top San Francisco Airport Testing Beacon System For Blind Travelers
It is also quite straightforward to make your app VoiceOver friendly. The tricky part is managing some of the VoiceOver gestures. Aside from the equivalent of alt tags on images, the most important part is getting the step forward/backward and continuous scroll gestures working right in your app. Turn on VoiceOver, swipe down with two fingers and see what happens. Then swipe one finger left or right. If you've done your job right, the cursor will move in the order you expect. Scrolling on lists that go below the fold are also subtle, yet important.
top Fooling a Mercedes Into Autonomous Driving With a Soda Can
This actually isn't that big of a leap from a technical difficulty level. A pair of Carnegie Mellon researchers
drove across the country in 1995 using a forward camera based system. 98.2% of the trip was autonomous. The non-autonomous parts of the NHAA drive are the same which would be needed under this approach.
top If Ridesharing Is Banned, What About Ride-Trading?
ITNAmerica has already sorted this out. It is possible to load your parent's ride account with credits from cash or your own driving contributions. For example, you can drive people around your area and the credits are used by your parents who live halfway across the country. If you fall short one month, just fill up the credits with cash.
top Government To Require Vehicle-to-vehicle Communication
In many US States, the local Departments of Transportation want nothing to do with enforcement actions. They will let the police/town/etc install red light cameras, but they don't want to be involved beyond that. In fact many red light cameras are operated by private companies under contract with local municipalities.
Here's an example of why DOTs don't want to be involved in enforcement. A while back some politician in New Jersey, not part of the local DOT, floated the idea of using EZPass toll data to automatically issue speeding tickets. This was almost certainly a money grab. Massive amounts of drivers started asking how to get rid of their EZPass accounts and turn in their transponders. DOT knew lower market penetration would negatively impact congestion at toll booths. They, thankfully, squashed the idea quickly.
top Why Competing For Tenure Is Like Trying To Become a Drug Lord
The main "duty" of most non-tenured professors is to produce research. If you do that best by working regular 9am-5pm hours or by only coming in in the middle of the night, nobody's going to care much. Aside from that, you need to attend occasional meetings and turn your grades in at the end of the semester. Once you have tenure, the obligation to produce continuous research is lessened a bit, and most of the schedule on which you "fulfill your duties" is really up to you.
From my perspective in the trenches, the reduction is not as big as most people might think for CS and the sciences. If you worked like crazy while building your credentials, either for tenure or to a senior position in a non-tenure research track, you can't really slack off too much. You still need to bring in the cash to cover your team, grad student tuitions, and your own salary, which are now more expensive too. This means just as much research effort and proposal writing. This is exacerbated when research funding is cut at a large scale (sequestration). The reduction really comes from i) having established robust lab practices, methods, and management skills and ii) improved proposal writing skills combined with a track record. Junior faculty expend a lot of time finding and developing the right models, processes, and skills.
Another problem is that you spend your early career developing and reinforcing workaholic habits. It is very hard to step away from work, even for a regular weekend. Unlike most high intensity jobs, the flexible time is great for scheduling around family so they actually see you. You can insulate them from the worst of it.
top Valve Announces Linux-Based SteamOS
Um, no. It has been in living rooms for over a decade. TiVo runs linux. Now get off my lawn!
top Cadillac SRX Converted Into Self-Driving Car
V2V is peer-to-peer and really focused on reducing reaction times. It allows the car ahead to instantly tell the car behind it is braking. This means less latency for corrective action. This also helps non-autonomous cars since V2V equipped vehicles could, theoretically, suck up some of the shockwaves present in current highway driving.
top Cadillac SRX Converted Into Self-Driving Car
The main advance is the progression towards real-world sensor selection and packaging. If you look at all the cars which completed the Urban Challenge, and the Google cars, you'll notice the spinning Velodyne laser sensor on the roof. It is a great sensor and makes autonomous driving much easier. Unfortunately, that sucker costs more than most luxury cars and would never be deployed the real-world since nobody wants a spinning can on their roof.
Carnegie Mellon would not have won the Urban Challenge without that sensor
or the others littered all over the exterior of the car. The major advance for this new Carnegie Mellon car is comparable performance with cheaper sensors fully packaged within the car. This is a big deal since (a) economics limits which sensors you can buy and (b) the car body and shape limit the size and location of sensors. These obviously limit your overall sensing capability.
The new car also has better computer packaging. Most autonomous vehicles have no trunk space and frequently have no back seat room. For a historical perspective, Carnegie Mellon's Navlab 1, which found a spot and parallel parked autonomously in 1992, had racks of computers and an extra air conditioning system to handle the heat load. Urban Challenge vehicles also had racks in their trunk areas. The Cadillac SRX team was able to cram all the computational gear out of sight. This is really Moore's Law, etc but it is still a respectable achievement.
top Cadillac SRX Converted Into Self-Driving Car
Actually, Google built the All-Star team from the Urban Grand Challenge. The Google group has lots of CMU (winner) and Stanford (2nd place) team members, including the technical lead from CMU.
top SOPA Creator Now In Charge of NSF Grants
It's not that simple. A lot of groundbreaking work is the result of side project within a larger research effort. Google is a good example of this. The ideas and approach had their origins in the
NSF project Larry and Sergey were working on. While the SDLP project probably had an impact on digital libraries, the stated goal of the work, the larger impact was the creation of a technology behemoth with thousands of US jobs and a major influence on the digital economy. Using your model, would Google have happened? Probably not.
Also, the way you posed the question is interesting for other reasons. Whether a person changes their behavior is often based on far more than just basic science and technology advancements. Issues like federal policy (political science) can have a huge impact. For example, I'm working on technology research related to the aging of the population. This is a very real societal need and it is easy to justify the work from a financial perspective (take a look at nursing home and caregiver costs). However, many health and independence technologies are intertwined with privacy, whether Medicare will pay, and other non-technical issues. We rely on the insights of our colleagues doing research in the social sciences to help us understand the interplay between functionality and barriers to acceptance and commercialization. Without their research, we'd probably make very expensive paperweights.
about a year and a half ago
top Researchers Analyze Twitter To Find Happiest Parts of the United States
No if you're posting from a vacation destination you're probably happy.
Exactly. If the researchers didn't account for traveling behavior (i.e., check to see if the person was posting from their typical geographical region) then the results would be heavily skewed by vacations. Hawaii, Maine, Nevada, Utah and Vermont are all popular vacation locations.
about a year and a half ago
top Researchers Opt To Limit Uses of Open-access Publications
It won't stop them, but it gives you the ability to get the plagiarized material pulled. Remember, plagiarism is antithetical to academia. Mashups are fine, as long as you give appropriate credit.
about a year and a half ago
top Ask Slashdot: What Would It Take For Developers To Start Their Own Union?
... which, of course already
top French Science and Higher Education Programs Avoid Austerity
Research is -- on the scale that government or really large corporations operate -- cheap. It is a relatively small portion of the budget and yet returns value over decades and centuries.
I'm not so sure about that.
PhD Comics broke down the FY 2009 budget. It's worth taking a look. $68B out of $3,518B, or about 19%. Now, having said this, there are impacts from government research funds that reach far beyond a large corporation research budget. Research can produce massive savings that last decades, grow whole new industries, or create entirely new "really large corporations" (Google came from an NSF grant). A typical corporate lab is not focused on problems that lead to societal-level savings or whole new fields. They lack the large-scale funds and are often constrained to the company's best interest. Government funded research generally covers a sector of the research map that companies avoid due to risk, long-term payback, or scale.
top Stem Cells Turn Hearing Back On
An important distinction is that Deaf Culture members are not speaking on behalf of all deaf people. There are many, many deaf and hard of hearing people who use cochlear implants and hearing aids. Most of this population were raised by hearing parents who opted for their child to grow up in their world, rather than the world of the DC-oriented state school for the deaf system. Since 90% of the children born deaf have hearing parents, it is not surprising that many of their parents choose an oral, mainstream route. If an insular community that spoke a different language told you that, because of a physical feature on your new baby, your child should grow up in a culture other than your own, would you? It is also hard to dissuade a hearing parent of a newborn when they see an older implanted kid talking and singing - the evidence is staring them in the face.
While some kids in the oral mainstream education path end up migrating away from technology for various reasons, most stay on this track and are very technology friendly. This isn't surprising given the outcomes. Extensive, longitudinal research shows the vast majority of children implanted with a CI in their first few years and enrolled in an oral school (e.g.,
Option Schools) are mainstreamed into regular classrooms by kindergarten/1st grade. Mainstreaming is a huge predictor of English reading literacy (ASL is not English), which as we all know, is important for many higher income employment opportunities. You don't hear about this population because most of them melt into society.
The advance mentioned in TFA is likely to receive the same attack the DC crowd is waging on cochlear implants. They claim deaf kids should make the decision for themselves. This is a smokescreen. Kids implanted after the early language development windows (pre-5) have a much harder time learning to understand and use the sound provided by the implant due to reduced brain plasticity. If they are much older, they are also less likely to be mainstreamed and therefore behind the curve on literacy. Therefore, it is not surprising kids who are "given a choice when they are older" would have poorer outcomes and are more likely to abandon technology.
Having said all this, many adult cochlear implant and hearing aid users are unlikely to opt for this advance. If they are like my wife, they are comfortable with their hearing loss, get good use out of their cochlear implant, and don't see a strong need to change. However, this advance would have a huge impact on newborns and kids still in the language development window.
top DHS Best-and-Brightest STEM Program Under Fire
I've seen OPT used properly and effectively for very talented foreign students. I've been around very good universities and I can confirm OPT is critical at keeping top-tier foreign students here in the US. The most common cases are (a) the summer grad school gap when changing schools and (b) a gap between graduation and an employment visa. The former may seem trivial, but it can allow a student to finish up a research project at University A before moving on to University B (e.g., undergrad to grad, MS to PhD, etc). Losing three months of an integrated, talented student has significant impact on a research project. For the latter, I've met numerous students who used OPT as key step towards gaining eventual permanent status.
top Subdermal Magnets Allow You To Wear an IPod Like a Watch
Magnets are used to keep the
internal and external antennas aligned for cochlear implants. CIs have been around for a while, so the community has learned some important lessons. First, you need to plan for magnet removal in the event of an emergency MRI. Most CI users don't get MRIs but sometimes there is a critical need. Therefore, newer implant models allow a qualified doctor to make a small incision, pop the magnet out without damaging the implant, and then put everything back after the MRI. This is extremely rare for obvious reasons.
Second, the magnet in the external component is usually tailored to the individual. The need for different strengths is due to magnet depth, hair, etc. There are several strength levels (e.g., high, medium, low) and you want one that will hold the coil tight, but not so tight that it leads to skin damage.
top Blind Man Test Drives Google's Autonomous Car
I want to see the CVs of these engineers first. On how many high-integrity systems have they worked so far? I know plenty of people with an AI background, and trust me, I don't want these to program my car. I'd also need to know which programming languages and development tools they have used, see the source code, and would like to know which formal software and hardware verification methods were used to verify the code.
I'm fortunate enough to know some of the people on the team. They come from robotics backgrounds and are very experienced in highly reliable systems. For example, a big chunk of the Google team was hired out of the Carnegie Mellon Tartan Racing team. That team tested their systems so much that they wore out hardware. The Carnegie Mellon team also originates from the Field Robotics Center which sends robots to Antarctica, volcanoes, sinkhole lakes, the mountain deserts of Chile, and abandoned mines. These are hardened systems developed using hard core systems engineering and serious software engineering methods. Trust me, this is not a group of ivory tower AI people ignoring version control and using latest beta release of LISP. One of the technical leads, Chris Urmson gave a talk last year at Carnegie Mellon. It was apparent that they have continued these practices in Google. The incredible volume of field testing is the outward evidence.
top Pay the TSA $100 and Bypass Airport Security
Actually, the screening is Global Entry, which is also associated with border crossing things like NEXUS and SENTRI. These programs are very popular for people who drive over the US border a lot since they let you go through a faster customs line. For example, NEXUS can shave 20+ minutes off a border crossing in/out of Canada (it works both ways). The time savings can really add up if you drive across the border a lot. Also, the fee is only once every few years.Therefore, people of all classes get NEXUS, SENTRI, etc. For example, when I went for my interview the waiting room was overwhelmingly middle class. These programs automatically enroll you into Global Entry. Global Entry gets you faster entry into the US when arriving through certain airports, which is a nice byproduct for people living in border towns.
In short, most people get Global Entry for border crossing and could care less about TSA. Given the background check needed for these Global Entry programs, it is not surprising that someone finally realized they could re-use the prescreening for TSA purposes.
top Carmakers Prepare For Augmented Reality Driving
In reality, the auto companies and their partners in university labs have been doing work research on HUDs for a while. The augmented reality approach has been tried in research studies as a result of successes in the aviation community. However, there are huge differences between augmented reality for cars when comparing to planes and pedestrians. The point of this post is not that HUDs are bad or unlikely to succeed, but rather that the designers of trade show concepts are ignoring much of the existing research. The concepts in TFA are unlikely to be used for actual products due to safety issues. Expect simpler HUDs focused on safety oriented problems. Here are some of the safety problems:
First, cars tend to hit things quicker. This is a crude point about recovery time, but a major one.
Second, there is considerably more variation in scene brightness due to driving speeds and local factors like buildings and trees. This leads to challenges perceiving the HUD imagery. Demos on trade show floors and labs usually gloss over this factor.
Perhaps the biggest concern is that there humans have perception errors due to the way our brains integrate augmented reality with the real world. First is the issue of cognitive capture. This is when you ignore the real world and just use the HUD for your information. For example, the collision warning system may highlight all the moving vehicles, so you learn to just look for the highlighting. Unlike a video game where every object is known, automotive sensing doesn't work 100% and objects will be missed. Cognitive capture is when you fail to perceive the kid running into the middle of the street because he wasn't highlighted. This can be demonstrated easily in the lab and many studies have concrete evidence of this.
The second perception problem is that HUDs can lead to misperception of distance. A HUD can only have one focal length while the real world has an infinite amount. Mismatches can lead to the driver misinterpreting the distance of an object. This isn't a problem when flying (everything is at optical infinity) or walking (you're moving to slow), but can cause problems when driving.
The third perception problem is masking. This is when the information about the new cafe covers the pedestrian crossing the street.
If your organization has access to
this paper, it is an excellent primer on the issues. And yes, it was written in 1997.