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: A scary story from the real world.
2007-11-10 23:03 ` Jerry
2007-11-11 0:11 ` tmoran
2007-11-11 8:46 ` Dmitry A. Kazakov
@ 2007-11-12 7:36 ` Harald Korneliussen
2007-11-12 8:27 ` Brian May
2 siblings, 1 reply; 16+ messages in thread
From: Harald Korneliussen @ 2007-11-12 7:36 UTC (permalink / raw)
On Nov 11, 12:03 am, Jerry <lancebo...@qwest.net> wrote:
> "...I don't believe in it." Aha--now we're getting somewhere. Language
> choice is apparently like religion, based on a belief system. People's
> choice of religion is almost always the religion in which they were
> born into. It must be the same for programming languages. But religion
> is inherently belief-based whereas programmers take pride in being
> logical and rational. So why _do_ programmers act as though language
> choice is a belief system?
Why invoke religion? anon below comments that the future for Microsoft
is Visual Basic. While I'm no fan of MS technologies, I know enough
about them to know that this is nonsense: Microsoft's strategy right
now is going for a multiple-lanuage virtual machine runtime, with a
Java-like language as the main development language.
Now, one may argue whether the Java/C# approach of virtual machines,
sandboxes, bytecode verification etc. is better than the Ada approach,
but it is at least an attempt. It's not as if they're doing nothing.
But the fact that anon doesn't even mention this, and instead invokes
the age-old spectre of Visual Basic, tells me that not everyone on
this side of the fence is interested in other approaches, to put it
like that.
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: A scary story from the real world.
2007-11-12 7:36 ` Harald Korneliussen
@ 2007-11-12 8:27 ` Brian May
2007-11-12 13:04 ` Stephen Leake
0 siblings, 1 reply; 16+ messages in thread
From: Brian May @ 2007-11-12 8:27 UTC (permalink / raw)
>>>>> "Harald" == Harald Korneliussen <vintermann@gmail.com> writes:
Harald> Now, one may argue whether the Java/C# approach of virtual machines,
Harald> sandboxes, bytecode verification etc. is better than the Ada approach,
Harald> but it is at least an attempt. It's not as if they're doing nothing.
There is nothing stopping Ada from having virtual machines, sandboxes,
bytecode verification, etc, either. If they really wanted to do so.
Also it is possible to compile Java code direct to machine code (gcj).
The issue of choice of language has nothing to do with if you want or
don't want these technologies.
--
Brian May <bam@snoopy.apana.org.au>
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: A scary story from the real world.
2007-11-12 8:27 ` Brian May
@ 2007-11-12 13:04 ` Stephen Leake
2007-11-12 15:09 ` Georg Bauhaus
0 siblings, 1 reply; 16+ messages in thread
From: Stephen Leake @ 2007-11-12 13:04 UTC (permalink / raw)
Brian May <bam@snoopy.apana.org.au> writes:
>>>>>> "Harald" == Harald Korneliussen <vintermann@gmail.com> writes:
>
> Harald> Now, one may argue whether the Java/C# approach of
> Harald> virtual machines, sandboxes, bytecode verification etc.
> Harald> is better than the Ada approach, but it is at least an
> Harald> attempt. It's not as if they're doing nothing.
>
> There is nothing stopping Ada from having virtual machines, sandboxes,
> bytecode verification, etc, either. If they really wanted to do so.
AdaCore has ported GNAT to .NET. There was even a small ad in the
current Dr Dobbs for it.
--
-- Stephe
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: A scary story from the real world.
2007-11-12 13:04 ` Stephen Leake
@ 2007-11-12 15:09 ` Georg Bauhaus
0 siblings, 0 replies; 16+ messages in thread
From: Georg Bauhaus @ 2007-11-12 15:09 UTC (permalink / raw)
On Mon, 2007-11-12 at 08:04 -0500, Stephen Leake wrote:
> Brian May <bam@snoopy.apana.org.au> writes:
>
> >>>>>> "Harald" == Harald Korneliussen <vintermann@gmail.com> writes:
> >
> > Harald> Now, one may argue whether the Java/C# approach of
> > Harald> virtual machines, sandboxes, bytecode verification etc.
> > Harald> is better than the Ada approach, but it is at least an
> > Harald> attempt. It's not as if they're doing nothing.
> >
> > There is nothing stopping Ada from having virtual machines, sandboxes,
> > bytecode verification, etc, either. If they really wanted to do so.
>
> AdaCore has ported GNAT to .NET. There was even a small ad in the
> current Dr Dobbs for it.
AppletMagic is integrated with SofCheck's Inspector (which is
made to help verification).
^ permalink raw reply [flat|nested] 16+ messages in thread
end of thread, other threads:[~2007-11-12 15:09 UTC | newest]
Thread overview: 16+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2007-11-09 17:33 A scary story from the real world Per Sandberg
2007-11-09 20:46 ` Jerry
2007-11-09 21:15 ` Per Sandberg
2007-11-09 21:25 ` Larry Kilgallen
2007-11-10 23:03 ` Jerry
2007-11-11 0:11 ` tmoran
2007-11-11 8:46 ` Dmitry A. Kazakov
2007-11-12 7:36 ` Harald Korneliussen
2007-11-12 8:27 ` Brian May
2007-11-12 13:04 ` Stephen Leake
2007-11-12 15:09 ` Georg Bauhaus
2007-11-11 14:24 ` Brian Drummond
2007-11-09 22:09 ` anon
2007-11-10 0:06 ` Brian May
2007-11-10 5:44 ` Larry Kilgallen
2007-11-10 21:40 ` Harald Korneliussen
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox