Board index » cppbuilder » Re: DevCo Luminaries Chat

Re: DevCo Luminaries Chat


2006-04-05 06:50:28 AM
cppbuilder54
"Kenneth de Camargo" <INVERT:rb.moc.arret@jcrk>writes:
Quote
I wonder if it would fly in court. "It was a computer failure, you
honour".
As I said before, the Therac-25 case medical negligence case ( caused
by poorly designed software )is famous. There are many other simpler cases.
Here is just one simple example:
Source: e-Health Insider Primary Care
Medical News Summary (summary of medical news story as reported by
e-Health Insider Primary Care): A second patient has been found to be
given the wrong medication following a computer error at a Glascow
practice. The 90 year old patient was wrongly given
phenobarbiturates. The woman noticed the mistake which could have
proved fatal. The practice claims to have instigated changes to
prevent repeat occurrences.
URL: www.ehiprimarycare.com/news/item.cfm
 
 

Re:Re: DevCo Luminaries Chat

"Brion L. Webster" < XXXX@XXXXX.COM >writes:
Quote
Actually, Ken, didn't someone in your neck of the woods do just that
when the chemotherapy machine couldn't figure out the right radiation
dose when using three plates instead of two, some simple topology
problem but the software wasn't designed to accept that number of
inputs?
Therac-25
 

Re:Re: DevCo Luminaries Chat

< XXXX@XXXXX.COM >wrote:
Quote
That is because we view dentists are those who were not bright enough
to be doctors and so should not be trusted with anything too
important.
<joke>Is that the legacy of UK dental care speaking? </joke>
FWIW, I find that a pretty horrendously insulting statement. I know quite
a few dentists in the U.S. who chose to go to Dental school, who encounter
a much wider variety of clinical issues than "just cleaning teeth", and
are just as competent if not more so in their field than other
professionals. Several made the reasoned decision that dentistry was just
plain more profitable too. Hard to argue with that logic, especially when
you compare their take home to mine, and the number of hours.
--
-Brion
There's no such thing as 'one, true way;'
- Mercedes Lackey
 

{smallsort}

Re:Re: DevCo Luminaries Chat

"Kenneth de Camargo" <INVERT:rb.moc.arret@jcrk>writes:
Quote
>Never heard of Therac-25 ?

Hey, ever heard of Zomax? I win!
It is an NSAID.
So ?
 

Re:Re: DevCo Luminaries Chat

"Kenneth de Camargo" <INVERT:rb.moc.arret@jcrk>writes:
Quote
>>Or if you don't stop drilling in time...
>Instant {*word*7}fistula :-)
Depends on the set of teeth.
My cousin ( a gynaecologist ) told me about an older lady with a
{*word*266}l discharge that he treated. He found that she had put her
false teeth in her {*word*266} for safe keeping and then had forgotten
about them.
She was effusively grateful when he found them and pulled them out,
since she had not been able to eat her favorite foods for several
weeks or so.
 

Re:Re: DevCo Luminaries Chat

"Brion L. Webster" < XXXX@XXXXX.COM >writes:
Quote
>That is because we view dentists are those who were not bright enough
>to be doctors and so should not be trusted with anything too
>important.

<joke>Is that the legacy of UK dental care speaking? </joke>
I am in Australia.
Quote
FWIW, I find that a pretty horrendously insulting statement.
I guess that it is different in the US, however dentists have a rather
low status here. My neighbour's son is doing dentistry. His parents
were devastated when he failed to get into med school and had to go to
dental school instead.
 

Re:Re: DevCo Luminaries Chat

At 00:23:13, 05.04.2006, < XXXX@XXXXX.COM >wrote:
Quote
"Rudy Velthuis [TeamB]" < XXXX@XXXXX.COM >writes:

>(*)or if I make another mistake, like pulling the tooth of a bleeder,
>or prescribing the wrong medicine, someone dies.

Not sure about Germany, but here in Australia dentists are only
allowed to prescribe amoxycilling ( for endocarditis prophylaxis ) and
erythromycin ( to those allergic to penicillin ).
It's Amoxycillin, and no serious dentist would prescribe that, if he can
avoid it. Amoxycillin is a broad spectrum antibiotic, and not nearly as
effective as antibiotics specific for the bacteria found in the mouth. So
a dentist generally prescribes Penicillin V here, if the patient has no
problems with that. Only if that is not possible, he will try
Erythromycin, or Doxycyclin.
Here, we are also allowed to prescribe pain killers, or other
antibiotics, etc. I can kill people with that.
Quote
That is because we view dentists are those who were not bright enough
to be doctors and so should not be trusted with anything too
important.
I see. How do you view programmers, over there? <g>
--
Rudy Velthuis [TeamB] rvelthuis.de/
"The man who goes alone can start today; but he who travels with another
must wait till that other is ready."
-- Henry David Thoreau (1817-1862)
 

Re:Re: DevCo Luminaries Chat

A Digital Matter of Life and Death
by Ivars Peterson
Science News, 12 March 1988
The radiation-therapy machine, a Therac 25 linear accelerator, was designed to
send a penetrating X-ray or electron beam deep into a cancer patient's body to
destroy embedded tumors without injuring skin tissue. But in three separate
instances in 1985 and 1986, the machine failed. Instead of delivering a safe
level of radiation, the Therac 25 administered a dose that was more than 100
times larger then the typical treatment dose. Two patients died and a third
was severely burned.
The malfunction was caused by an error in the computer program controlling
the machine. It was a subtle error that no one had picked up during the
extensive testing the machine had undergone. The error surfaced only when a
technician happened to use a specific, unusual combination of keystrokes to
instruct the machine.
The Therac incidents and other cases of medical device failures caused by
computer errors have focused attention on the increasingly important role
played by computers in medical applications. Computers or machines with
built-in microprocessors perform functions that range from keeping track of
patients to diagnosing ailments and providing treatments.
 

Re:Re: DevCo Luminaries Chat

David Erbas-White wrote:
Quote
I believe the statement
was originally made in the context of demonstrating why it is
important for the BCB compiler to be 'compliant', and for bugs in it
to be fixed, was it not? One poster had indicated that there was
nothing 'mission critical' about Borland's tools, and the response
(which you seem to have taken offense to) was attempting to
demonstrate that, indeed, there are 'mission critical' programs out
there -- I don't think it was intended as "my work is more important
than your work".
And in any event, Borland's products are not stated as being certified
for use in mission critical situations, nor is Windows, btw.
Quote
Again, I disagree. As a general rule, I find that customers of an
industry have a better 'feel' for the industry than the persons in it
(and that is true of any industry, not just the medical profession).
I have to disagree here. User feedback, wherever the area, is
fundamental, but the expertise simply is not there.
Quote
I've witnessed (and been a party too) too many examples of
incompetence in the medical industry to assume that just because they
have training/certification, that they really "know what they're
doing". That's because medical professionals are just as human as
anyone else, and just as prone to make mistakes, and just as prone to
not want to admit them. They also tend (just as everyone else) to
have "tunnel vision", and not necessarily see things outside their
immediate focus.
No argument there.
Quote
That's why I don't understand this mindset that excuses Borland for
the problems with their product. It's really unconscionable for them
not to be expending every effort to make their product work correctly
-- but I have NEVER (repeat, NEVER) seen a software company as
difficult to work with as Borland has been for the past several
years. They have (had) some great products, but the support that has
been provided has been horrifyingly poor.
I don't think that anyone should give (or is giving) Borland a free
pass on anything, particularly wrt customer relations (and there has
been a visible effort to improve that area) even more than with
support. That being said, it's unreasonable to expect their products a
degree of reliability (again, mission critical) that is not available
even in the OS they run on/generate code for.
--
Ken
planeta.terra.com.br/educacao/kencamargo/
* this is not a sig *
 

Re:Re: DevCo Luminaries Chat

"Kenneth de Camargo" <INVERT:rb.moc.arret@jcrk>writes:
Quote
>I've witnessed (and been a party too) too many examples of
>incompetence in the medical industry to assume that just because they
>have training/certification, that they really "know what they're
>doing". That's because medical professionals are just as human as
>anyone else, and just as prone to make mistakes
I have an amusing story about a child who was prescribed a lethal dose
of codiene by the local children's hospital for a broken leg.....
 

Re:Re: DevCo Luminaries Chat

At 01:26:47, 05.04.2006, < XXXX@XXXXX.COM >wrote:
Quote

A Digital Matter of Life and Death
Yeah, heard of that, in my radiology fresh-up course. Now like I said, I
doubt it was (or should have been) programmed in C++ on Windows. There
are special, fail-safe languages and systems for such operations.
That is not the same scale as a DB app handling prescriptions, though.
This was hardware controlled by software.
--
Rudy Velthuis [TeamB] rvelthuis.de/
"Believe those who are seeking the truth. Doubt those who find it."
-- Andr?Gide
 

Re:Re: DevCo Luminaries Chat

At 00:46:09, 05.04.2006, David Erbas-White wrote:
Quote
Again, I disagree. As a general rule, I find that customers of an
industry have a better 'feel' for the industry than the persons in it
I very much doubt that. Most patients have absolutely no feel for the
medical industry, except for the very superficial contacts they have with
it.
--
Rudy Velthuis [TeamB] rvelthuis.de/
"The secret of a good sermon is to have a good beginning and a good
ending, then having the two as close together as possible."
-- George Burns.
 

Re:Re: DevCo Luminaries Chat

Kenneth de Camargo wrote:
Quote


And in any event, Borland's products are not stated as being certified
for use in mission critical situations, nor is Windows, btw.


Let's be honest. I have never, not anywhere, found an off-the-shelf
product that states that it CAN be used in mission-critical or
life-support areas. If you look at ANY parts data sheet, there will be
extremely fine print that specifically states that it is not warranted
for that purpose.
So, what's the bottom line? Engineers end up using the parts anyway,
but their legal department either declaims responsibility (as well), or
they jack up the price enough to 'insure' against any possible lawsuits.
There's not a medical device out there that doesn't include some kind of
'not allowed' integrated circuit -- it's a simple matter of reality
brought on by the current legal climate.
The same is true of Windows, DOS, Borland tools, etc. In fact, with
software, the EULAs already declaim any liability for anything other
than the distribution media being faulty. But, again, the reality is
that these items can/must be used, anyway.
The argument that Borland is not responsible for 'mission critical'
applications is fallacious, because the simple truth is that the errors
in Borland product relate to non-critical software development as well.
It's simply been pointed out that one of the ramifications of the poor
level of tools support is that society as a whole ends up paying the
price, due to the reality that these 'non allowed' tools MUST be
utilized, or progress grinds to a halt.
David Erbas-White
 

Re:Re: DevCo Luminaries Chat

At 01:07:55, 05.04.2006, < XXXX@XXXXX.COM >wrote:
Quote
My cousin ( a gynaecologist ) told me about an older lady with a
{*word*266}l discharge that he treated. He found that she had put her
false teeth in her {*word*266} for safe keeping and then had forgotten
about them.
Er...
--
Rudy Velthuis [TeamB] rvelthuis.de/
"I think 'Hail to the Chief' has a nice ring to it."
- John F. Kennedy (1917-1963) when asked what is his favorite song
 

Re:Re: DevCo Luminaries Chat

Rudy Velthuis [TeamB] wrote:
Quote
At 01:26:47, 05.04.2006, < XXXX@XXXXX.COM >wrote:


>A Digital Matter of Life and Death


Yeah, heard of that, in my radiology fresh-up course. Now like I said, I
doubt it was (or should have been) programmed in C++ on Windows. There
are special, fail-safe languages and systems for such operations.
Really? Please enlighten me. What are these special fail-safe
languages and systems?
The software used on the space shuttle is generally recognized as the
most thoroughly reviewed/debugged software in history. And given that,
they estimate there is still one bug per thousand lines of code.
So, what, exactly, are these special fail-safe languages? And even if a
'standard' exists for such a language, the fact of language compliance
has been demonstrated to be problematic over and over and over again, so
please point out and actual, fail-safe language AND IMPLEMENTATION...
David Erbas-White
Quote

That is not the same scale as a DB app handling prescriptions, though.
This was hardware controlled by software.