blinded by light

On 9/30/2019 9:10 PM, Jeff Liebermann wrote:
On Tue, 01 Oct 2019 07:52:24 +0700, John B.
wrote:

On Mon, 30 Sep 2019 06:00:24 -0700, Jeff Liebermann
wrote:

As the driver who appeared sound asleep in his self driven car
(recently highlighted here) demonstrates , "rely on the
intelligence of the operator" may not be the best path to take.

Yep. The "nut that holds the wheel" tends to be the major problem.
For every advance in automation, safety devices, and product liability
litigation, there is an equal advance in operator idiocy, inattention,
product abuse, and assumption of safety. I suspect the biggest
problem is that Joe Sixpack tends to believe the manufacturers claims,
instead of his own common sense. Of course, the early adopters tend
to be those with more money than common sense, so disasters are
probably inevitable. Until science invents an evolution accelerator,
we are faced with an apparently endless supply of better idiots.

In fact, as I think Frank will testify, industrial safety is largely
concerned with eliminating "reliance on the intelligence of the
operator" :-)

As usual, I beg to differ. I covered this in a previous rant about
the INCREASE in accidents that appeared after safety interlocks were
installed on a terminal to wire crimping machine at a former employer.
Before the safety interlocks were installed, the operators were all
trained and told that the machine will chop off a hand or several
fingers if they get anywhere near the moving parts. There were no
accidents for about 5 years of operation. After OSHA demanded that
the machine be retrofitted with hand safety interlocks, accidents
started to appear. Fortunately, none resulted in the loss of any
fingers, but did produce some impressive "pinch" injuries caused by
the clear plastic safety shields.

The problem was that with the safety interlocks, the operators felt
like they were safe. That gave them the license to do some stupid
things, which resulted in the injuries. So, which is better? To
terrorize the operator with visions of injuries if they screw up, or
to have them half-asleep assuming that they would be safe no matter
what they do wrong? The current fashion in the latter, but I prefer
the former.


Which brings us back to Chalo's comment form years ago on
RBT that auto steering wheels ought to have a pointy spike
in the center.

--
Andrew Muzi
www.yellowjersey.org/
Open every day since 1 April, 1971

On Tue, 1 Oct 2019 07:54:50 +0000 (UTC), David Scheidt
wrote:

https://youtu.be/tCX9K0Jk6ME?t=146

I like that. "Bicycles, they're not meant to be safe."

--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

On 10/1/2019 9:14 AM, AMuzi wrote:
On 9/30/2019 9:10 PM, Jeff Liebermann wrote:
On Tue, 01 Oct 2019 07:52:24 +0700, John B.
wrote:

On Mon, 30 Sep 2019 06:00:24 -0700, Jeff Liebermann
wrote:

As the driver who appeared sound asleep in his self driven car
(recently highlighted here) demonstrates , "rely on the
intelligence of the operator" may not be the best path to take.

Yep.Â* The "nut that holds the wheel" tends to be the major problem.
For every advance in automation, safety devices, and product liability
litigation, there is an equal advance in operator idiocy, inattention,
product abuse, and assumption of safety.Â* I suspect the biggest
problem is that Joe Sixpack tends to believe the manufacturers claims,
instead of his own common sense.Â* Of course, the early adopters tend
to be those with more money than common sense, so disasters are
probably inevitable.Â* Until science invents an evolution accelerator,
we are faced with an apparently endless supply of better idiots.

In fact, as I think Frank will testify, industrial safety is largely
concerned with eliminating "reliance on the intelligence of the
operator" :-)

As usual, I beg to differ.Â* I covered this in a previous rant about
the INCREASE in accidents that appeared after safety interlocks were
installed on a terminal to wire crimping machine at a former employer.
Before the safety interlocks were installed, the operators were all
trained and told that the machine will chop off a hand or several
fingers if they get anywhere near the moving parts.Â* There were no
accidents for about 5 years of operation.Â* After OSHA demanded that
the machine be retrofitted with hand safety interlocks, accidents
started to appear.Â* Fortunately, none resulted in the loss of any
fingers, but did produce some impressive "pinch" injuries caused by
the clear plastic safety shields.

The problem was that with the safety interlocks, the operators felt
like they were safe.Â* That gave them the license to do some stupid
things, which resulted in the injuries.Â* So, which is better?Â* To
terrorize the operator with visions of injuries if they screw up, or
to have them half-asleep assuming that they would be safe no matter
what they do wrong?Â* The current fashion in the latter, but I prefer
the former.


Which brings us back to Chalo's comment form years ago on RBT that auto
steering wheels ought to have a pointy spike in the center.

To which, years ago, I probably wrote "+1".


--
- Frank Krygowski

On 9/30/2019 10:10 PM, Jeff Liebermann wrote:
On Tue, 01 Oct 2019 07:52:24 +0700, John B.
wrote:

On Mon, 30 Sep 2019 06:00:24 -0700, Jeff Liebermann
wrote:

As the driver who appeared sound asleep in his self driven car
(recently highlighted here) demonstrates , "rely on the
intelligence of the operator" may not be the best path to take.

Yep. The "nut that holds the wheel" tends to be the major problem.
For every advance in automation, safety devices, and product liability
litigation, there is an equal advance in operator idiocy, inattention,
product abuse, and assumption of safety. I suspect the biggest
problem is that Joe Sixpack tends to believe the manufacturers claims,
instead of his own common sense. Of course, the early adopters tend
to be those with more money than common sense, so disasters are
probably inevitable. Until science invents an evolution accelerator,
we are faced with an apparently endless supply of better idiots.

In fact, as I think Frank will testify, industrial safety is largely
concerned with eliminating "reliance on the intelligence of the
operator" :-)

As usual, I beg to differ. I covered this in a previous rant about
the INCREASE in accidents that appeared after safety interlocks were
installed on a terminal to wire crimping machine at a former employer.
Before the safety interlocks were installed, the operators were all
trained and told that the machine will chop off a hand or several
fingers if they get anywhere near the moving parts. There were no
accidents for about 5 years of operation. After OSHA demanded that
the machine be retrofitted with hand safety interlocks, accidents
started to appear. Fortunately, none resulted in the loss of any
fingers, but did produce some impressive "pinch" injuries caused by
the clear plastic safety shields.

The problem was that with the safety interlocks, the operators felt
like they were safe. That gave them the license to do some stupid
things, which resulted in the injuries. So, which is better? To
terrorize the operator with visions of injuries if they screw up, or
to have them half-asleep assuming that they would be safe no matter
what they do wrong? The current fashion in the latter, but I prefer
the former.

What you're discussing has been termed "risk compensation" or "risk
homeostasis." Give someone a feeling of protection and they will take
more risks.

It's actually perfectly logical. Of _course_ someone will descend stairs
faster if they can hold on to a handrail. Of _course_ a worker will try
for higher production if he's on incentive and the machine won't chop
off his fingers. Of _course_ a person will jump out of a plane with a
parachute, but won't do it without one. All that is reasonable.

But it gets more complicated in some situations. Nobody recognized that
of _course_ British drivers would take more risks when they were made to
wear seatbelts. (Their KSI counts dropped. But passengers - who were not
required to wear them - had their KSIs increase. So did pedestrians.)

It's also more complicated when the feeling of protection exceeds the
actual protection. My two favorites, of course, are bike helmets ("They
prevent 85% of head injuries!!!") and "protected" cycletracks ("Finally,
someplace _safe_ to ride!").

Results? Bike concussions have risen over 60% since helmets became
popular. And the latest study by the Insurance Institute for Highway
Safety found one-directional "protected" cycletracks had about 19%
higher crash risk than a major road with NO bike lanes. And a
bi-directional "protected" cycletrack had over 11 TIMES the crash risk
as a major road without bike lanes.

"Ah yes, I can relax. It doesn't matter if I zoom into this intersection
without looking for cross traffic, turning traffic, motorists pulling
out and stuff like that. I'm protected! Besides, I'm wearing a helmet!!"


--
- Frank Krygowski

On 9/30/2019 8:52 PM, John B. wrote:
On Mon, 30 Sep 2019 06:00:24 -0700, Jeff Liebermann
wrote:

On Mon, 30 Sep 2019 13:26:05 +0700, John B.
wrote:

The problems with all the modern, marvelous, systems is, "sometimes
they don't work".

True. However, no system works 100% of the time in 100% of all
possible (contrived) situations. A bicycle that is quite suitable for
riding on pavement would probably do badly in dirt, mud, rain, etc.
One has to design for either a specific situation, or as in this case,
the greatest number of reasonable situations, and rely on the
intelligence of the operator to know when to disarm the monster. It
is impossible to design out clueless operators because we continue to
produce better clueless operators.

As the driver who appeared sound asleep in his self driven car
(recently highlighted here) demonstrates , "rely on the
intelligence of the operator" may not be the best path to take.

In fact, as I think Frank will testify, industrial safety is largely
concerned with eliminating "reliance on the intelligence of the
operator" :-)

I think it's worse than that. Industrial safety nowadays has to prevent
deliberate suicide if it's at all possible to do so. As in "He crawled
under the barrier on his belly and stood in front of the robot so it
would stab him to death. That's the company's fault."


--
- Frank Krygowski

On 9/30/2019 10:49 PM, Jeff Liebermann wrote:
On Tue, 01 Oct 2019 08:00:10 +0700, John B.
wrote:

"Back in the day" everything that Cadillac did was considered a
feature. Remember the tail fins? Caddy had them first :-)

Cadillac’s IntelliBeam System Automatically Enables High-Beams:
http://gmauthority.com/blog/2014/11/cadillacs-intellibeam-system-automatically-enables-high-beams-feature-spotlight/

It doesn't seem to claim to turn off the brights if a bicyclist or
pedestrian is approaching. :-(

(Bicycles are defined as vehicles in some states, but not in others.)


--
- Frank Krygowski

On 9/30/2019 9:16 PM, John B. wrote:
On Mon, 30 Sep 2019 09:30:18 -0700 (PDT), Frank Krygowski
wrote:

On Monday, September 30, 2019 at 10:11:05 AM UTC-4, duane wrote:
Here lights are required at
night. There's no specification except one white in front and one red
behind.

I think the NHTSA codes and most U.S. state codes say the bike lights must be
visible from 500 feet. It's true that's a pretty vague spec. Visible under what
conditions? Total darkness or surrounded by city glare? And it's high time we
had upper limits, as well.

- Frank Krygowski

Visible from a stated distance is a fairly standard specification.
Marine navigation lights have used this standard for more than a
hundred years.

I think it makes more sense in a marine environment. There's rarely
enough bright surrounding light to cause an observer's pupils to stop down.


--
- Frank Krygowski

On 10/1/2019 2:54 PM, Frank Krygowski wrote:
On 9/30/2019 8:52 PM, John B. wrote:
On Mon, 30 Sep 2019 06:00:24 -0700, Jeff Liebermann

wrote:

On Mon, 30 Sep 2019 13:26:05 +0700, John B.
wrote:

The problems with all the modern, marvelous, systems is,
"sometimes
they don't work".

True. However, no system works 100% of the time in 100%
of all
possible (contrived) situations. A bicycle that is quite
suitable for
riding on pavement would probably do badly in dirt, mud,
rain, etc.
One has to design for either a specific situation, or as
in this case,
the greatest number of reasonable situations, and rely on
the
intelligence of the operator to know when to disarm the
monster. It
is impossible to design out clueless operators because we
continue to
produce better clueless operators.

As the driver who appeared sound asleep in his self driven
car
(recently highlighted here) demonstrates , "rely on the
intelligence of the operator" may not be the best path to
take.

In fact, as I think Frank will testify, industrial safety
is largely
concerned with eliminating "reliance on the intelligence
of the
operator" :-)

I think it's worse than that. Industrial safety nowadays has
to prevent deliberate suicide if it's at all possible to do
so. As in "He crawled under the barrier on his belly and
stood in front of the robot so it would stab him to death.
That's the company's fault."



With a mind like that you might consider a second career as
plaintiff's attorney.

--
Andrew Muzi
www.yellowjersey.org/
Open every day since 1 April, 1971

On Tue, 1 Oct 2019 15:57:25 -0400, Frank Krygowski
wrote:

On 9/30/2019 10:49 PM, Jeff Liebermann wrote:
On Tue, 01 Oct 2019 08:00:10 +0700, John B.
wrote:

"Back in the day" everything that Cadillac did was considered a
feature. Remember the tail fins? Caddy had them first :-)

Cadillac’s IntelliBeam System Automatically Enables High-Beams:
http://gmauthority.com/blog/2014/11/cadillacs-intellibeam-system-automatically-enables-high-beams-feature-spotlight/

It doesn't seem to claim to turn off the brights if a bicyclist or
pedestrian is approaching. :-(

(Bicycles are defined as vehicles in some states, but not in others.)

That was 5 years ago. Maybe General Motors has discovered that there
are such things as bicycles by now. Yep:
"General Motors starts taking orders for its first e-bike"
https://www.theverge.com/2019/2/14/18224778/gm-ebike-ariv-specs-price-europe

What bugs me is that Intellibeam will not allow the high beams to turn
on unless a list of conditions are met. The last one is rather odd:
"You are driving on winding or hilly roads"
Hills and blind turns are where I need high beams the most.

Perhaps this condition:
"The other vehicle’s lamps are missing, damaged, obstructed
from view, or otherwise undetected"
might qualify for detecting bicycles without lights. As I read it,
the Cadillac's brights will only activate if the bicyclist has rather
bright and blinding headlights. The author of the specs for the GM
Intellibeam doesn't seem to understand how high beams work.



[Driver] Caddy, turn on headlight high beams.
[Caddy] I'm sorry Dave, but I can't do that.
[Driver] Why not?
[Caddy] You have not met the required conditions for enabling the
high beams.
[Driver] Which conditions?
[Caddy] The one's listed in your owners manual. Would you like me
to read them to you?
[Driver] Never mind. I might as well drive blind folded.
[Caddy] I'm sorry Dave, but I can't let you do that. I'll take
control of the vehicle until the driver analysis computer can
determine if it is safe for you to continue driving.
(Passenger eject activates and launches driver through the sunroof).
[Caddy] Oops.

--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

On Tue, 01 Oct 2019 08:14:17 -0500, AMuzi wrote:

Which brings us back to Chalo's comment form years ago on
RBT that auto steering wheels ought to have a pointy spike
in the center.

It won't work. Drivers will become accustomed and immunized against
the threat of being impaled after a few months of driving without any
fatal incidents. You can't convince someone by merely repeating a
threat over and over on a daily basis with nothing happening. It's
much like Chicken Little warning everyone constantly that the sky is
falling until everyone is so tired of hearing the warning that they
ignore it when the problem becomes real. Same with the spike. The
driver that impales himself on the steering wheel will probably act
surprised and announce "Gee, it's never done that before".

If you want a lasting effect, you have to hit the driver with a single
spectacular blow that will leave a lasting impression of fear, and not
a continuous irritation. Driver training tries to do that by showing
movies of mangled accident victims. The military does the same by
showing movies of battlefield carnage. I did the same thing to
several pre-teens caught smoking by giving them a tour of the local
hospital oncology ward and having the lung cancer and emphysema
victims terrorize the kids from their hospital beds. The production
manager at a former employer made it clear that the machines were
dangerous by inserting a sausage into the crimper and waving around
the results. (Unfortunately, he ruined the effect by munching on the
remains of the sausage).

Maybe if prospective bicycle buyers were required to negotiate and
survive 30 minutes of downtown traffic, complete with simulated
injuries and fatalities, before a store is allowed to sell them a
bicycle, we might see a better class of cyclists. Same for car
drivers.
--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558