He talked a little about the Google driverless car but lots of other very interesting material as well. I think it can be seen at the CR website - takes about an hour. He said the traffic in the Silicon Valley is worse now than in Manhattan, where he was today. Google is starting up an office in Manhattan that will be home for some research people from Cornell and the Technion in Israel.
My son who works for Google is in New York right now and his hike from the
hotel to the Google site is 20 minutes. He tried his California shoes the
first two days and wore blisters in his feet so now he has dressed down to
tennis shoes in a city where they all dress up compared to the other coast.
Having a walking perspective of the traffic flow in that part of NY is
different way of looking at this problem. I assume there is not a subway
to hop on if he is hiking 20 full minutes in that part of New York. I
guess the subway is only for some subset of the city.
I would love to build a 180 mph guideway between the new Google New York
office and reasonably priced apartments. My son said one of his co-workers
who is there pays $3,000 a month for his little apartment and walks further
than his hotel at 20 minutes. I paid $9,000 for my first mansion just out
of school. (in a federal target area in south central Fort Worth)
If Google could fix the housing issue for Manhattan they would not need to
pay those guys a big premium to work at that office. The premium for those
high tech workers at the top of their game must be tremendous. Certainly
more money spent going no where (neither Google nor the employees) on pay
premiums than buying a few miles of extrusion at $200,000 per mile.
The hard part is getting to the level within Google where ideas are
accepted. I have talked with the worker dudes inside the car program and
they have to protect their turf and advise that what they are working on is
the best thing ever. If a clear communication could be made to Larry Page
about the gain to Google New York office specifically by building a link
between affordable apartments and their new office he would certainly make
more money for Google and New York would have incrementally less foot
traffic to their office. They see the value of the Google bus obviously
above the price of the bus where they let workers work on the bus. Walking
40 minutes a day is non-productive and you can't wear your California shoes
apparently. 4mph walking speed for a younger worker says the trip is 2.66
miles wasted effort and at 180 mph that is 30 seconds per trip plus 18.6
seconds to accelerate and decelerate.
Housing and office space that are accessible are required to make the pay
premium for attracting workers to that office without breaking the bank.
Advanced transportation like a grown up Shweeb would solve that problem
and improve the life style of Google workers over all other workers in New
York City by thousands of dollars per worker per month's rent.
Many large companies have used the model where they create a company store
and company living arrangements. This typically was a reverse situation
where the company was located in a remote location like a copper mine or a
chemical plant. This is trying to place a big company in a city that is
already full and has rent controlled weirdness screwing up the free market
for living accommodations. Only Google is big enough and has enough vision
and thought for their workers to build a company in a major full city but
look at it like a remote company installation in an empty space. If they
built the apartments and built the offices as well as the physical link
between the two it would be very much like a Dow Chemical building a
magnesium plant on the gulf coast. They have to supply both the factory
and housing as well as a way to get from one to the other. To get land
prices compatible with housing you have to move further from the city
center and that creates mobility need. If one mile is good then 9 miles is
better for getting a better land price to build employee housing or
contracting for a developer to build. At 9 miles the commute is 3 minutes
so the wasted time for workers would not justify pulling out the lap top
and rebooting their Windows 7 if they are not using their Apples. At land
prices in Northern New York the commute could be 27 miles away or 9
minutes. Somewhere there is an optimum for land price rent and
transportation cost. In all cases it takes Google traffic off the
Manhattan sidewalks. Freeing them up for New Yorkers who already live
there.
On Tue, May 22, 2012 at 11:26 PM, Jerry Schneider <j...@peak.org> wrote:
> He talked a little about the Google driverless car but lots of other very
> interesting material as well. I think it can be seen at the CR website -
> takes about an hour. He said the traffic in the Silicon Valley is worse now
> than in Manhattan, where he was today. Google is starting up an office in
> Manhattan that will be home for some research people from Cornell and the
> Technion in Israel.
> --
> You received this message because you are subscribed to the Google Groups
> "transport-innovators" group.
> To post to this group, send email to transport-innovators@**
> googlegroups.com <transport-innovators@googlegroups.com>.
> To unsubscribe from this group, send email to transport-innovators+**
> unsubscribe@googlegroups.com<transport-innovators%2Bunsubscribe@googlegroup s.com>
> .
> For more options, visit this group at http://groups.google.com/** > group/transport-innovators?hl=**en<http://groups.google.com/group/transport-innovators?hl=en>
> .
on 5/23/12 9:34 AM, Jerry Roane at jerry.ro...@gmail.com wrote:
Jerry
My son who works for Google is in New York right now and his hike from the
hotel to the Google site is 20 minutes. He tried his California shoes the
first two days and wore blisters in his feet so now he has dressed down to
tennis shoes in a city where they all dress up compared to the other coast.
Having a walking perspective of the traffic flow in that part of NY is
different way of looking at this problem. I assume there is not a subway to
hop on if he is hiking 20 full minutes in that part of New York. I guess
the subway is only for some subset of the city.
I would love to build a 180 mph guideway between the new Google New York
office and reasonably priced apartments. My son said one of his co-workers
who is there pays $3,000 a month for his little apartment and walks further
than his hotel at 20 minutes. I paid $9,000 for my first mansion just out
of school. (in a federal target area in south central Fort Worth)
Just because Google was formed by a couple of computer nerds who worked
out a means to build a better internet search engine and were able cash in
on it in the marketplace does not necessarily mean that all of the people
subsequently employed by and now run Google have anything close to the right
answers for most other problems as is illustrated by the situation that your
son has discovered in NYC. (I don't mean to cast an doubt on your son's
capability.)
I am afraid that the same goes for whoever at Google cooked up that
self-driving car scheme. It sounds like a great concept until you wake up
from your dream and spend about a minute of fully-awake time trying to think
through the idea. Perhaps this fellow would do much better at working on
something else on his computer.
>on 5/23/12 9:34 AM, Jerry Roane at jerry.ro...@gmail.com wrote:
>Jerry
>My son who works for Google is in New York right now and his hike >from the hotel to the Google site is 20 minutes. He tried his >California shoes the first two days and wore blisters in his feet so >now he has dressed down to tennis shoes in a city where they all >dress up compared to the other coast. Having a walking perspective >of the traffic flow in that part of NY is different way of looking >at this problem. I assume there is not a subway to hop on if he is >hiking 20 full minutes in that part of New York. I guess the subway >is only for some subset of the city.
>I would love to build a 180 mph guideway between the new Google New >York office and reasonably priced apartments. My son said one of >his co-workers who is there pays $3,000 a month for his little >apartment and walks further than his hotel at 20 minutes. I paid >$9,000 for my first mansion just out of school. (in a federal >target area in south central Fort Worth)
> Just because Google was formed by a couple of computer nerds who > worked out a means to build a better internet search engine and > were able cash in on it in the marketplace does not necessarily > mean that all of the people subsequently employed by and now run > Google have anything close to the right answers for most other > problems as is illustrated by the situation that your son has > discovered in NYC. (I don't mean to cast an doubt on your son's capability.)
> I am afraid that the same goes for whoever at Google cooked up > that self-driving car scheme. It sounds like a great concept until > you wake up from your dream and spend about a minute of fully-awake > time trying to think through the idea. Perhaps this fellow would > do much better at working on something else on his computer.
Judging by the way he perked up when asked about the Google car, I think that Larry Page has been a driving force behind the development effort. Perhaps it would help if KH would spend some time getting acquainted with LP and the way he thinks. http://www.charlierose.com/guest/view/2594
On May 23, 2012, at 11:06 AM, Jerry Schneider wrote:
> Judging by the way he perked up when asked about the Google car, I
> think that Larry Page has been a driving force behind the
> development effort. Perhaps it would help if KH would spend some
> time getting acquainted with LP and the way he thinks.
I presume that I am the KH that you refer to. If so, I doubt it
because we obviously live in different worlds. (We may not even speak
the same language.) I have enough trouble with mid-western English. I may not have worked for the great Googe, but I have a lot of years
experience developing advanced systems using computers, including the
means that modern, distributed computers communicate and are programmed.
I also am beginning to accumulate a significant amount of experience
in developing guideway-based vehicles wherein the computing
requirements are relatively modest but by the time that you develop
systems with the essential levels of automated self-testing and
failure tolerance to make them safe for human use, you find a lot of
hurdles working with only a single dimension wherein you have near
absolute control over the operating environment. If you add all of
the other dimensions and multitude of environmental factors, the
problem becomes impractical to solve. Not only that, if you manage to
do it, just how much have you contributed to solving the problems of
congestion and air pollution caused by the current set of automobiles
and big trucks?
Kirston, I am with you on this one. The best example is this, and I must explain why I say it. My background is mostly aviation, and I am familiar with the automated systems on aircraft. Having said that, here is what I feel about taking a trip on a plane that is fully automated, that is NO Pilot.
I would have to be 'Nuts". Computers can be programmed to deal with situations that are known or at least anticipated. With every emergency that happens, there is always some factor that was unusual: something happened that both the builder and the operator did not anticipate. This means that the computer would not be programmed to handle it either.
The same applies to cars. With over sixty years of driving all over North America you might think that I would have seen everything bad that can happen on the road. Think again. Every week somebody out there does something different from what I have ever seen before.
So: Since I can only program a computer to deal with problems that I know, or even anticipate, how do I program it to deal with the ones that I do not know about, or one that has never happened before?
>To: transport-innovators@googlegroups.com >Sent: Thursday, May 24, 2012 12:48:20 AM >Subject: Re: [t-i] Larry Page, Google CEO, was the guest today on Charlie Rose show
>On May 23, 2012, at 11:06 AM, Jerry Schneider wrote:
>> Judging by the way he perked up when asked about the Google car, I think that Larry Page has been a driving force behind the development effort. Perhaps it would help if KH would spend some time getting acquainted with LP and the way he thinks.
> I presume that I am the KH that you refer to. If so, I doubt it because we obviously live in different worlds. (We may not even speak the same language.) I have enough trouble with mid-western English. I may not have worked for the great Googe, but I have a lot of years experience developing advanced systems using computers, including the means that modern, distributed computers communicate and are programmed.
> I also am beginning to accumulate a significant amount of experience in developing guideway-based vehicles wherein the computing requirements are relatively modest but by the time that you develop systems with the essential levels of automated self-testing and failure tolerance to make them safe for human use, you find a lot of hurdles working with only a single dimension wherein you have near absolute control over the operating environment. If you add all of the other dimensions and multitude of environmental factors, the problem becomes impractical to solve. Not only that, if you manage to do it, just how much have you contributed to solving the problems of congestion and air pollution caused by the current set of automobiles and big trucks?
>Kirston Henderson
>--You received this message because you are subscribed to the Google Groups "transport-innovators" group. >To post to this group, send email to transport-innovators@googlegroups.com. >To unsubscribe from this group, send email to transport-innovators+unsubscribe@googlegroups.com. >For more options, visit this group at http://groups.google.com/group/transport-innovators?hl=en.
I think that the google approach is to slow down or stop if faced with something unanticipated, perhaps demanding driver intervention. It's harder to just stop in an airplane, but acceptable (to some degree) on the road. It may learn/record the correct response given the circumstance in the future. If cars are made to communicate with Google (instant tracking, crowd sourcing), reviewed interventions can be propagated to other cars in the same situation.
I know everyone here wants it to fail and don't want to hear different, but it will probably "work" in the statistical sense where fatalities match or better human fatalities. It's a shame this is what will be acceptable when near zero fatalities, no emissions, greater throughput and less overall cost would be possible with guideway based systems.
On Thursday, 24 May 2012 02:32:31 UTC-4, Jack Slade wrote:
> Kirston, I am with you on this one. The best example is this, and I > must explain why I say it. My background is mostly aviation, and I am > familiar with the automated systems on aircraft. Having said that, here is > what I feel about taking a trip on a plane that is fully automated, that > is NO Pilot.
> I would have to be 'Nuts". Computers can be programmed to deal with > situations that are known or at least anticipated. With every emergencythat happens, there is always some factor that was unusual: something > happened that both the builder and the operator did not anticipate. This > means that the computer would not be programmed to handle it either.
> The same applies to cars. With over sixty years of driving all over North > America you might think that I would have seen everything bad that can > happen on the road. Think again. Every week somebody out there does > something different from what I have ever seen before.
> So: Since I can only program a computer to deal with problems that I > know, or even anticipate, how do I program it to deal with the ones that I > do not know about, or one that has never happened before?
> Jack Slade
> *From:* Kirston Henderson > *To:* transport-innovators@googlegroups.com > *Sent:* Thursday, May 24, 2012 12:48:20 AM > *Subject:* Re: [t-i] Larry Page, Google CEO, was the guest today on > Charlie Rose show
> On May 23, 2012, at 11:06 AM, Jerry Schneider wrote:
> > Judging by the way he perked up when asked about the Google car, I think > that Larry Page has been a driving force behind the development effort. > Perhaps it would help if KH would spend some time getting acquainted with > LP and the way he thinks.
> I presume that I am the KH that you refer to. If so, I doubt it > because we obviously live in different worlds. (We may not even speak the > same language.) I have enough trouble with mid-western English. I may not > have worked for the great Googe, but I have a lot of years experience > developing advanced systems using computers, including the means that > modern, distributed computers communicate and are programmed.
> I also am beginning to accumulate a significant amount of experience > in developing guideway-based vehicles wherein the computing requirements > are relatively modest but by the time that you develop systems with the > essential levels of automated self-testing and failure tolerance to make > them safe for human use, you find a lot of hurdles working with only a > single dimension wherein you have near absolute control over the operating > environment. If you add all of the other dimensions and multitude of > environmental factors, the problem becomes impractical to solve. Not only > that, if you manage to do it, just how much have you contributed to solving > the problems of congestion and air pollution caused by the current set of > automobiles and big trucks?
> Kirston Henderson
> --You received this message because you are subscribed to the Google > Groups "transport-innovators" group. > To post to this group, send email to transport-innovators@googlegroups.com > . > To unsubscribe from this group, send email to transport-innovators+ > unsubscribe@googlegroups.com. > For more options, visit this group at > http://groups.google.com/group/transport-innovators?hl=en.
I don't want to be grouped with "everyone here wants it to fail". That is
a broad brush. I believe that the high end computing can easily solve
simple mechanical motion problems well enough to "work". I believe that
they will do better than human drivers not because the human driver is that
bad of a controller* when paying attention.* The data suggests that an
attentive driver is an incredibly good driver. It is distractions that
turn a human driver into junk. That said my objection was of my
reservations of the litigation associated with computer taking control will
bring on massive judgments against the company who dares. If my little
company were to take on the ambulance chasers I would lose. Perhaps a
powerhouse with big deep pockets could take it on. Moving a wafer through
a semiconductor plant is not much different than moving a few cars down a
few roads from the computer's perspective. It has been done for quite some
time and computers are significantly faster than anything mechanical.
I would like to mention out of order that DM is not just putting regular
cars on a transport truck. DM is about two modes (or more) roadway and
guideway. That does not have any bearing on the weight of the guideway car
unless you look at a single design.
Back on topic I believe that self-driving cars are perfect for parking and
that is what is already on the market being sold today but in a bare bones
form. It is nothing to expand the parallel parking task to parking on the
roof so I think it is a given to be able to remote park cars autonomously.
That part of self-driving is an enabling technology for TriTrack's vision
of how you park your private individual seatcovers and associated cup
holder. (Your private space and personal clutter). With the ability to
self-drive you transform the architecture of the buildings we all use.
Back in the days of horse drawn carriages the kitchen often had raised
floors that were dock high to let you leave the carriage and step into the
home easily and it kept the mud out of the passenger path. Additionally it
had a cover above so the rain was no longer getting you wet moving from
your carriage to the house. It is easy to date a home built before cars
and after. These entrances show the age of the antique home. Same thing
will happen with new houses as the entrance to the home will return to
these porticoes and the car will go park itself wherever it can talking
with all other cars parked in the vicinity. Packing density of parked cars
is what self-driving cars can do right off the bat. No need to open the
doors taking up expensive real estate. The cars can park inches from each
other and return even if one car is blocking another under present
conditions of human parking.
While on the guideway I think we are all eventually talking about
self-driving cars. I think Kirston is thinking second generation but still
in the plan. The difference is the road portion and the increased
complexity and litigation that it brings. I think our little group should
embrace the idea of computer capability and recognize that the hardware and
software and do this in the next few months. The bigger task is outside
the computer field. Same with PRT DM whatever. It is not the hardware or
software that is the issue at all but marketing and sales of the concepts.
I use that term in the broadest of context.
I believe it will take all of the above to get the nation's economy healthy
and thriving. Guideway and advanced computer control. Just my take.
On Thu, May 24, 2012 at 7:59 AM, eph <rhapsodi...@yahoo.com> wrote:
> I think that the google approach is to slow down or stop if faced with
> something unanticipated, perhaps demanding driver intervention. It's
> harder to just stop in an airplane, but acceptable (to some degree) on the
> road. It may learn/record the correct response given the circumstance in
> the future. If cars are made to communicate with Google (instant tracking,
> crowd sourcing), reviewed interventions can be propagated to other cars in
> the same situation.
> I know everyone here wants it to fail and don't want to hear different,
> but it will probably "work" in the statistical sense where fatalities match
> or better human fatalities. It's a shame this is what will be acceptable
> when near zero fatalities, no emissions, greater throughput and less
> overall cost would be possible with guideway based systems.
> F.
> On Thursday, 24 May 2012 02:32:31 UTC-4, Jack Slade wrote:
>> Kirston, I am with you on this one. The best example is this, and I
>> must explain why I say it. My background is mostly aviation, and I am
>> familiar with the automated systems on aircraft. Having said that, here is
>> what I feel about taking a trip on a plane that is fully automated, that
>> is NO Pilot.
>> I would have to be 'Nuts". Computers can be programmed to deal with
>> situations that are known or at least anticipated. With every emergencythat happens, there is always some factor that was unusual: something
>> happened that both the builder and the operator did not anticipate. This
>> means that the computer would not be programmed to handle it either.
>> The same applies to cars. With over sixty years of driving all over
>> North America you might think that I would have seen everything bad that
>> can happen on the road. Think again. Every week somebody out there does
>> something different from what I have ever seen before.
>> So: Since I can only program a computer to deal with problems that I
>> know, or even anticipate, how do I program it to deal with the ones that I
>> do not know about, or one that has never happened before?
>> Jack Slade
>> *From:* Kirston Henderson
>> *To:* transport-innovators@**googlegroups.com<transport-innovators@googlegroups.c om>
>> *Sent:* Thursday, May 24, 2012 12:48:20 AM
>> *Subject:* Re: [t-i] Larry Page, Google CEO, was the guest today on
>> Charlie Rose show
>> On May 23, 2012, at 11:06 AM, Jerry Schneider wrote:
>> > Judging by the way he perked up when asked about the Google car, I
>> think that Larry Page has been a driving force behind the development
>> effort. Perhaps it would help if KH would spend some time getting
>> acquainted with LP and the way he thinks.
>> I presume that I am the KH that you refer to. If so, I doubt it
>> because we obviously live in different worlds. (We may not even speak the
>> same language.) I have enough trouble with mid-western English. I may not
>> have worked for the great Googe, but I have a lot of years experience
>> developing advanced systems using computers, including the means that
>> modern, distributed computers communicate and are programmed.
>> I also am beginning to accumulate a significant amount of experience
>> in developing guideway-based vehicles wherein the computing requirements
>> are relatively modest but by the time that you develop systems with the
>> essential levels of automated self-testing and failure tolerance to make
>> them safe for human use, you find a lot of hurdles working with only a
>> single dimension wherein you have near absolute control over the operating
>> environment. If you add all of the other dimensions and multitude of
>> environmental factors, the problem becomes impractical to solve. Not only
>> that, if you manage to do it, just how much have you contributed to solving
>> the problems of congestion and air pollution caused by the current set of
>> automobiles and big trucks?
>> Kirston Henderson
>> --You received this message because you are subscribed to the Google
>> Groups "transport-innovators" group.
>> To post to this group, send email to transport-innovators@**
>> googlegroups.com <transport-innovators@googlegroups.com>.
>> To unsubscribe from this group, send email to transport-innovators+
>> unsubscri**be@googlegroups.com <unsubscribe@googlegroups.com>.
>> For more options, visit this group at http://groups.google.com/** >> group/transport-innovators?hl=**en<http://groups.google.com/group/transport-innovators?hl=en>
>> .
> To post to this group, send email to transport-innovators@googlegroups.com
> .
> To unsubscribe from this group, send email to
> transport-innovators+unsubscribe@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/transport-innovators?hl=en.
You seem to be asking for perfection. With the criteria you suggest how can PRT ever be good enough?
Dennis
From: Jack Slade Sent: Wednesday, May 23, 2012 11:32 PM
To: transport-innovators@googlegroups.com Subject: Re: [t-i] Larry Page, Google CEO, was the guest today on Charlie Rose show
Kirston, I am with you on this one. The best example is this, and I must explain why I say it. My background is mostly aviation, and I am familiar with the automated systems on aircraft. Having said that, here is what I feel about taking a trip on a plane that is fully automated, that is NO Pilot.
I would have to be 'Nuts". Computers can be programmed to deal with situations that are known or at least anticipated. With every emergency that happens, there is always some factor that was unusual: something happened that both the builder and the operator did not anticipate. This means that the computer would not be programmed to handle it either.
The same applies to cars. With over sixty years of driving all over North America you might think that I would have seen everything bad that can happen on the road. Think again. Every week somebody out there does something different from what I have ever seen before.
So: Since I can only program a computer to deal with problems that I know, or even anticipate, how do I program it to deal with the ones that I do not know about, or one that has never happened before?
Jack Slade
From: Kirston Henderson <kirston.hender...@megarail.com>
To: transport-innovators@googlegroups.com Sent: Thursday, May 24, 2012 12:48:20 AM
Subject: Re: [t-i] Larry Page, Google CEO, was the guest today on Charlie Rose show
On May 23, 2012, at 11:06 AM, Jerry Schneider wrote:
> Judging by the way he perked up when asked about the Google car, I think that Larry Page has been a driving force behind the development effort. Perhaps it would help if KH would spend some time getting acquainted with LP and the way he thinks.
I presume that I am the KH that you refer to. If so, I doubt it because we obviously live in different worlds. (We may not even speak the same language.) I have enough trouble with mid-western English. I may not have worked for the great Googe, but I have a lot of years experience developing advanced systems using computers, including the means that modern, distributed computers communicate and are programmed.
I also am beginning to accumulate a significant amount of experience in developing guideway-based vehicles wherein the computing requirements are relatively modest but by the time that you develop systems with the essential levels of automated self-testing and failure tolerance to make them safe for human use, you find a lot of hurdles working with only a single dimension wherein you have near absolute control over the operating environment. If you add all of the other dimensions and multitude of environmental factors, the problem becomes impractical to solve. Not only that, if you manage to do it, just how much have you contributed to solving the problems of congestion and air pollution caused by the current set of automobiles and big trucks?
Kirston Henderson
--You received this message because you are subscribed to the Google Groups "transport-innovators" group.
To post to this group, send email to transport-innovators@googlegroups.com.
To unsubscribe from this group, send email to transport-innovators+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/transport-innovators?hl=en.
-- You received this message because you are subscribed to the Google Groups "transport-innovators" group.
To post to this group, send email to transport-innovators@googlegroups.com.
To unsubscribe from this group, send email to transport-innovators+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/transport-innovators?hl=en.
Please don't put me in the "he wishes it will fail column". I just think it's more difficult to do than say Brad Templeton professes, and that it will take much longer to implement.
Dennis
From: eph Sent: Thursday, May 24, 2012 5:59 AM
To: transport-innovators@googlegroups.com Subject: Re: [t-i] Larry Page, Google CEO, was the guest today on Charlie Rose show
I think that the google approach is to slow down or stop if faced with something unanticipated, perhaps demanding driver intervention. It's harder to just stop in an airplane, but acceptable (to some degree) on the road. It may learn/record the correct response given the circumstance in the future. If cars are made to communicate with Google (instant tracking, crowd sourcing), reviewed interventions can be propagated to other cars in the same situation.
I know everyone here wants it to fail and don't want to hear different, but it will probably "work" in the statistical sense where fatalities match or better human fatalities. It's a shame this is what will be acceptable when near zero fatalities, no emissions, greater throughput and less overall cost would be possible with guideway based systems.
F.
On Thursday, 24 May 2012 02:32:31 UTC-4, Jack Slade wrote:
Kirston, I am with you on this one. The best example is this, and I must explain why I say it. My background is mostly aviation, and I am familiar with the automated systems on aircraft. Having said that, here is what I feel about taking a trip on a plane that is fully automated, that is NO Pilot.
I would have to be 'Nuts". Computers can be programmed to deal with situations that are known or at least anticipated. With every emergency that happens, there is always some factor that was unusual: something happened that both the builder and the operator did not anticipate. This means that the computer would not be programmed to handle it either.
The same applies to cars. With over sixty years of driving all over North America you might think that I would have seen everything bad that can happen on the road. Think again. Every week somebody out there does something different from what I have ever seen before.
So: Since I can only program a computer to deal with problems that I know, or even anticipate, how do I program it to deal with the ones that I do not know about, or one that has never happened before?
Jack Slade
From: Kirston Henderson To: transport-innovators@googlegroups.com Sent: Thursday, May 24, 2012 12:48:20 AM
Subject: Re: [t-i] Larry Page, Google CEO, was the guest today on Charlie Rose show
On May 23, 2012, at 11:06 AM, Jerry Schneider wrote:
> Judging by the way he perked up when asked about the Google car, I think that Larry Page has been a driving force behind the development effort. Perhaps it would help if KH would spend some time getting acquainted with LP and the way he thinks.
I presume that I am the KH that you refer to. If so, I doubt it because we obviously live in different worlds. (We may not even speak the same language.) I have enough trouble with mid-western English. I may not have worked for the great Googe, but I have a lot of years experience developing advanced systems using computers, including the means that modern, distributed computers communicate and are programmed.
I also am beginning to accumulate a significant amount of experience in developing guideway-based vehicles wherein the computing requirements are relatively modest but by the time that you develop systems with the essential levels of automated self-testing and failure tolerance to make them safe for human use, you find a lot of hurdles working with only a single dimension wherein you have near absolute control over the operating environment. If you add all of the other dimensions and multitude of environmental factors, the problem becomes impractical to solve. Not only that, if you manage to do it, just how much have you contributed to solving the problems of congestion and air pollution caused by the current set of automobiles and big trucks?
Kirston Henderson
--You received this message because you are subscribed to the Google Groups "transport-innovators" group.
To post to this group, send email to transport-innovators@googlegroups.com.
To unsubscribe from this group, send email to transport-innovators+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/transport-innovators?hl=en.
>I think that the google approach is to slow down or stop if faced >with something unanticipated, perhaps demanding driver >intervention. It's harder to just stop in an airplane, but >acceptable (to some degree) on the road. It may learn/record the >correct response given the circumstance in the future. If cars are >made to communicate with Google (instant tracking, crowd sourcing), >reviewed interventions can be propagated to other cars in the same situation.
>I know everyone here wants it to fail and don't want to hear >different, but it will probably "work" in the statistical sense >where fatalities match or better human fatalities. It's a shame >this is what will be acceptable when near zero fatalities, no >emissions, greater throughput and less overall cost would be >possible with guideway based systems.
I'd be interested to know more about why you think "less overall cost" - as it seems to me that lower cost (i.e. no guideway required) is regarded as a very prominent reason that its supporters always cite up front.
>Please don't put me in the "he wishes it will fail column". I just >think it's more difficult to do than say Brad Templeton professes, >and that it will take much longer to implement.
If it is implemented in phases, a little at a time, as I suspect it will (must) - how long will it take to get some experience with Phase I?
--------------------------------------------------
From: "Jerry Schneider" <j...@peak.org>
Sent: Thursday, May 24, 2012 9:14 AM
To: <transport-innovators@googlegroups.com>
Subject: Re: [t-i] Larry Page, Google CEO, was the guest today on Charlie Rose show
> At 08:34 AM 5/24/2012, you wrote:
>>Please don't put me in the "he wishes it will fail column". I just think >>it's more difficult to do than say Brad Templeton professes, and that it >>will take much longer to implement.
> If it is implemented in phases, a little at a time, as I suspect it will > (must) - how long will it take to get some experience with Phase I?
> -- > You received this message because you are subscribed to the Google Groups > "transport-innovators" group.
> To post to this group, send email to > transport-innovators@googlegroups.com.
> To unsubscribe from this group, send email to > transport-innovators+unsubscribe@googlegroups.com.
> For more options, visit this group at > http://groups.google.com/group/transport-innovators?hl=en.
Not perfection, but close to it. To do this for PRT is not easy...nobody ever said it was...but it has far fewer problems to deal with than a road vehicle has. Problems on a guideway are easier to define, and are not subject to the same variations as roads.
Would I like to be able to tell my car where to take me today, and to go park itself when I get there? You bet I would. However, I am not in such a frenzy of anticipation that I am willing ignore obstacles that are real, in the hope that future technology will soon solve them.
I suggest you pose a couple of safety questions to these people. 1. Have you given any thought to what laser and radar pulses may do to the eyes of perestrians over a long period of time?
I have posted that question here on the list, and everybody ignores it. Why?
>To: transport-innovators@googlegroups.com >Sent: Thursday, May 24, 2012 11:29:13 AM >Subject: Re: [t-i] Larry Page, Google CEO, was the guest today on Charlie Rose show
>You seem to be asking for perfection. With the criteria you suggest how can PRT ever be good enough? > >Dennis
>From: Jack Slade >Sent: Wednesday, May 23, 2012 11:32 PM >To: transport-innovators@googlegroups.com >Subject: Re: [t-i] Larry Page, Google CEO, was the guest today on Charlie Rose show
>Kirston, I am with you on this one. The best example is this, and I must explain why I say it. My background is mostly aviation, and I am familiar with the automated systems on aircraft. Having said that, here is what I feel about taking a trip on a plane that is fully automated, that is NO Pilot. > >I would have to be 'Nuts". Computers can be programmed to deal with situations that are known or at least anticipated. With every emergency that happens, there is always some factor that was unusual: something happened that both the builder and the operator did not anticipate. This means that the computer would not be programmed to handle it either. > >The same applies to cars. With over sixty years of driving all over North America you might think that I would have seen everything bad that can happen on the road. Think again. Every week somebody out there does something different from what I have ever seen before. > >So: Since I can only program a computer to deal with problems that I know, or even anticipate, how do I program it to deal with the ones that I do not know about, or one that has never happened before? > >Jack Slade
>From: Kirston Henderson <kirston.hender...@megarail.com> >>To: transport-innovators@googlegroups.com >>Sent: Thursday, May 24, 2012 12:48:20 AM >>Subject: Re: [t-i] Larry Page, Google CEO, was the guest today on Charlie Rose show
>>On May 23, 2012, at 11:06 AM, Jerry Schneider wrote:
>>> Judging by the way he perked up when asked about the Google car, I think that Larry Page has been a driving force behind the development effort. Perhaps it would help if KH would spend some time getting acquainted with LP and the way he thinks.
>> I presume that I am the KH that you refer to. If so, I doubt it because we obviously live in different worlds. (We may not even speak the same language.) I have enough trouble with mid-western English. I may not have worked for the great Googe, but I have a lot of years experience developing advanced systems using computers, including the means that modern, distributed computers communicate and are programmed.
>> I also am beginning to accumulate a significant amount of experience in developing guideway-based vehicles wherein the computing requirements are relatively modest but by the time that you develop systems with the essential levels of automated self-testing and failure tolerance to make them safe for human use, you find a lot of hurdles working with only a single dimension wherein you have near absolute control over the operating environment. If you add all of the other dimensions and multitude of environmental factors, the problem becomes impractical to solve. Not only that, if you manage to do it, just how much have you contributed to solving the problems of congestion and air pollution caused by the current set of automobiles and big trucks?
>>Kirston Henderson
>>--You received this message because you are subscribed to the Google Groups "transport-innovators" group. >>To post to this group, send email to transport-innovators@googlegroups.com. >>To unsubscribe from this group, send email to transport-innovators+unsubscribe@googlegroups.com. >>For more options, visit this group at http://groups.google.com/group/transport-innovators?hl=en.
>>-- >You received this message because you are subscribed to the Google Groups "transport-innovators" group. >To post to this group, send email to transport-innovators@googlegroups.com. >To unsubscribe from this group, send email to transport-innovators+unsubscribe@googlegroups.com. >For more options, visit this group at http://groups.google.com/group/transport-innovators?hl=en. >-- >You received this message because you are subscribed to the Google Groups "transport-innovators" group. >To post to this group, send email to transport-innovators@googlegroups.com. >To unsubscribe from this group, send email to transport-innovators+unsubscribe@googlegroups.com. >For more options, visit this group at http://groups.google.com/group/transport-innovators?hl=en.
Here's a question (Probably meaningless in the long run), if one were to pick a trip in an automated vehicle, what vehicle would you choose?
Obviously, most, if not all, would pick a trip on a dedicated guideway, like PRT. But, my question would give a person only 2 choices, an aircraft or a robocar.
I'll start off with considerations;
* Weather
* Speed
* Drunks and other operators
* Number of dimensions to use (3 vs. 2)
* Type of limiting and controlling organizations
* Inspecting and licensing
* Condition of equipment
Any more considerations? Why would your choice be best?
> Kirston, I am with you on this one. The best example is this, and I > must explain why I say it. My background is mostly aviation, and I > am familiar with the automated systems on aircraft. Having said that, > here is what I feel about taking a trip on a plane that is fully > automated, that is NO Pilot.
> I would have to be 'Nuts". Computers can be programmed to deal with > situations that are known or at least anticipated. With > every emergencythat happens, there is always some factor that was > unusual: something happened that both the builder and the operator > did not anticipate. This means that the computer would not be > programmed to handle it either.
> The same applies to cars. With over sixty years of driving all over > North America you might think that I would have seen everything bad > that can happen on the road. Think again. Every week somebody out > there does something different from what I have ever seen before.
> So: Since I can only program a computer to deal with problems that I > know, or even anticipate, how do I program it to deal with the ones > that I do not know about, or one that has never happened before?
> Jack Slade
> *From:* Kirston Henderson <kirston.hender...@megarail.com>
> *To:* transport-innovators@googlegroups.com
> *Sent:* Thursday, May 24, 2012 12:48:20 AM
> *Subject:* Re: [t-i] Larry Page, Google CEO, was the guest today
> on Charlie Rose show
> On May 23, 2012, at 11:06 AM, Jerry Schneider wrote:
> > Judging by the way he perked up when asked about the Google car,
> I think that Larry Page has been a driving force behind the
> development effort. Perhaps it would help if KH would spend some
> time getting acquainted with LP and the way he thinks.
> I presume that I am the KH that you refer to. If so, I doubt
> it because we obviously live in different worlds. (We may not
> even speak the same language.) I have enough trouble with
> mid-western English. I may not have worked for the great Googe,
> but I have a lot of years experience developing advanced systems
> using computers, including the means that modern, distributed
> computers communicate and are programmed.
> I also am beginning to accumulate a significant amount of
> experience in developing guideway-based vehicles wherein the
> computing requirements are relatively modest but by the time that
> you develop systems with the essential levels of automated
> self-testing and failure tolerance to make them safe for human
> use, you find a lot of hurdles working with only a single
> dimension wherein you have near absolute control over the
> operating environment. If you add all of the other dimensions and
> multitude of environmental factors, the problem becomes
> impractical to solve. Not only that, if you manage to do it, just
> how much have you contributed to solving the problems of
> congestion and air pollution caused by the current set of
> automobiles and big trucks?
>Here's a question (Probably meaningless in the long run), if one >were to pick a trip in an automated vehicle, what vehicle would you choose?
>Obviously, most, if not all, would pick a trip on a dedicated >guideway, like PRT. But, my question would give a person only 2 >choices, an aircraft or a robocar.
>I'll start off with considerations;
> * Weather
> * Speed
> * Drunks and other operators
> * Number of dimensions to use (3 vs. 2)
> * Type of limiting and controlling organizations
> * Inspecting and licensing
> * Condition of equipment
>Any more considerations? Why would your choice be best?
Good question. Without stating your assumptions about the nature of the trip, it's a difficult question to answer. If you assume that the deployment of robocars at development stage X will be "everywhere" there are a large number of factors to consider, not all of them anticipated and programmed. I expect to see a phased deployment and considerable evolution of robocars over time that will provide whatever limiting conditions or regulations need to be adopted, probably dictated by some unfortunate events. That said, I generally favor man/machine combinations (partnerships) as I think they are likely to be superior, overall, for some time yet.
I think that if you roll in the environmental cost, health care costs, insurance costs, fuel costs, vehicle costs, maintenance cost, etc. AND extra infrastructure to handle all these robocars, a case can be made for guideway + cheaper, smaller, shorter-range electric cars (DM). Unfortunately, many of these impacts aren't reflected in the perceived cost of car driving. My statement was "overall cost" - which includes societal costs as I see it. Google has made selling DM that much more difficult by pushing it's vision that is seen as "good enough". That's a shame.
On Thursday, 24 May 2012 11:44:42 UTC-4, jschneider wrote:
> At 05:59 AM 5/24/2012, you wrote: > >I think that the google approach is to slow down or stop if faced > >with something unanticipated, perhaps demanding driver > >intervention. It's harder to just stop in an airplane, but > >acceptable (to some degree) on the road. It may learn/record the > >correct response given the circumstance in the future. If cars are > >made to communicate with Google (instant tracking, crowd sourcing), > >reviewed interventions can be propagated to other cars in the same > situation.
> >I know everyone here wants it to fail and don't want to hear > >different, but it will probably "work" in the statistical sense > >where fatalities match or better human fatalities. It's a shame > >this is what will be acceptable when near zero fatalities, no > >emissions, greater throughput and less overall cost would be > >possible with guideway based systems.
> I'd be interested to know more about why you think "less overall > cost" - as it seems to me that lower cost (i.e. no guideway required) > is regarded as a very prominent reason that its supporters always > cite up front.
>I think that if you roll in the environmental cost, health care >costs, insurance costs, fuel costs, vehicle costs, maintenance cost, >etc. AND extra infrastructure to handle all these robocars, a case >can be made for guideway + cheaper, smaller, shorter-range electric >cars (DM). Unfortunately, many of these impacts aren't reflected in >the perceived cost of car driving. My statement was "overall cost" >- which includes societal costs as I see it. Google has made >selling DM that much more difficult by pushing it's vision that is >seen as "good enough". That's a shame.
Fair enough - if you include the potential savings that might arise from greatly improved safety and mobility services to groups that would benefit from it (elderly non-drives, blind people, drunk driving reductions, and so on). I would hope that someone on the Google team is doing such calculations on societal impact issues. So far, I've not see much on the subject with the exception of Tyler Folsom's excellent paper on the subject.
Link to paper entitled<http://faculty.washington.edu/jbs/itrans/tyler-article.pdf> Social Ramifications of Autonomous Urban Land Vehicles, by Tyler C. Folsom (posted in 2010, my What's New page).
>On Thursday, 24 May 2012 11:44:42 UTC-4, jschneider wrote:
>At 05:59 AM 5/24/2012, you wrote:
> >I think that the google approach is to slow down or stop if faced
> >with something unanticipated, perhaps demanding driver
> >intervention. It's harder to just stop in an airplane, but
> >acceptable (to some degree) on the road. It may learn/record the
> >correct response given the circumstance in the future. If cars are
> >made to communicate with Google (instant tracking, crowd sourcing),
> >reviewed interventions can be propagated to other cars in the same > situation.
> >I know everyone here wants it to fail and don't want to hear
> >different, but it will probably "work" in the statistical sense
> >where fatalities match or better human fatalities. It's a shame
> >this is what will be acceptable when near zero fatalities, no
> >emissions, greater throughput and less overall cost would be
> >possible with guideway based systems.
>I'd be interested to know more about why you think "less overall
>cost" - as it seems to me that lower cost (i.e. no guideway required)
>is regarded as a very prominent reason that its supporters always
>cite up front.
>- Jerry Schneider -
> Innovative Transportation Technologies
100 years from now,if Jack's statement is published in the "100 years ago" section of Aviation week, what wold be the reaction?
I realize one definition of a piolot;s life is"Hours and hourd of bordom, with a few cases of sheer terror in between".
But what about real time ground response to automated passenger flight problems; and maybe all landings to start?
In flight anomoly allerts the ground "pilot" who manages the correction with what is available, and if nesessart initiates emergency automated landing.
Could such hamdle the landing in the Hudson?
In the safety comparison, how do you count accidents caused by the on-board pilot? The Buffalo crash couple years was cased by the pilot and would not have happened under automatic control.
----- Original Message ----- From: Richard Gronning To: transport-innovators@googlegroups.com Sent: Thursday, May 24, 2012 2:07 PM
Subject: Re: [t-i] Larry Page, Google CEO, was the guest today on Charlie Rose show
Here's a question (Probably meaningless in the long run), if one were to pick a trip in an automated vehicle, what vehicle would you choose?
Obviously, most, if not all, would pick a trip on a dedicated guideway, like PRT. But, my question would give a person only 2 choices, an aircraft or a robocar.
I'll start off with considerations;
a.. Weather b.. Speed
c.. Drunks and other operators
d.. Number of dimensions to use (3 vs. 2) e.. Type of limiting and controlling organizations f.. Inspecting and licensing g.. Condition of equipment Any more considerations? Why would your choice be best?
Dick
On 5/24/2012 1:32 AM, Jack Slade wrote: Kirston, I am with you on this one. The best example is this, and I must explain why I say it. My background is mostly aviation, and I am familiar with the automated systems on aircraft. Having said that, here is what I feel about taking a trip on a plane that is fully automated, that is NO Pilot.
I would have to be 'Nuts". Computers can be programmed to deal with situations that are known or at least anticipated. With every emergency that happens, there is always some factor that was unusual: something happened that both the builder and the operator did not anticipate. This means that the computer would not be programmed to handle it either.
The same applies to cars. With over sixty years of driving all over North America you might think that I would have seen everything bad that can happen on the road. Think again. Every week somebody out there does something different from what I have ever seen before.
So: Since I can only program a computer to deal with problems that I know, or even anticipate, how do I program it to deal with the ones that I do not know about, or one that has never happened before?
Jack Slade
From: Kirston Henderson <kirston.hender...@megarail.com>
To: transport-innovators@googlegroups.com Sent: Thursday, May 24, 2012 12:48:20 AM
Subject: Re: [t-i] Larry Page, Google CEO, was the guest today on Charlie Rose show
On May 23, 2012, at 11:06 AM, Jerry Schneider wrote:
> Judging by the way he perked up when asked about the Google car, I think that Larry Page has been a driving force behind the development effort. Perhaps it would help if KH would spend some time getting acquainted with LP and the way he thinks.
I presume that I am the KH that you refer to. If so, I doubt it because we obviously live in different worlds. (We may not even speak the same language.) I have enough trouble with mid-western English. I may not have worked for the great Googe, but I have a lot of years experience developing advanced systems using computers, including the means that modern, distributed computers communicate and are programmed.
I also am beginning to accumulate a significant amount of experience in developing guideway-based vehicles wherein the computing requirements are relatively modest but by the time that you develop systems with the essential levels of automated self-testing and failure tolerance to make them safe for human use, you find a lot of hurdles working with only a single dimension wherein you have near absolute control over the operating environment. If you add all of the other dimensions and multitude of environmental factors, the problem becomes impractical to solve. Not only that, if you manage to do it, just how much have you contributed to solving the problems of congestion and air pollution caused by the current set of automobiles and big trucks?
Kirston Henderson
-- You received this message because you are subscribed to the Google Groups "transport-innovators" group.
To post to this group, send email to transport-innovators@googlegroups.com.
To unsubscribe from this group, send email to transport-innovators+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/transport-innovators?hl=en.
>On May 23, 2012, at 11:06 AM, Jerry Schneider wrote:
>>Judging by the way he perked up when asked about the Google car, I
>>think that Larry Page has been a driving force behind the
>>development effort. Perhaps it would help if KH would spend some
>>time getting acquainted with LP and the way he thinks.
> I presume that I am the KH that you refer to. If so, I doubt it
>because we obviously live in different worlds. (We may not even speak
>the same language.) I have enough trouble with mid-western English.
>I may not have worked for the great Googe, but I have a lot of years
>experience developing advanced systems using computers, including the
>means that modern, distributed computers communicate and are programmed.
> I also am beginning to accumulate a significant amount of > experience
>in developing guideway-based vehicles wherein the computing
>requirements are relatively modest but by the time that you develop
>systems with the essential levels of automated self-testing and
>failure tolerance to make them safe for human use, you find a lot of
>hurdles working with only a single dimension wherein you have near
>absolute control over the operating environment. If you add all of
>the other dimensions and multitude of environmental factors, the
>problem becomes impractical to solve. Not only that, if you manage to
>do it, just how much have you contributed to solving the problems of
>congestion and air pollution caused by the current set of automobiles
>and big trucks?
I have always been an advocate of "questioning authority" - it's a great way to learn.
I also like to challenge inappropriate analogies.
I think until we know more about the evolutionary path to deployment that the Google
team maps out, it will be hard to know which of our many societal problems, that are
related to our mobility desires and needs, might be targeted by their evolving product. I don't think
"solving congestion" is at the top of their list at the moment. Autonomous trucks have
not been mentioned so far as I know. Autonomous golf carts have been. I thought you
were intending to start with a human driver. Hope you leave something for that person
to do other than staying awake.
I am not sure you two guys are reading each other's posts correctly. The
present price of the Google auto-driving car is significant. There is a
hope that the astronomical hardware cost might come down by a factor of
... not-going-to-happen. If there are a billion $40,000.00 cars in the
world and you want to add a $6000 option to it that is 6 trillion dollars
as their incremental cost? Please check my math. Each car has to have a
module and the data network has to be paid for and maintained at some cost
around $100.00 per month in today dollars. TriTrack specifically is 5
pennies a mile for energy or 1 billion cars driving 14,000 miles a year.
14,000 miles equals $700/year times a billion cars is $700 billion which
is less than 6 trillion. Additionally the billion cars get 20 mpg not
TriTrack performance so the 6 trillion is just the adder for self driving
lazy driving. 14,000 miles at 20 mpg with $5.00 gas is $3500/year in
addition to the Google hardware on a billion cars. $1200 per year per car
for Internet. The world is not that rich. Gasoline and diesel in addition
to the cost of the self-drivomatic will bankrupt us period. If the
self-drivomatic was free and paid each billion user $4000/year it still
will cost more than TriTrack specifically.
The problem with "illion" math is the public has no idea what those numbers
mean. They are all just big numbers.
You do point out the externalities and they too are valid and add to the
"illion" math problem but you forgot some major ones that dwarf the ones
you listed. Same outcome.
On Thu, May 24, 2012 at 2:03 PM, eph <rhapsodi...@yahoo.com> wrote:
> I think that if you roll in the environmental cost, health care costs,
> insurance costs, fuel costs, vehicle costs, maintenance cost, etc. AND
> extra infrastructure to handle all these robocars, a case can be made for
> guideway + cheaper, smaller, shorter-range electric cars (DM).
> Unfortunately, many of these impacts aren't reflected in the perceived cost
> of car driving. My statement was "overall cost" - which includes societal
> costs as I see it. Google has made selling DM that much more difficult by
> pushing it's vision that is seen as "good enough". That's a shame.
> F.
> On Thursday, 24 May 2012 11:44:42 UTC-4, jschneider wrote:
>> At 05:59 AM 5/24/2012, you wrote:
>> >I think that the google approach is to slow down or stop if faced
>> >with something unanticipated, perhaps demanding driver
>> >intervention. It's harder to just stop in an airplane, but
>> >acceptable (to some degree) on the road. It may learn/record the
>> >correct response given the circumstance in the future. If cars are
>> >made to communicate with Google (instant tracking, crowd sourcing),
>> >reviewed interventions can be propagated to other cars in the same
>> situation.
>> >I know everyone here wants it to fail and don't want to hear
>> >different, but it will probably "work" in the statistical sense
>> >where fatalities match or better human fatalities. It's a shame
>> >this is what will be acceptable when near zero fatalities, no
>> >emissions, greater throughput and less overall cost would be
>> >possible with guideway based systems.
>> I'd be interested to know more about why you think "less overall
>> cost" - as it seems to me that lower cost (i.e. no guideway required)
>> is regarded as a very prominent reason that its supporters always
>> cite up front.
> To post to this group, send email to transport-innovators@googlegroups.com
> .
> To unsubscribe from this group, send email to
> transport-innovators+unsubscribe@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/transport-innovators?hl=en.
I would not want forecast what 100 years of tehchnological improvements may bring, and I certainly am not a pessimist. Anything that I say applies to present-day conditions, and anything new that may help us in the near future. I think it is stupid to put all resources into one technology while ignoring all others, yet this is what is happening. History repeating itself: Remember the push for fuel cells and the hydrogen technology of only 10 years ago? Where is it all now?
>To: transport-innovators@googlegroups.com >Sent: Thursday, May 24, 2012 4:15:09 PM >Subject: Re: [t-i] Larry Page, Google CEO, was the guest today on Charlie Rose show
>100 years from now,if Jack's statement is published in the "100 years ago" section of Aviation week, what wold be the reaction? > >I realize one definition of a piolot;s life is"Hours and hourd of bordom, with a few cases of sheer terror in between". > >But what about real time ground response to automated passenger flight problems; and maybe all landings to start? >In flight anomoly allerts the ground "pilot" who manages the correction with what is available, and if nesessart initiates emergency automated landing. > >Could such hamdle the landing in the Hudson? >In the safety comparison, how do you count accidents caused by the on-board pilot? The Buffalo crash couple years was cased by the pilot and would not have happened under automatic control. > >Walt Brewer >----- Original Message ----- >>From: Richard Gronning >>To: transport-innovators@googlegroups.com >>Sent: Thursday, May 24, 2012 2:07 PM >>Subject: Re: [t-i] Larry Page, Google CEO, was the guest today on Charlie Rose show
>>Here's a question (Probably meaningless in the long run), if one were to pick a trip in an automated vehicle, what vehicle would you choose?
>>Obviously, most, if not all, would pick a trip on a dedicated guideway, like PRT. But, my question would give a person only 2 choices, an aircraft or a robocar.
>>I'll start off with considerations;
>> * Weather >> * Speed
>> * Drunks and other operators
>> * Number of dimensions to use (3 vs. 2) >> * Type of limiting and controlling organizations >> * Inspecting and licensing >> * Condition of equipment >>Any more considerations? Why would your choice be best?
>>Dick
>>On 5/24/2012 1:32 AM, Jack Slade wrote: >>Kirston, I am with you on this one. The best example is this, and I must explain why I say it. My background is mostly aviation, and I am familiar with the automated systems on aircraft. Having said that, here is what I feel about taking a trip on a plane that is fully automated, that is NO Pilot. >>> >>>I would have to be 'Nuts". Computers can be programmed to deal with situations that are known or at least anticipated. With every emergency that happens, there is always some factor that was unusual: something happened that both the builder and the operator did not anticipate. This means that the computer would not be programmed to handle it either. >>> >>>The same applies to cars. With over sixty years of driving all over North America you might think that I would have seen everything bad that can happen on the road. Think again. Every week somebody out there does something different from what I have ever seen before. >>> >>>So: Since I can only program a computer to deal with problems that I know, or even anticipate, how do I program it to deal with the ones that I do not know about, or one that has never happened before? >>> >>>Jack Slade
>>>From: Kirston Henderson <kirston.hender...@megarail.com> >>>>To: transport-innovators@googlegroups.com >>>>Sent: Thursday, May 24, 2012 12:48:20 AM >>>>Subject: Re: [t-i] Larry Page, Google CEO, was the guest today on Charlie Rose show
>>>>On May 23, 2012, at 11:06 AM, Jerry Schneider wrote:
>>>>> Judging by the way he perked up when asked about the Google car, I think that Larry Page has been a driving force behind the development effort. Perhaps it would help if KH would spend some time getting acquainted with LP and the way he thinks.
>>>> I presume that I am the KH that you refer to. If so, I doubt it because we obviously live in different worlds. (We may not even speak the same language.) I have enough trouble with mid-western English. I may not have worked for the great Googe, but I have a lot of years experience developing advanced systems using computers, including the means that modern, distributed computers communicate and are programmed.
>>>> I also am beginning to accumulate a significant amount of experience in developing guideway-based vehicles wherein the computing requirements are relatively modest but by the time that you develop systems with the essential levels of automated self-testing and failure tolerance to make them safe for human use, you find a lot of hurdles working with only a single dimension wherein you have near absolute control over the operating environment. If you add all of the other dimensions and multitude of environmental factors, the problem becomes impractical to solve. Not only that, if you manage to do it, just how much have you contributed to solving the problems of congestion and air pollution caused by the current set of automobiles and big trucks?
>>>>Kirston Henderson
>>-- >>You received this message because you are subscribed to the Google Groups "transport-innovators" group. >>To post to this group, send email to transport-innovators@googlegroups.com. >>To unsubscribe from this group, send email to transport-innovators+unsubscribe@googlegroups.com. >>For more options, visit this group at http://groups.google.com/group/transport-innovators?hl=en. >>-- >You received this message because you are subscribed to the Google Groups "transport-innovators" group. >To post to this group, send email to transport-innovators@googlegroups.com. >To unsubscribe from this group, send email to transport-innovators+unsubscribe@googlegroups.com. >For more options, visit this group at http://groups.google.com/group/transport-innovators?hl=en.
Hmm, mentions the Lille metro as an autonomous vehicle (guideway based!) - this glosses over the potential (especially in the early days) of computer induced errors (Toyota brake example). Also the potential to plow into crowds and drive off cliffs (perhaps due to a computer virus or weather) - not the case with guideway based vehicles.
Though driverless vehicles can be electric, they probably won't be in this critical time where we need to act on climate change. Not likely that minimalist vehicles will be available because highway speeds tend to favour large vehicles, especially in bad or windy weather.
It is likely that extra infrastructure for driverless cars will be exactly as it is now - large and maintenance hungry. Guideway carrying small vehicles can be optimized and on a life-cycle cost likely end up ahead compared to roads, though we have no examples to prove this yet and this is design dependent.
Higher speeds can be achieved safely in most weather conditions, increasing productivity and reliability for workforce and supplies.
We're missing our chance to create a new, better transportation mode.
On Thursday, 24 May 2012 15:59:35 UTC-4, jschneider wrote:
> At 12:03 PM 5/24/2012, you wrote: > >I think that if you roll in the environmental cost, health care > >costs, insurance costs, fuel costs, vehicle costs, maintenance cost, > >etc. AND extra infrastructure to handle all these robocars, a case > >can be made for guideway + cheaper, smaller, shorter-range electric > >cars (DM). Unfortunately, many of these impacts aren't reflected in > >the perceived cost of car driving. My statement was "overall cost" > >- which includes societal costs as I see it. Google has made > >selling DM that much more difficult by pushing it's vision that is > >seen as "good enough". That's a shame.
> Fair enough - if you include the potential savings that might arise > from greatly improved safety and mobility services to groups that > would benefit from it (elderly non-drives, blind people, drunk > driving reductions, and so on). I would hope that someone on the > Google team is doing such calculations on societal impact issues. So > far, I've not see much on the subject with the exception of Tyler > Folsom's excellent paper on the subject. > Link to paper > entitled<http://faculty.washington.edu/jbs/itrans/tyler-article.pdf> > Social Ramifications of Autonomous Urban Land Vehicles, by Tyler C. > Folsom (posted in 2010, my What's New page).
> >F.
> >On Thursday, 24 May 2012 11:44:42 UTC-4, jschneider wrote: > >At 05:59 AM 5/24/2012, you wrote: > > >I think that the google approach is to slow down or stop if faced > > >with something unanticipated, perhaps demanding driver > > >intervention. It's harder to just stop in an airplane, but > > >acceptable (to some degree) on the road. It may learn/record the > > >correct response given the circumstance in the future. If cars are > > >made to communicate with Google (instant tracking, crowd sourcing), > > >reviewed interventions can be propagated to other cars in the same > > situation.
> > >I know everyone here wants it to fail and don't want to hear > > >different, but it will probably "work" in the statistical sense > > >where fatalities match or better human fatalities. It's a shame > > >this is what will be acceptable when near zero fatalities, no > > >emissions, greater throughput and less overall cost would be > > >possible with guideway based systems. > >I'd be interested to know more about why you think "less overall > >cost" - as it seems to me that lower cost (i.e. no guideway required) > >is regarded as a very prominent reason that its supporters always > >cite up front.
> >- Jerry Schneider - > > Innovative Transportation Technologies
>I am not sure you two guys are reading each other's posts >correctly. The present price of the Google auto-driving car is >significant. There is a hope that the astronomical hardware cost >might come down by a factor of ... not-going-to-happen. If there >are a billion $40,000.00 cars in the world and you want to add a >$6000 option to it that is 6 trillion dollars as their incremental >cost? Please check my math. Each car has to have a module and the >data network has to be paid for and maintained at some cost around >$100.00 per month in today dollars. TriTrack specifically is 5 >pennies a mile for energy or 1 billion cars driving 14,000 miles a >year. 14,000 miles equals $700/year times a billion cars is $700 >billion which is less than 6 trillion. Additionally the billion >cars get 20 mpg not TriTrack performance so the 6 trillion is just >the adder for self driving lazy driving. 14,000 miles at 20 mpg >with $5.00 gas is $3500/year in addition to the Google hardware on a >billion cars. $1200 per year per car for Internet. The world is >not that rich. Gasoline and diesel in addition to the cost of the >self-drivomatic will bankrupt us period. If the self-drivomatic was >free and paid each billion user $4000/year it still will cost more >than TriTrack specifically.
This is an example of taking the argument to an extreme level and using it to try to take down the other view - commonly used. The Google car is going to have to follow an evolutionary path that will have lots of checks based on experience, not extreme projections and will be subject to termination if it doesn't pass muster. As best I can tell, the concept resonates
very well with a large section of the public (not including the people who have rival inventions) and that popular support is now producing activity that is intended to encourage its deployment. How wise the Google team is going to be in formulating a phased implementation in locations that make sense is yet to be determined. No amount of calculation based on extreme assumptions is going to have much impact on what they think and do.
>The problem with "illion" math is the public has no idea what those >numbers mean. They are all just big numbers.
I agree that the public interest is largely emotional, not analytical.
>You do point out the externalities and they too are valid and add to >the "illion" math problem but you forgot some major ones that dwarf >the ones you listed. Same outcome.
Let's consider your assertion that the cost of equipping the vehicles (will they all be stock Toyotas).will not come down over time. Please explain why you believe this to be the case. Won't maintaining the data base be limited to those locations where the Phase I vehicles are allowed to travel without restriction? That's a good point and needs investigation. I would assume that the bigger the database, the more costly it will be to keep updated. Some areas will be growing, some declining and others not changing much at all.
>On Thu, May 24, 2012 at 2:03 PM, eph ><<mailto:rhapsodi...@yahoo.com>rhapsodi...@yahoo.com> wrote:
>I think that if you roll in the environmental cost, health care >costs, insurance costs, fuel costs, vehicle costs, maintenance cost, >etc. AND extra infrastructure to handle all these robocars, a case >can be made for guideway + cheaper, smaller, shorter-range electric >cars (DM). Unfortunately, many of these impacts aren't reflected in >the perceived cost of car driving. My statement was "overall cost" >- which includes societal costs as I see it. Google has made >selling DM that much more difficult by pushing it's vision that is >seen as "good enough". That's a shame.
>F.
>On Thursday, 24 May 2012 11:44:42 UTC-4, jschneider wrote:
>At 05:59 AM 5/24/2012, you wrote:
> >I think that the google approach is to slow down or stop if faced
> >with something unanticipated, perhaps demanding driver
> >intervention. It's harder to just stop in an airplane, but
> >acceptable (to some degree) on the road. It may learn/record the
> >correct response given the circumstance in the future. If cars are
> >made to communicate with Google (instant tracking, crowd sourcing),
> >reviewed interventions can be propagated to other cars in the same > situation.
> >I know everyone here wants it to fail and don't want to hear
> >different, but it will probably "work" in the statistical sense
> >where fatalities match or better human fatalities. It's a shame
> >this is what will be acceptable when near zero fatalities, no
> >emissions, greater throughput and less overall cost would be
> >possible with guideway based systems.
>I'd be interested to know more about why you think "less overall
>cost" - as it seems to me that lower cost (i.e. no guideway required)
>is regarded as a very prominent reason that its supporters always
>cite up front.
>Hmm, mentions the Lille metro as an autonomous vehicle (guideway >based!) - this glosses over the potential (especially in the early >days) of computer induced errors (Toyota brake example). Also the >potential to plow into crowds and drive off cliffs (perhaps due to a >computer virus or weather) - not the case with guideway based vehicles.
>Though driverless vehicles can be electric, they probably won't be >in this critical time where we need to act on climate change. Not >likely that minimalist vehicles will be available because highway >speeds tend to favour large vehicles, especially in bad or windy weather.
>It is likely that extra infrastructure for driverless cars will be >exactly as it is now - large and maintenance hungry. Guideway >carrying small vehicles can be optimized and on a life-cycle cost >likely end up ahead compared to roads, though we have no examples to >prove this yet and this is design dependent.
>Higher speeds can be achieved safely in most weather conditions, >increasing productivity and reliability for workforce and supplies.
>We're missing our chance to create a new, better transportation mode.
Well, it has been at least 40 years since some reasonable PRT concepts were defined Since then there have been alternative designs, and some minimal levels of cussing and discussing. Widespread visibility and public interest levels are still dismal.
Autonomous vehicles are relatively new on the scene and have gotten more broad and high level attention in the last few years that PRT has gotten in 40 years. We need to see what might be possible in terms of an evolutionary path, some decent cost estimates, some decent societal impact estimates (i.e. benefit/cost) and a whole lot more before a credible "better" judgment can be made.
>On Thursday, 24 May 2012 15:59:35 UTC-4, jschneider wrote:
>At 12:03 PM 5/24/2012, you wrote:
> >I think that if you roll in the environmental cost, health care
> >costs, insurance costs, fuel costs, vehicle costs, maintenance cost,
> >etc. AND extra infrastructure to handle all these robocars, a case
> >can be made for guideway + cheaper, smaller, shorter-range electric
> >cars (DM). Unfortunately, many of these impacts aren't reflected in
> >the perceived cost of car driving. My statement was "overall cost"
> >- which includes societal costs as I see it. Google has made
> >selling DM that much more difficult by pushing it's vision that is
> >seen as "good enough". That's a shame.
>Fair enough - if you include the potential savings that might arise
>from greatly improved safety and mobility services to groups that
>would benefit from it (elderly non-drives, blind people, drunk
>driving reductions, and so on). I would hope that someone on the
>Google team is doing such calculations on societal impact issues. So
>far, I've not see much on the subject with the exception of Tyler
>Folsom's excellent paper on the subject.
>Link to paper
>entitled<<http://faculty.washington.edu/jbs/itrans/tyler-article.pdf>http://faculty.washington.edu/jbs/itrans/tyler-article.pdf>
>Social Ramifications of Autonomous Urban Land Vehicles, by Tyler C.
>Folsom (posted in 2010, my What's New page).
> >F.
> >On Thursday, 24 May 2012 11:44:42 UTC-4, jschneider wrote:
> >At 05:59 AM 5/24/2012, you wrote:
> > >I think that the google approach is to slow down or stop if faced
> > >with something unanticipated, perhaps demanding driver
> > >intervention. It's harder to just stop in an airplane, but
> > >acceptable (to some degree) on the road. It may learn/record the
> > >correct response given the circumstance in the future. If cars are
> > >made to communicate with Google (instant tracking, crowd sourcing),
> > >reviewed interventions can be propagated to other cars in the same
> > situation.
> > >I know everyone here wants it to fail and don't want to hear
> > >different, but it will probably "work" in the statistical sense
> > >where fatalities match or better human fatalities. It's a shame
> > >this is what will be acceptable when near zero fatalities, no
> > >emissions, greater throughput and less overall cost would be
> > >possible with guideway based systems.
> >I'd be interested to know more about why you think "less overall
> >cost" - as it seems to me that lower cost (i.e. no guideway required)
> >is regarded as a very prominent reason that its supporters always
> >cite up front.
> >- Jerry Schneider -
> > Innovative Transportation Technologies
> >To post to this group, send email to > <mailto:transport-innovators@googlegroups.com>transport-innovators@googlegr oups.com.
> >To unsubscribe from this group, send email to
> ><mailto:transport-innovators%2Bunsubscribe@googlegroups.com>transpo > rt-innovators+unsubscribe@googlegroups.com.
> >For more options, visit this group at
> ><http://groups.google.com/group/transport-innovators?hl=en>http://g > roups.google.com/group/transport-innovators?hl=en.
>- Jerry Schneider -
> Innovative Transportation Technologies
