visibility of DRL

On 4/7/2019 5:34 PM, wrote:

snip

I find it strange. You keep making statements, such as "multiple
studies" but for some reason you always neglect to provide any
references to actual studies.

If you check the Usenet archives (deja.com) you can find the links to
studies on the effectiveness of DRLs. Don't limit your search to just
bicycle DRLs, also look at motorcycle DRLs. You may find it hard to
believe, but there are not hundreds of government funded surveys on
bicycle DRLs.

But even if there were hundreds of double-blind studies, it would be
immaterial. Anytime a statistically significant survey, or a
scientifically valid study is presented, you will have a small number of
people that will absolutely refuse to believe it based on some excuse.
That's why there are still climate change deniers, flat-earthers, and
those that don't believe that there is any benefit to wearing a helmet
in a head-impact crash.

The fact that companies selling DRLs for bicycles tout the benefits of
DRLs, does not negate the research conclusions, but the logic you see in
this thread is along the lines of "Trek states "all existing research
indicates that the single best way for a cyclist to increase the
likelihood of being seen by a driver is to use a flashing light that’s
daylight visible." Trek sells lights with a flash mode for daytime use.
Therefore anything Trek says regarding the benefits of DRLs must be false."

On 4/7/2019 3:53 PM, wrote:

But they didn't use to be. And in foreign countries where wheel driven
generators were normally fitted to "transportation" bikes they still
are today.

When you're just trying to illuminate a small patch of road directly in
front of slow moving commuter bicycle then a low-power light mounted
lower makes a lot of sense.

When you're trying to illuminate further ahead, as well as off to the
sides and slightly upwards, a handlebar mounted light makes more sense.

I definitely remember Schwin bikes, the good strong ones with the
double top tubes, that had a front light mount mounted under the head
bearing lock nut.

That's a good position, and many lights can still mount onto a bracket
under the lock nut for the threaded stem. You can also buy a threadless
headset spacer with light bracket, i.e.the Minoura CS-500, but these
aren't very common.

On Mon, 8 Apr 2019 01:18:20 -0700, sms
wrote:

On 4/7/2019 5:34 PM, wrote:

snip

I find it strange. You keep making statements, such as "multiple
studies" but for some reason you always neglect to provide any
references to actual studies.

If you check the Usenet archives (deja.com) you can find the links to
studies on the effectiveness of DRLs. Don't limit your search to just
bicycle DRLs, also look at motorcycle DRLs. You may find it hard to
believe, but there are not hundreds of government funded surveys on
bicycle DRLs.

Are you saying that you have not "checked the archives" and thus are
unable to underwrite your claims with facts?

Why is that? After all one does prefer that anyone who expounds
supposed "facts" has a bit more than just a loud voice and numerous
reputations to justify their arguments.

Or perhaps you would have us believe that you are somehow aware of
some sort of secret information, only available to the initiated, and
that you are now divulging these secrets to the unwashed masses?

But even if there were hundreds of double-blind studies, it would be
immaterial. Anytime a statistically significant survey, or a
scientifically valid study is presented, you will have a small number of
people that will absolutely refuse to believe it based on some excuse.

That is hardly true in your case. No one is denying anything. What you
are being asked is to backup your fevered claims with facts. A feat
that, to date, appears to be beyond your capability.

That's why there are still climate change deniers, flat-earthers, and
those that don't believe that there is any benefit to wearing a helmet
in a head-impact crash.

The fact that companies selling DRLs for bicycles tout the benefits of
DRLs, does not negate the research conclusions, but the logic you see in
this thread is along the lines of "Trek states "all existing research
indicates that the single best way for a cyclist to increase the
likelihood of being seen by a driver is to use a flashing light that’s
daylight visible." Trek sells lights with a flash mode for daytime use.
Therefore anything Trek says regarding the benefits of DRLs must be false."

The fact that a company sell something proves nothing except that said
company wishes to market a product, hopefully at a profit. It says
nothing for the usability or value of the item.

You can, for one example, buy a pet rock, for $7.99, on Amazon, which
comes in a ventilated cardboard box and can be attached to your
computer USB port. And does absolutely nothing.
--
cheers,

John B.

On Sun, 07 Apr 2019 05:49:43 +0700, wrote:

Yes the often mentioned Odense study (of some 1,845 users and some
2,000 non users over a 12 month period) showed permanent running
lights reduced accidents by 19%.

Did you perhaps forget about my cursory and superficial analysis of
that report in answer to your comments on the same?
https://groups.google.com/d/msg/rec.bicycles.tech/CVAE2CuCHo8/VRIyIZ-cAgAJ
If you accept my suspicions, the actual results of the study, ignoring
a 28% fudging of the treatment group data, would be a 9% INCREASE in
accidents by the treatment group over the control group.

The study was initiated and funded by the company that made the
permanent running lights.

Yeah, that's a problem. Nobody else would consider funding such a
study so it was left to the manufacturer to provide the necessary
incentives.

Topic drift:
I recently had an entertaining example of how that can go wrong. In
the past, the wireless phone providers would fund studies that
conveniently prove that cell phone radiation is harmless. To give the
impression of being objective, the study would be performed by a
university or commercial laboratory with the funding companies or
organization distancing themselves as much as possible. The resulting
reports all tended to favor the point of view of the funding
companies, which was duly noticed by their critics. The result was
that such reports were generally considered "slanted" and ignored by
the press and some regulatory authorities. So, the cell phone
companies simply stopped funding such studies because nobody accepted
the results.

Roll forward a few years and the FCC released some new frequencies for
5G phones and devices some of which were in the millimeter wavelength
region. This novelty allowed individuals and organizations who
consider any form of RF to be hazardous to recycle all the old FUD
(fear, uncertainty, doubt) from previous battles over RF
proliferation. Eventually, the politicians got involved, and held a
hearing where an official of the CTIA (Cellular Telecommunications
Industry Association) was grilled by a clueless senator over why there
had been no research done by the industry on the alleged dangers and
health effects of 5G wireless. The official had been properly coached
to not mention anything about how such studies are funded, so all he
could do was admit that no studies had been performed. He couldn't
testify that a study would have been a waste of time and money because
nobody would believe the results of a study funded by the telecom
industry.

"At Senate Commerce Hearing, Blumenthal Raises Concerns on 5G Wireless
Technology's Potential Health Risks"
https://www.blumenthal.senate.gov/newsroom/press/release/at-senate-commerce-hearing-blumenthal-raises-concerns-on-5g-wireless-technologys-potential-health-risks

"US Senator Blumenthal Raises Concerns on 5G Wireless Technology
Health Risks at Senate Hearing"
https://www.youtube.com/watch?v=ekNC0J3xx1w (4:51)

For those who believe in DRL, no proof is necessary. For those who do
not believe in DRL, no proof is possible.

End of topic drift and I return now to working on my taxes.

--
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 Sat, 6 Apr 2019 23:22:06 +0000 (UTC), Ralph Barone
wrote:

They probably could have handed out St Christopher medals and gotten
similar results.

True. In college in the late 1960's, I was involved in such a study.
I won't mention all the particulars because it would surely produce an
irrelevant debate on the topic. The test group was given a device
intended to produce a specific result. The control group was given a
placebo device that did nothing. The study was a double blind study
where neither the participants or researchers were told which was the
real device or the placebo. What only a few people knew was that both
devices were identical placebos that differed only slightly in the
package coloring and markings.

I vaguely recall about 80 students in each group. The test ran for
about 6 weeks. Participants were expected to keep detailed logs of
the effectiveness of the device. After about one week, the difference
in the packaging and coloring was "discovered" and duly noted by the
participants. A rumor was "accidentally" leaked indicating which
package was real and which was a placebo.

A few weeks later, the logged results were collected and tabulated
(literally on Hollerith cards), fed to the IBM tabulator, summarized
by the IBM 1620 computah, and printed for distribution among the
participants. The test group produced about 40% better results than
the control group. So much for the value of such studies.

The problem with DRL is that anyone can see that the light is flashing
or not flashing. Everyone in the test group knows that they have a
functional device and everyone in the control group knows that they
don't have anything to test. Expected results soon turn into actual
results. In other words, there's no way I could arrange a double
blind study for DRL safety and the results will tend to favor the
expected results.


Drivel: Last week, I was driving home late in my car over my usual
route. It was very dark, but no fog or rain. In the distance was a
bright red pattern of about four flashing light. I couldn't tell if
they were moving or stationary. I watched them carefully trying to
identify if it was a road hazard flasher, bicycle, or alien flying
saucer. Finally, I was close enough to see that it was a bicycle with
four asynchronously flashing lights on the helmet, belt, and
left/right arms. I also noticed that I had drifted across the center
divider and into the oncoming traffic lanes. Nobody was in those
lanes or I might have had a rather spectacular head-on auto accident.

It would seem that exotic patterns of red lights are a form of
distracted driving. Perhaps changing the law to allow bicycles and
other human powered vehicles to use a different color flashing tail
light, or perhaps a specific pattern of flashes, might be helpful.


--
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 Mon, 08 Apr 2019 18:19:47 -0700, Jeff Liebermann
wrote:

On Sat, 6 Apr 2019 23:22:06 +0000 (UTC), Ralph Barone
wrote:

They probably could have handed out St Christopher medals and gotten
similar results.

True. In college in the late 1960's, I was involved in such a study.
I won't mention all the particulars because it would surely produce an
irrelevant debate on the topic. The test group was given a device
intended to produce a specific result. The control group was given a
placebo device that did nothing. The study was a double blind study
where neither the participants or researchers were told which was the
real device or the placebo. What only a few people knew was that both
devices were identical placebos that differed only slightly in the
package coloring and markings.

I vaguely recall about 80 students in each group. The test ran for
about 6 weeks. Participants were expected to keep detailed logs of
the effectiveness of the device. After about one week, the difference
in the packaging and coloring was "discovered" and duly noted by the
participants. A rumor was "accidentally" leaked indicating which
package was real and which was a placebo.

A few weeks later, the logged results were collected and tabulated
(literally on Hollerith cards), fed to the IBM tabulator, summarized
by the IBM 1620 computah, and printed for distribution among the
participants. The test group produced about 40% better results than
the control group. So much for the value of such studies.

The problem with DRL is that anyone can see that the light is flashing
or not flashing. Everyone in the test group knows that they have a
functional device and everyone in the control group knows that they
don't have anything to test. Expected results soon turn into actual
results. In other words, there's no way I could arrange a double
blind study for DRL safety and the results will tend to favor the
expected results.


A friend, now deceased, was in the financial research business and ran
innumerable studies to determine the feasibility of things ranging
(from memory) the building a copper smelter to whether refrigerates
should be sold on time payment up country.

One week end when we were eating pizza and listening to his jazz
records a girl came by the house asking questions for a survey. My
friend refused to answer her questions as he said he was aware of the
purpose of the survey and that would slant his answers. Anyway, after
the girl left he commented that "it is easy to design a study to prove
anything you want".


Drivel: Last week, I was driving home late in my car over my usual
route. It was very dark, but no fog or rain. In the distance was a
bright red pattern of about four flashing light. I couldn't tell if
they were moving or stationary. I watched them carefully trying to
identify if it was a road hazard flasher, bicycle, or alien flying
saucer. Finally, I was close enough to see that it was a bicycle with
four asynchronously flashing lights on the helmet, belt, and
left/right arms. I also noticed that I had drifted across the center
divider and into the oncoming traffic lanes. Nobody was in those
lanes or I might have had a rather spectacular head-on auto accident.

It would seem that exotic patterns of red lights are a form of
distracted driving. Perhaps changing the law to allow bicycles and
other human powered vehicles to use a different color flashing tail
light, or perhaps a specific pattern of flashes, might be helpful.
--
cheers,

John B.

I want a bike-mounted death ray. Please let me know if you source one of those, Andy.

I set up one of my bikes with a front hub dynamo/drum brake. Another one has had a Sanyo bottom bracket dynamo for ages. Both these bikes now sport rectified LED lights I made that are always on (though the BB dynamo is "switched" mechanically). The light on the hub dynamo bike mounts to the front axle so it grazes the road surface; it has a collimated front white LED and an uncollimated red rear LED on the same unit, both supported by supercapacitors for standlight.

I can't say that the daytime running lights have any appreciable effect on drivers, but I know that the ready-to-roll quality of a bike whose lights are always on when it's being used makes me choose that one more often than I otherwise would.

Am 09.04.2019 um 03:19 schrieb Jeff Liebermann:
The problem with DRL is that anyone can see that the light is flashing
or not flashing. Everyone in the test group knows that they have a
functional device and everyone in the control group knows that they
don't have anything to test. Expected results soon turn into actual
results. In other words, there's no way I could arrange a double
blind study for DRL safety and the results will tend to favor the
expected results.

The only meaningful data with DLR is a "regional" test pattern. Germany
investigated this a while ago, imposing DLR for cars on one island and
comparing accident rates with a similar-sized control region.
In both regions, the accident rates reduced by similar amounts.

On 4/8/2019 6:19 PM, Jeff Liebermann wrote:

The problem with DRL is that anyone can see that the light is flashing
or not flashing. Everyone in the test group knows that they have a
functional device and everyone in the control group knows that they
don't have anything to test. Expected results soon turn into actual
results. In other words, there's no way I could arrange a double
blind study for DRL safety and the results will tend to favor the
expected results.

Another issue is that the group with the DRLs will be taking more risks.
It's called " “risk compensation." In the Odense study, the results
would have been skewed by risk compensation. If all the riders had
somehow been forced to take the same risks, the conclusion would have
been different, with the riders using DRLs having even fewer incidents.

On Tuesday, April 9, 2019 at 6:39:51 AM UTC-7, sms wrote:
On 4/8/2019 6:19 PM, Jeff Liebermann wrote:

The problem with DRL is that anyone can see that the light is flashing
or not flashing. Everyone in the test group knows that they have a
functional device and everyone in the control group knows that they
don't have anything to test. Expected results soon turn into actual
results. In other words, there's no way I could arrange a double
blind study for DRL safety and the results will tend to favor the
expected results.

Another issue is that the group with the DRLs will be taking more risks.
It's called " “risk compensation." In the Odense study, the results
would have been skewed by risk compensation. If all the riders had
somehow been forced to take the same risks, the conclusion would have
been different, with the riders using DRLs having even fewer incidents.

Yes, I turn on my front flasher and say "f*** that, I'm going for it!" Then I start racing down the middle of the road, going the wrong way -- "I'm king of the world!" And with my helmet, it's no holds barred!

The safety mavens around here with their five flashers and three headlights, flags, vests, etc., etc. don't seem to be doing any compensating. The most consistently crazy are the ones with the least safety equipment.

-- Jay Beattie.