Skip to Content

Prototype

Straight from the Lab: Technology’s First Draft

Micro-Mover

A tiny motorized table that is able to slide sideways in increments measured in atom-widths could offer a way to vastly increase computers’ memory capacity. The “micro-mover,” as it’s called by its inventors at Agilent Laboratories in Palo Alto, CA, is carved into a silicon chip three millimeters wide. The mover sits on two sets of flexible legs that bend slightly under electrostatic forces; one set flexes in the north-south direction, the other east-west. In this way, the device can skate to any of a billion positions in increments of five nanometers-about the width of a dozen atoms, says Farid Matta, manager of the group that built the device. Researchers have long envisioned memory devices that would pack bits as closely as individual molecules in a solid material, but any such storage device requires “something that moves under the read-write tool to allow you to write a one or a zero in a new spot,” Matta explains. The micro-mover is sprightly and precise enough to do that with individual molecules, he says-potentially writing up to 125 megabytes of data in an area only 50 micrometers on a side.

Picture Perfect

Even Ansel Adams struggled with light, employing tricky developing techniques to preserve detail and avoid large areas of a print appearing washed out or too dark. Now, amateur photographers can produce artfully balanced images with a new digital-processing algorithm.

Developed by MIT computer scientist Frdo Durand, the process reduces excessive contrast in the picture without losing detail and takes only a couple of seconds. It starts with a digitized image, either from a digital camera or a scan of a negative, and partitions the image data in unique ways. The first partition is into color (which the algorithm does not touch) and light-intensity components. The computer next subdivides the intensity component into one comprising the picture’s details and one containing a map of large-scale variations in luminosity. The algorithm then reduces the variations in this final component. Durand hopes the technology will first be employed in software that digital photographers use to download their pictures from camera to computer, which could transform those vacation snapshots from awful to art.

Flat-Screen 3-D

Father and son inventors Paul and Ilan Kleinberger have developed a flat-screen 3-D display that eliminates the flicker, low resolution, and bulky glasses that detract from other 3-D methods. 3-D effects are created by showing slightly different images to each eye. At startup 3ality in Jerusalem, Israel, the Kleinbergers use their own software to combine left- and right-eye image data so that each pixel on a liquid-crystal display contains the information for both eyes. 3ality adds an extra liquid-crystal layer that “twists” the resulting light, giving a different polarization to each eye’s image. That way, a viewer wearing lightweight polarizing glasses gets an uninterrupted, full-resolution image for each eye. 3ality has built prototypes in cooperation with the U.S. Army; company president Yosh Mantinband says the displays should be ready to market by 2005.

Building Browser

Imagine walking down the street and browsing the contents of a store by pointing a handheld computer at it. Up pops a list of available products and services, and even comments from customers. At Hewlett-Packard Laboratories in Palo Alto, CA, engineer Salil Pradhan’s team has developed such a mobile interface. The software runs on a personal digital assistant or cell phone, automatically linking to Web sites associated with nearby buildings. Unlike tracking systems that use bar codes or radio frequency ID tags, Pradhan’s device works at distances greater than 10 meters; it calculates the user’s position and orientation using a Global Positioning System receiver and a digital compass embedded in the handheld. Pradhan is expanding the system to work indoors. A product could be available in three years, he says.

Safer Soil

Off-the-shelf devices that detect contaminants are bulky and expensive and require frequent recalibration. At the University of California, Los Angeles, engineers Tom Harmon and Jack Judy have built a tiny sensor that could improve soil monitoring of nitrates, which can leak into ground water and cause health problems. The sensor consists of a carbon rod coated with a mix of polymer and nitrate. Placing it in water or moist soil generates a voltage that corresponds to the difference between the nitrate concentrations of the soil and the coating.

The hair-thin, 1.5-centimeter-long device could be mass-produced on silicon chips, says Harmon, lowering the cost to one-tenth that of conventional nitrate detectors. Eventually, the sensor will self-calibrate based on models of how environmental factors, such as temperature, affect the signal. In five years, Harmon says, the effort could yield arrays of cheap, reliable sensors that need little human intervention once deployed.

Data-Mining Drives

You want to find that digital shot of your belly-flopping husband from last summer’s vacation, but you can’t remember where you filed it on your hard drive. What if, rather than making you open folder after folder, your hard drive acted more like a database, quickly serving up the file you want? That’s the idea behind a project in Mahadev Satyanarayanan’s Intel-sponsored lab at Carnegie Mellon University to create software for special processors inside hard drives. The software would speed searches by examining hard-drive data as they’re read and suppressing all the data that have no chance of fitting the search parameters before they reach a computer’s main processor. If assigned to look for a suspected terrorist in video surveillance data, for example, the system could block frames showing empty sidewalks. If intelligence officers, radiologists, astronomers, or other professionals who depend on data from images express interest in the group’s early simulations, then Intel could start working with them within a year or two to build specialized hard-drive software, and eventually hardware, Satyanarayanan says.

Squirm Scan

Making brain scans of children and of people with conditions like Parkinson’s disease is difficult because the patients have trouble keeping still. Soon, though, technology initially designed to let geneticists scan even-squirmier patients-mice-just might help. Being developed jointly at Oak Ridge National Laboratory in Tennessee and the Thomas Jefferson National Accelerator Lab Facility in Newport News, VA, the system tracks-and compensates for-patient movement during a scan. Reflective markers are attached to the head; an infrared strobe light illuminates these markers, and two infrared cameras monitor the reflections. The position data derived from these cameras are then used to correct the scan information by determining where an x-ray, in the case of a single-photon emission tomography scan, would have hit the detector had the head not moved. Sorting out the data allows the researchers to deliver a crisp, accurate image. Tests of the technology began on mice in March, and the team hopes a fully functional version will be helping humans in two to three years.

Aluminum Bone

Tissue engineers have made great strides in growing bone parts in the lab, but it is proving much more difficult to replace whole sections of leg or arm bones, which sustain constant pounding. Researchers at Rice University have developed a technique for growing bone tissue strong enough to withstand the stresses of everyday activity. Conventional bone-tissue engineering involves replacing lost bone with a biodegradable polymer scaffold seeded with cells. As the polymer degrades, new tissue develops. But in load-bearing parts of the skeleton, cells are constantly breaking down and forming new bone in response to mechanical stimuli. If the polymer scaffold placed in a patient’s leg is too weak, the material falls apart under this stress. To reinforce their scaffold material, Rice bioengineer Antonios Mikos and chemist Andrew Barron added nanoparticles of alumoxane (an aluminum-based compound) to a photosensitive polymer. Shining light on this blend spurs the nanoparticles to fix themselves to the polymer chains. The resulting material’s compressive strength is three times that of the polymer alone. Mikos hopes to start testing the material in rabbits this summer.

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.