In contrast to the optimism I usually present here, and last week’s article about a self-driving Mercedes just a year away it’s worth noting this interview with various BMW folks where they provide a much more cautious timeline of at least a decade. Part of their concern comes from the use of computer vision systems. These are much cheaper than laser scanners but do not provide the reliability needed; it’s no accident that all the successful teams in the Darpa urban challenge relied very heavily on laser scanning.
I’m enough of an optimist that I am ready to bring forward the question “When will a child be born that never drives because of robocars?” Of course there are many people in the developed world who never get a licence for a variety of reasons, particularly people who live their lives in Manhattan and other transit-heavy cities. But for most of us, getting a licence and getting on the road is a rite of passage. Yet studies are showing that teens are now waiting longer to get a licence with various reasons speculated.
Nonetheless eventually we will see somebody who would normally have been jumping at the chance to get a licence and get out on the road who never gets one because they have a robocar. It won’t be easy of course, since even those who have robocars will still need to travel to places that don’t have them and rent cars, but many people who don’t have licences today just make use of taxis and transit in those situations.
I will put forward the proposal that this child may already have been born. When I see a baby today, I wonder, “will this child ever learn to drive?” While 16 years is aggressive for the ubiquitous fully autonomous operation needed for this, I do think we’re on the cusp, and if that child has not yet been born, it’s not too far away.
One reason for this is all the forces that are already reducing teen driving. A teen debating whether to take the effort to learn to drive might easily be swayed not to because mom has bought him a robocar. Once a successful safety record for robocars is demonstrated, parents will buy them for teens — instead of buying them driving lessons — and pressure the teens to not take the risk of driving themselves.
In other news, here’s a pointer to work by designer Charles Rattray on the look of future robocars. His designs match with my position that many robocars should be half the width of today’s cars, carrying only 1-2 people, since the vast majority of cars today only carry 1-2 people. Today’s car buyers insist on 5 passenger sedans (or larger) but when you have mobility-on-demand you can use the right vehicle for the trip on every trip, and that’s going to mostly be one person vehicles. This in turn, is the real key to efficient transportation, because while you can do great things with more efficient or electric power trains and more aerodynamic cars, nothing compares to making the car smaller, lighter and narrower in a major way. He has many design sketches and a video of how he sees the cars in action.
For the first time, a car company has put a date on shipment of a car with self-driving ability.
According to British site Auto Express, Mercedes has revealed that their 2013 S-class will feature self-driving. Not clear if there is an official company press release, though the company has been talking about such features, as have many other companies. Realize that the 2013 model year is just a year away.
The car will feature radar based automatic cruise control, combined with lane-marker following, and the automatic driving will only operate below 40kph. In other words, this is designed to let you take your hands off the wheel in stop-and-go traffic jams, not to drive you at actual open driving speeds. You’ll need to pay attention to the road, not read a book, but at that low speed you’ll have decent warning if something goes wrong and the car starts drifting, so I suspect that in spite of warnings not to do so, people will get away with minor tasks like reading a few e-mails or even sending some.
While a very basic level introduction, this is still a milestone and will pave the way (love those road metaphors) for other companies. While the focus of the DARPA grand challenges and most visions of the robocar future has been on cars that can drive completely on their own, there are now strong signals that the technology will arrive in the form of driving assist, and human drivers will be called upon to still do much of the driving, in particular the tricky bits the systems aren’t safe to handle. In my article a few years ago the roadmap to robocars I suspected we might see a few specialized applications first, such as robot valet parking and even autonomous vehicles for military delivery applications, but now the autopilot is on track for showing up commercially first.
I shoot with the Canon 5d Mark II. While officially not a pro camera, the reality is that a large fraction of professional photographers use this camera rather than the Eos-1D cameras which are faster but much bulkier and in some ways even inferior to the 5D. But it’s been out a long time now, and everybody is wondering when its successor will come and what features it will have.
Each increment in the DSLR world has been quite dramatic over the last decade. There’s always been a big increase in resolution with the new generation, but now at 22 megapixels there’s less call for that. While there are lenses that deliver more than 22 megapixels sharply, they are usually quite expensive, and while nobody would turn down 50mp for free, there just wouldn’t be nearly as much benefit from it than the last doubling. Here’s a look at features that might come, or at least be wished for.
More pixels may not be important, but everybody wants better pixels.
Low noise / higher ISO: The 5D2 astounded us with ISO 3200 shots that aren’t very noisy. Unlike megapixels, there is almost no limit to how high we would like ISO to go at low noise levels. Let’s hope we see 12,500 or more at low noise, plus even 50,000 noisy. Due to physics, smaller pixels have higher noise, so this is another reason not to increase the megapixel count.
3 colour: The value of full 3-colour samples at every pixel has been overstated in the past. The reason is that Bayer interpolation is actually quite good, and almost every photographer would rather have 18 million bayer pixels over 6 million full RGB pixels. It’s not even a contest. As we start maxing out our megapixels to match our lenses, this is one way to get more out of a picture. But if it means smaller pixels, it causes noise. The Foveon approach which stacked the 3 pixels would be OK here — finally. But I don’t expect this to be very likely.
Higher dynamic range: How about 16 bits per pixel, or even 24? HDR photography is cool but difficult. But nobody doesn’t want more range, if only for the ability to change exposure decisions after the fact and bring out those shadows or highlights. Automatic HDR in the camera would be nice but it’s no substitute for try high-range pixels.
Video & Audio
Due to the high quality video in the 5D2, many professional videographers now use it. Last week Canon announced new high-end video cameras aimed at that market, so they may not focus on improvements in this area. If they do, people might like to see things like 60 frame video, ability to focus while shooting, higher ISO, and 4K video. read more »
Last week, new studies came back on the California High Speed Rail project. They have raised the estimated cost to $99 billion, and dropped the ridership estimate to 36.8 million and $5.5 billion in annual revenue. Note that only around 20 million people currently fly the SF to LA corridor — they expect to not just capture most of those but large numbers of central valley trips.
Even at the earlier estimates the project was an obvious mistake, and there’s no way to financially justify spending $99 billion to pull in $5.3 billion/year even subbing zero in for the large operating cost. But for various political reasons involving getting federal money, some are still pushing for this project, and we may well build a short train to nowhere in the central valley just to get the federal bucks.
They’re planning there because the various cities in the populated areas have been fighting legal battles to block the train there, not wanting its disruption. Because the train can only stop if a very few places at the speed it wants to go, a lot of towns would end up having construction and noise and street blockage and not get a lot of use from the train.
The local opposition is a tough barrier, because the train ends up really only being useful where the people are. While I have doubts about how many people would ride the long haul, since few want to go from downtown SF to downtown LA, lots of people would ride a fast train in the urban areas. In particular, what nobody talks about is running the HSR primarily to the airport, and streamlining both security clearance and the connection with new technology. The only reason HSR is pushed as possibly competing with flights is because of the nightmare we have made of flying, where people have to get to airports 45 minutes ahead of even short-haul flights and take a fair bit of time to get out of airports on the other end and make it through traffic to their destinations. A fast train from a downtown to the airport where you clear security (and check bags) right on the train, and the train drops you right at the central gate areas post security would create an unbeatable trip from downtown anywhere to downtown anywhere.
For fast trains, the San Francisco to San Jose route is so short that a 250mph HSR could do the 48 mile trip between the towns in 12 minutes without stopping, call it 15 with the start and stop at each end. This opens up an interesting cost saving — you could build a single track, and have a train zip back and forth on it, and still provide service every 30 minutes. You could put a double-track section in the middle and have service every 15 minutes, with lots of safety interlocks of course. A single track requires less land, less of everything and could probably be built along easier routes, even highway medians in some cases. You could avoid turnaround time by having double track at the endpoints, so one train is leaving for opposite route the moment the other train arrives, giving each train quite a long turnaround — with double rolling stock.
Of course, having no stops is not that valuable because only a few people want to go from SJ to SF. People would want a stop at the airport as I have indicated, and at least one in Mountain View or Palo Alto. Each stop costs a bunch of time, and eventually the trip gets long enough that the single-track trick becomes less useful. For a while I’ve wondered if you could make trains that could dock, so that the main train runs non-stop and is able to shed cars which stop at local stops (not that hard) and to dock with cars coming from local stops (harder.) I proposed this 7 years ago near the start of this blog, and there are serious rail designers thinking along the same lines — see the video in that link.
In the Priestman Goode proposal, they have trains docking side to side. That seems much more challenging though it offers fast transfer. If you combine the two ideas, you would have two tracks — one for the nonstop trains and one for the docking shuttles which serve all the local stops. Indeed, if you could do this you could get rid of the old regular speed rail service running on existing track pairs because this would be superior in all ways except cost. My own proposals attempted to dock on a single track, which seems easier to me.
Robocars play a role in all this too. Even the HSR authority realizes they have a big problem, in that once people get quickly to an HSR station, they still have to get to their real destination. Using local transit may mean spending more time on a local bus than on the HSR. The mobility on demand of robocars is a great answer, and I’m pretty sure that with a 2030 forecast completion date (if they’re lucky) we’ll have robocars long before then. And the one thing cars can’t readily do is go very fast efficiently between cities.
The docking approach, should it work, has another advantage. The main train can take the best route (cheapest or shortest) without too much regard for where the stations are. People like stations in urban centers, but bringing the high speed train right through such areas (like Palo Alto) is hard and has caused the lawsuits. If the train goes through the industrial space along the Bay, and a spur goes into downtown for the shuttle that docks with it, you get a win all around.
Another approach that doesn’t require dock/undock works when you have a solid terminus like SF. You have 3 trains leave SF at the same time. The first one goes express to San Jose. The second goes express to Palo Alto and Mountain View and then switches to low speed tracks to go to Sunnyvale and Santa Clara. The third goes to SFO airport. Because SFO airport is also an origination point, it sends a train to SJ just before or after the one from SF, and another train to Mountain View right after that one. Mountain View to SJ service might be able to fit in or have to be local service. These sub-trains are just a few cars. This is not as energy efficient, though it can be if the trains are able to get close to one another and draft, sort of a virtual coupling without physical contact. You need perfect sync, and special long-spring collision bumpers in case the sync fails and they bump. The risk of higher-speed bumping must be prevented by failsafes that don’t even let the trains get on the same track until speed is matched close enough. This requires more than just a single track of course.
Congestion on the roads has a variety of sources. These include accidents of course, reductions in road capacity, irrational human driving behaviours and others, but most of all you get congestion when more cars are trying to use a road than it has capacity for.
That’s why the two main success stories in congestion today are metering lights and downtown congestion charging. Metering lights limit how fast cars can enter the highway, so that you don’t overload it and traffic flows smoothly. By waiting a bit at the metering light you get a fast ride once on the highway. Sometimes though, especially when the other factors like accidents come into play, things still gum up.
Now that more and more cars are connected (by virtue just of the smartphone the driver carries if nothing else) the potential will open up for something else in congestion — finding ways to encourage drivers to leave a congested road. read more »
A little self-plug. I have an article on an introduction to panoramic photographic technique the November issue of Photo Technique with a few panos in it. This is old world journalism, folks — you have to read it on paper at least for now.
We decided to go to Harvey’s pan in Savuti one afternoon and lucked upon a large breeding group of elephant just on their way there. I caught them in one of my first long lens panoramas. Long lens panos are fairly difficult due to the limited depth of field, but they get great detail on the baby elephant.
I’m just back from the “ITS World Congress” an annual meeting of people working on “Intelligent Transportation Systems” which means all sorts of applications of computers and networking to transportation, particularly cars. A whole bunch of stuff gets covered there, including traffic monitoring and management, toll collection, transit operations etc. but what’s of interest to robocar enthusiasts is what goes into cars and streets. People started networking cars with systems like OnStar, now known in the generic sense as “telematics” but things have grown since then.
The big effort involves putting digital radios into cars. The radio system, known by names like 802.11p, WAVE and DSRC involves an 802.11 derived protocol in a new dedicated band at 5.9ghz. The goal is a protocol suitable for safety applications, with super-fast connections and reliable data. Once the radios in the car, the car will be able to use it to talk to other cars (known as V2V) or to infrastructure facilities such as traffic lights (known as V2I.) The initial planned figured that the V2I services would give you internet in your car, but the reality is that 4G cellular networks have taken over that part of the value chain.
Coming up with value for V2V is a tricky proposition. Since you can only talk to cars very close to you, it’s not a reliable way to talk with any particular car. Relaying through the wide area network is best for that unless you need lots of bandwidth or really low latency. There’s not much that needs lots of bandwidth, but safety applications do demand both low latency and a robust system that doesn’t depend on infrastructure.
The current approach to safety applications is to have equipped cars transmit status information. Formerly called a “here I am” this is a broadcast of location, direction, speed and signals like brake lights, turn signals etc. If somebody else’s car is transmitting that, your car can detect their presence, even if you can’t see them. This lets your car detect and warn about things like:
The car 2 or 3 in front of you, hidden by the truck in front of you, that has hit the brakes or stalled
People in your blind spot, or who are coming up on you really fast when your’re about to change lanes
Hidden cars coming up when you want to turn left, or want to pass on a rural highway
Cars about to run red lights or blow stop signs at an intersection you’re about to go through
Privacy is a big issue. The boxes change their ID every minute so you can’t track a car over a long distance unless you can follow it over every segment, but is that enough? They say a law is needed so the police don’t use the speed broadcast to ticket you, but will it stay that way?
It turns out that intersection collisions are a large fraction of crashes, so there’s a big win there, if you can do it. The problem is one of critical mass. Installed in just a few cars, such a system is extremely unlikely to provide aid. For things like blindspot detection, existing systems that use cameras or radars are far better because they see all cars, not just those with radios. Even with 10% penetration, there’s only a 1% chance any given collision could be prevented with the system, though it’s a 10% chance for the people who seek out the system. (Sadly, those who seek out fancy safety systems are probably less likely to be the ones blowing through red lights, and indeed another feature of the system — getting data from traffic lights — already can do a lot to stop an equipped car from going through a red light by mistake.) read more »
They say that famous deaths come in threes. That’s no doubt just an artifact of our strange sense of coincidence, but after Jobs and Ritchie, tonight we learn of the death of John McCarthy, AI pioneer and creator of LISP.
My first personal encounter with John was part of a big story of my life, the banning of rec.humor.funny. In a short summary of what’s told there, RHF had been banned at Waterloo and later, due to a comedy of errors got banned at Stanford. Shortly after the ban John called me up and said he wanted to be a champion against the ban. He had been worried for some time about the growing tide of speech codes at supposed bastions of academic freedom, and the idea of banning publications on the internet was a new level. John used his sway to get some press, organize a protest march and have the matter fixed by the academic senate. Strangely, just a few days ago I was at a dinner for a group called FIRE which fights against crazy academic bans, and I was recounting the story of what John did at Stanford for the first time in many years.
Later, I moved to silicon valley and got to know John in person a bit more. He was an incredible force of character long after the age where most have shrunk away. If the AIs of the future are able to resurrect the figures of the past, you know he’ll be one of the first in line for them.
RIP John. And Dennis (who I praised over on Google+). And Steve Jobs. Let’s really limit it to three for a while.
Since getting involved with Google’s self-driving-car team, I’ve had to keep silent about its internals, but for those who are interested in the project, a recent presentation at the intelligent robotics conference in San Francisco is now up on youtube. The talk is by Sebastian Thrun (overall project leader) and Chris Urmson, lead developer. Sebastian led the Stanley and Junior teams in the Darpa Grand Challenge and Chris led CMU teams, including BOSS which won the urban challenge.
The talk begins in part one with the story of the grand challenges. If you read this blog you probably know most of that story.
Part two (above) shows video that’s been seen before in Sebastian’s TED talk and my own talks, and maps of some of the routes the car has driven. Then you get Chris showing some hard technical details about mapping and sensors.
Part three shows the never before revealed story of a different project called “Caddy”: self-driving, self-delivering golf carts for use in campus transportation. The golf carts are an example of what I’ve dubbed a WhistleCar — a car that delivers itself and then you drive it in any complex situations.
If you want to see what’s inside the project, these videos are a must-watch, particularly part 2 (embedded above) and the start of part 3.
There’s lots of other robocar news after the Intelligent Transportation Systems conference, which I attended this week in Orlando FL. The ITS community is paying only minimal attention to robocars, which is an error on their part, but a lot of the technology there will eventually affect how robocars develop — though a surprising amount of it will become obsolete because it focuses on the problems caused by lots of human driving.
The list of robocar teams grows again with a new project from Oxford university, led by Paul Newman. Nissan is also involved, though the base vehicle is a Bowler Wildcat off-road vehicle.
The project sports a LIDAR design I have not yet seen, with 4 laser units on a mount spinning at what looks like 1-2hz, but they claim a 40hz sampling rate and do have very nice mapping results. They claim their localizer is very good, and demos show it working on rough off-road terrain. Some videos also see it doing waypoint driving without the LIDAR but they talk about why GPS is not adequate.
The claims about the vehicle have a British understatement. They say it will be 10-15 years before it’s ready for the roads, and talk mostly about simple problems like handling traffic jams — something Audi, BWM and VW have all claimed they will release in the middle of this decade, using simpler sensor systems. He also envisions a future arms-race where a car that can do 10 minutes/day of self-driving competes with one that can do 15.
Congestion is their main message it seems, citing the Dept. for Transport’s figures of a 25 billion pound cost for congestion in 2025 in the UK.
Boston Dynamics has gone even further with their latest model, AlphaDog
The AlphaDog’s legs are hydraulic, and so adding legs like this to a car which has a motor and compressor is not so far fetched. In this design they could easily fold up into the sides of a single person wheeled vehicle. In the video, the robot is shown carrying 400lbs of weights, and a range of 20km is claimed. You might not quite want to ride it yet, but that’s coming.
Let’s look at some of the consequences for transportation and cities:
Houses need not be on streets to have full access by small vehicles and cargo delivery robots. They can be on the side of hills and up stairs. Neighbourhoods can be built with just small lightly paved or graded paths so that the robot’s legs don’t disturb the terrain.
The robots may well, in a controlled environment, be able to place their feet with good precision. As such the path for a walking robot might look like just a series of stone pads dotting the grass — the way some paths for people look. In reality they would be more sturdy, but that’s what they could look like.
In developing countries which do not have infrastructure, they may never have to put in that much infrastructure. Combined with flying robots, delivery of goods can become possible to any location, and at high speed.
The world’s tourist destinations may become swamped with people who can ride a walking robot to remote locations where before the daunting hike kept the crowds down. There will be efforts to ban walking chairs, but the elderly and disabled will be able to fight such bans as discriminatory.
Indeed, for the disabled and aged, the walking chair robot might well open up lots of the world that is now closed. The main issue would be power and noise. The motors that power BigDog are very noisy. AlphaDog in the video is using external power.
Robotic cargo delivery (deliverbots) need no longer be limited to places you can roll up to. That can include places inside buildings, even up stairs.
I’m actually not a fan of login and sessions on the web, and in fact prefer a more stateless concept I call authenticated actions to the more common systems of login and “identity.”
But I’m not going to win the day soon on that, and I face many web sites that think I should have a login session, and that session should in fact terminate if I don’t click on the browser often enough. This frequently has really annoying results — you can be working on a complex form or other activity, then switch off briefly to other web sites or email to come back and find that “your session has expired” and you have to start from scratch.
There are times when there is an underlying reason for this. For example, when booking things like tickets, the site needs to “hold” your pending reservation until you complete it, but if you’re not going to complete it, they need to return that ticket or seat to the pool for somebody else to buy. But many times sessions expire without that reason. Commonly the idea is that for security, they don’t want to leave you logged on in a way that might allow somebody to come to your computer after you leave it and take over your session to do bad stuff. That is a worthwhile concept, particularly for people who will do sessions at public terminals, but it’s frustrating when it happens on the computer in your house when you’re alone.
Many sites also overdo it. While airlines need to cancel your pending seat requests after a while, there is no reason for them to forget everything and make you start from scratch. That’s just bad web design. Other sites are happy to let you stay “logged on” for a year.
To help, it would be nice if the browser had a way of communicating things it knows about your session with the computer to trusted web sites. The browser knows if you have just switched to other windows, or even to other applications where you are using your mouse and keyboard. Fancier tools have even gone so far as to use your webcam and microphone to figure if you are still at your desk or have left the computer. And you know whether your computer is in a public space, semi-public space or entirely private space. If a browser, or browser plug-in, has a standardized way to let a site query session status, or be informed of session changes and per-machine policy, sites could be smarter about logging you out. That doesn’t mean your bank still should not be paranoid if you are logged in to a session where you can spend your money, but they can be more informed about it. read more »
But there’s one place this might make sense. I think you should get a chance to do a survey after every interaction with the police, as well as others who have some color of authority over you (judges, security guards, border patrol etc.) The data you enter would be anonymous, and the survey conducted by a different party bonded to protect your privacy. There would also be entry in some means (perhaps with different classes of card) about whether the encounter was assistive, or was a stop, or lead to arrest though there are limits on this while keeping the data anonymous. If you are required to identify yourself as part of the encounter, this can be your means to getting a card later, though again the data entered must not be tied to your name.
Police would get small cards which have a cryptographic code which allows the bearer to fill out the survey. They would be required to hand one out in any incident. The number handed out would need to be close to the count in their own incident report, so that they don’t just keep the cards to fill out positive surveys on themselves. If police won’t give you a card that’s a serious matter itself.
Of course, people who have been stopped, rather than assisted by police will have a naturally antagonistic view. What would matter in these surveys would be how each officer compares to the other officers. You would not judge officers on their absolute score, but their score relative to other officers with similar duties. These scores would be admissible in court when an officer testifies. An officer with a seriously bad record would become less trusted by judges and juries. The worst cops would have to leave the force, being unable to testify in court without being doubted. And the absolute numbers would also tell us something. On the forms, people could complain about misuse of authority and corruption, and could also leave positive remarks.
The 3rd party taking in the data would have to have impeccable credentials so people trust that it truly destroys any association between submitter and data. They would also have to be trained at how to protect against re-linking. (For example, if dates can be figured out, officers may well be able to connect people with forms. As such data must be released slowly, and only after a large enough number of forms are in the batch, and forms with unique profiles must be merged with care.) In most cases the 3rd party would have to be in another state, and possibly another country to assure it is not under the sway of those it is collecting data on.
We also would have to assure that people don’t try to sell the survey cards. That’s hard, if they are to be truly anonymous. You might have to use them quickly, to avoid giving you time to find a buyer. The 3rd party could run regular stings trying to buy and sell cards and pierce anonymity on just those. I’m sure that there are other ways officers would try to game the system that would have to be found and dealt with. Over time, the data should become public in amalgamated form, not just available to defence lawyers.
Some Robocar updates, since with Burning Man and Singularity U my posting volume has been down:
BMW Highly Automated Driving
BMW has announced a prototype car with a “Highly Automated Driving” feature on the 5 series. The vehicle, which has logged 3,000 miles of minimal driver involvement, uses a vision system from the lane departure camera, maps, the ACC radar and high-accuracy GPS. It is claimed to handle allowing and doing highway merges. The system is not meant to allow the driver to take their eyes off the road, though, and his highway-only. The system was also called “ConnectedDrive Connect (CDC)” which strikes me as a somewhat awkward name.
BMW also recently demonstrated some other self-driving technology on test tracks and has a very active lab. With Volkswagen’s Temporary Auto Pilot, Mercedes and Audi all promoting concept cars with autonomous modes, German companies seem to be the most engaged in the field. BMW has also shown a “traffic jam assistant” similar to the Audi system below.
Audi EV A2 has traffic jam autopilot
Another concept car (An earlier Audi was used in the Pikes Peak challenge) from Audi will feature an autopilot for stop and go traffic. Here they combine lane-keeping and automatic cruise control, except at lower speeds where the risk is low. The driver still has to pay some attention, and when traffic gets moving again they must take full control. Still, it’s aimed at one of the greatest annoyances of commuting.
Made in Germany continues on the roads
The AutoNOMOS team reports success with their vehicle on the streets of Berlin, though so far only 80km of operation. They have 6 laser sensors, and claim that they believe the vehicles will soon be ready to deploy on private roads, but will need a decade of legal work to be used on the public streets.
Autonomous tractor out working the fields
For some time the concept of “precision agriculture” has been the high-tech hotspot in farming. Here, tractors equipped with GPS and other location technologies take detailed maps of the fields and the crops and custom control spray of fertilizers, pesticides and other substances according to data about what’s right for each small square. However, farmers are still at the controls. Some day, tractors may drive themselves and engage in many farm activities. In the fields, there is not usually much to run into and you don’t go very fast. On the other hand, most family farmers today farm as a lifestyle and aren’t looking to automate it, but agribusiness operators are.
Yesterday, Don Henley (of the Eagles) penned an editorial in USA Today supporting the Protect IP Act (PIPA) which has serious free speech implications and turns web sites into copyright police. Don called out both the EFF (of which I am a Director) and Google (which is a consulting client of mine) so I have this whimsical response for him:
Take it Easy, Don. There’s a New Kid in Town, and it’s called the
Internet. Get Over It. I Can’t Tell you Why, but in The Long Run,
there isn’t going to be a Heartache Tonight. One of these Nights I hope
you’ll understand that for search engines to Take it To the Limit,
they can’t be forced to police every search result.
only grow when living Life in the Fast Lane, able to operate, innovate
and design products without needing to check for permission from the
music industry. If every time you wrote a song you had to worry about
what every user who plays it and every store that sells it might do with
it, you would lose your Peaceful, Easy Feeling quickly. Big companies
might run filters, but if the small ones had needed to they would be
The Best of My Love,
Brad Templeton, EFF
PS: Henley really is a great musician, and I do understand his frustration with people who don’t pay him for his work. But click the link above for EFF’s explanation of why this sort of approach isn’t really going to help musicians and will do real damage to the way people use computers and the internet.
Today an op-ed by John Sununu and Harold Ford Jr. of “Broadband For America” (a group of cable companies and other ISPs which says it is really a grass-roots organization) declared that the net needs a better pricing model for what Netflix is doing. For a group of ISPs, they really seem to not understand how the internet works and how pricing works, so I felt it was worthwhile to describe how things work with a remarkably close analogy. (I have no association with Netflix, I am not even a customer, but I do stream video on the net.)
You can liken the internet to a package delivery service that works somewhat differently from traditional ones like the postal service or FedEx. The internet’s pricing model is “I pay for my line to the middle, and you pay for your line to the middle and we don’t account for the costs of individual traffic.”
In the package model, imagine a big shipping depot. Shippers send packages to this depot, and it’s the recipient’s job to get the package from the depot to their house. The shippers pay for their end, you pay for your end, and both share the cost of creating the depot.
Because most people don’t want to go directly to the depot to get their packages, a few “last mile” delivery companies have sprung up. For a monthly fee, they will deliver anything that shows up at the depot addressed to you directly to your house. They advertise in fact, that for the flat fee, they will deliver as many packages as show up, subject to a fairly high maximum rate per unit of time (called bandwidth in the internet world.) They promote and compete on this unlimited service.
To be efficient, the delivery companies don’t run a private truck from the depot to your house all the time. Instead, they load up a truck with all the packages for your neighbourhood, and it does one delivery run. Some days you have a lot of packages and your neighbours have few. Other days you have few and they have a lot. The truck is sized to handle the high end of the total load for all the neighbours. However, it can’t handle it if a large number of the neighbours all want to use a large fraction of their total load on the same day, they just didn’t buy enough trucks for that, even though they advertised they were selling that.
This is not unreasonable. A majority of the businesses in the world that sell flat rate service work this way, not just internet companies. Though there are a few extra twists in this case:
The last mile companies have a government granted franchise. Only a couple can get permission to operate. (In reality — only a few companies have got permission to have wires strung on poles or under the street.)
Some of the last mile companies also used to be your exclusive source for some goods (in this case phone service and TV) and are concerned that now there are competitors delivering those things to the customers.
The problem arises because new services like Netflix suddenly have created a lot more demand to ship packages. More than the last mile companies counted on. They’re seeing the truck fill up and need to run more trucks. But they proudly advertised unlimited deliveries from the depot to their customers. So now, in the op-ed, they’re asking that companies like Netflix, in addition to paying the cost of shipping to the depot, pay some of the cost for delivery from the depot to the customer. If they did this, companies would pass this cost on to the customer, even though the customer already paid for that last mile delivery. read more »
Tuesday we and Aneesh Chopra, CTO of the USA come to Singularity University and among many things, he was asked about immigration. (In part because our class comes from 35 countries and many of them would love to be entrepreneurs in the USA.) Chopra announced some immigration rule clarifications that had come out that day which will help things somewhat. They did rule clarifications because getting congress to do meaningful reform is very hard.
I gave him one suggestion, inspired by something I saw long ago at a high-roller CEO conference for the PC industry. In a room of 400 top executives and founders of PC and internet startups, it was asked that all who were born outside the USA stand up. A considerable majority (including myself) stood. I wished every opponent of broader immigration could see that.
So I proposed that he, or others, go around to conferences of high-tech entrepreneurs and founders, and ask this question, and film the people standing up. Then edit that into a nice rapid-fire video to play for members of congress and immigration opponents. It’s hard to argue with. The hope of US leadership and job creation is severely diminished without immigrants.
It’s often been lamented that some of the brightest from the world, who beg to come to the USA to get degrees, are then kicked out when their education ends, and they go home with their PhD. “Staple a green card to a PhD” is one approach.
A different approach, possibly to be combined, is to create a provisional permanent resident status. The main rule, perhaps the only rule, is that if you are on it you must pay a minimum tax averaging $30,000 per year for six years. You can go below once or twice but must make it up to bring up the average. And once you hit the total of $180,000 you get a real green card. If you don’t hit it, you can leave. If you don’t leave and don’t pay, it’s tax evasion, for which you can be both jailed and deported.
Yes, this is access for the rich. It means the wealthy could buy green cards, but also that entrepreneurs or people will well funded partners could easily get in. More to the point, the people coming in would clearly not be a drain on U.S. society. And with the salary needed to pay $30K in tax being about $130,000 (based on the real tax rate including all deductions) these are largely not people “stealing US jobs by working cheap.” Admittedly an employer could offer to pay an immigrant a low wage and pick up the tax but that could be illegal if difficult to enforce.
This price is below the prices often cited for “buy immigrant status” programs in the US and around the world, and the values in investor visas. I think you want to get young, less well-heeled entrepreneurs so I think million dollar prices are the wrong idea. But it may be that there is some price that even the biggest xenophobes will agree results in a net win. What is that number?
A report from China Daily details a successful 286km trip by a vehicle from the National University of Defense Technology.
According to the report the vehicle did fully autonomous highway driving, and did not use GPS, just LIDAR, radar and vision systems. Modern robocars do not rely exclusively on GPS, as it is both inaccurate and goes out in certain regions, but they do like to use it to help them localize, so not using it at all suggests a good localizer. They also report handling fog, thunderstorms and poor lane markers, all good acheivements.
There is what appears to be a machine translated article with some photos. The external photo seems to show an unmodified car, the internal one shows what appear to be two stereo cameras left and right, though the article talks mostly about LIDAR and radar, not vision systems. Dai Bin, spokesman for the team has published papers on both LIDAR based obstacle avoidance and vision systems similar to the one used by Stanley in the Darpa desert challenge for off-road use.
It will be interesting to hear more about the work of this team.
In other news, BMW has announced that its 2014 i3 battery-electric car will also offer a combination of lane-keeping and adaptive cruise control](http://www.greencarreports.com/news/1064191_2014-bmw-i3-moves-us-closer-to-autonomous-driving-in-cities) while under 25mph, effectively being a “stop and go traffic autopilot.” This is an interesting early offering, since it’s easier to put this together at low speeds in the controlled highway environment.
Also included will be a detection system for impeding collisions which can apply the brakes, and a fully automatic parallel parking system (most earlier systems still needed human control of the brake, causing amusing-if-it’s-not-you accidents for people who didn’t think they had to do anything.)
This is a challenge to blog readers to come up with (or find examples in practice) of good systems to allocate students to parallel sessions based on their preferences. I’ve just concluded one round of this, and the bidding system I built worked OK, but is not perfect.
The problem: Around 80 students. On 10 days over 4 weeks they will be split into 3-5 different parallel sessions on those days. Many sessions have a cap on the number of students, and more students will have them as a 1st choice than can fit. Some sessions can take many students and won’t fill up. The students can express their preference as ranking, or with numeric values.
This is known in the literature as the Allocation problem, and there are various approaches, though none I found seemed to fit just right, either being easy to code or having existing running code. But I am keen on pointers.
Maximize student satisfaction/minimize disappointment. Giving a student their 1st choice is good. Giving 3rd or 4th choices is bad.
The system must be easy for the students to understand and use.
Fairness. This has many meanings, but ideally mismatches that can’t be avoided should be distributed. If somebody gets a 4th choice one day, they perhaps should have a better shot at a 1st choice on another day.
It’s nice if there’s a means of applying penalties to students who violate rules, sneak into sessions etc. Academic violations can result in less chance at getting your 1st choice.
It should be flexible. Sessions may have to be changed or many not fully finalize until a week before the session.
It is nice to handle quirks, like duplicated sessions a student takes only once, but where the student might have preferences for one of the instances over another. There may also be pre-requisites, so only students who take one session can have the sequel.
Things change and manual tweaking can be advised.
Rank sessions in order, 1st come, 1st served
This was used in the prior year. Much like a traditional sign-up sheet in some ways, students could indicate their choices in order. If more students had a session as 1st choice than would fit, the ones who filled out their form first got in. This gave priority over all 10 days and so it was changed to rotate each week to distribute who was first in line. read more »