JWST recently snapped this infrared test image of a star, which also shows fainter background stars and galaxies — a testament to the telescope’s power.
In early February, NASA engineers began to remotely align the 18 hexagonal segments of the James Webb Space Telescope’s primary mirror, which had been folded away for launch. The goal of this meticulous, three-month-long process is to perfectly position the mirror segments relative to each other, creating a single, smooth, 6.5-meter-wide surface that can gather and focus light from the distant cosmos.
You may recall earlier snapshots that marked previous milestones. For example, the second of seven milestones was punctuated with a shot taken before the mirrors were fully aligned; it featured multiple images of a single star. Now, NASA has announced the fifth major alignment milestone is complete. Called fine phasing, this step helped to identify and correct small differences between individual mirror segments to bring the infrared universe into sharp, clear focus.
A newly discovered data wiper malware that wipes routers and modems has been deployed in the cyberattack that targeted the KA-SAT satellite broadband service to wipe SATCOM modems on February 24, affecting thousands in Ukraine and tens of thousands more across Europe.
The malware, dubbed AcidRain by researchers at SentinelOne, is designed to brute-force device file names and wipe every file it can find, making it easy to redeploy in future attacks.
SentinelOne says this might hint at the attackers’ lack of familiarity with the targeted devices’ filesystem and firmware or their intent to develop a reusable tool.
Alpha Centauri seems almost within grasp as promising research soars into reality.
Lightsails were once a thing of science fiction, evolving through several variations over the last 40 years. Now, science fiction is becoming reality. Advances in laser technology and new ultrastrong, ultralight materials open up the possibility of venturing beyond our solar system in the not-too-distant future.
Researchers from UCLA and the University of Pennsylvania recently published two papers outlining various shapes and heat-dissipating materials they tested to evaluate lightsails beyond previous limits. The research was conducted in conjunction with the Breakthrough Starshot Initiative, a project with the goal of sending a microchip-sized probe to the Alpha Centauri system, which, at just over 4 light-years away, is the closest and possibly most habitable neighboring star system. Breakthrough Starshot plans to use a high-powered laser array to propel tiny lightsail probes through space at a top speed of some 20 percent the speed of light. Incorporated into the sails would be minuscule scientific instruments, such as cameras, magnetometers, and communicators that could beam information back to Earth as they fly through the Alpha Centuari system.
Almost 1,000 adults and children with type 1 diabetes have been given a potentially life-altering ‘artificial pancreas’ by the NHS in England as part of the first nationwide test into the effectiveness of this technology in the world.
Health and high quality care for all, now and for future generations.
We’ve seen this funky dual disk polar printer already recently, but [Heinz Loepmeier] has been busy working on it, so here’s an update. The primary focus here is nozzleboss, a blender plugin which enables the surface textures of already sliced objects to be manipulated. The idea is to read in the gcode for the object, and convert it to an internal mesh representation that blender needs in order to function. From there the desired textures can be applied to the surfaces for subsequent stages to operate upon. One trick that nozzleboss can do is to create weight maps to tweak the extrusion flow rate or print velocity value according to the pixel value at the surface — such ‘velocity painting’ can produce some very subtle surface effects on previously featureless faces. Another
trick is to use the same weight maps and simply map colours to blender text blocks which are injected into the gcode at export time. These gcode blocks can be used swap tool heads or extruders, enabling blending of multiple filament colours or types in the same object.
Some nice examples of such printing manipulation can be seen on [Heinz’s] instagram page for the project. So, going back to the hardware again, the first video embedded below shows the ‘dual disk polar printer’ fitted with a crazy five-extruders-into-one-nozzle mixing hotend setup, which should be capable of full CMYK colour mixing and some. The second video below shows an interesting by-product of the wide horizontal motion range of the machine, that the whole printing area can be shifted to a nozzle at the other end of the gantry. This enables a novel way to switch extruders, by just moving the whole bed and print under the nozzle of interest! One final observation — is that of the print surface — it does look rather like they’re printing direct onto a slab of marble, which I think is the first time we’ve seen that.
Interesting printer designs are being worked on a lot these days, here’s a really nice 5-axis prusa i3 hack, and if you want to stay in the cartesian world, but your desktop machine is just too small, then you can always supersize it.
The young man has made several statements that compromise the scientific community, but the most shocking is related to CERN and how it could have destroyed our universe.
Regarded as a genius child and listed as the “most intelligent young man in the world”, Max Laughlin surprised the world with his great intellectual abilities.
With only 13 years old he could develop from scratch its own device for energy access. A system that is capable of providing all the necessary energy without the need for oil, coal or solar energy.
Training datasets are very important for experimenting with varied data to train new AI models. However, many commonly used public data sets contain labeling errors. This makes it challenging to train robust models, particularly for novel tasks. Many researchers use techniques such as employing a variety of data quality control procedures to overcome these shortcomings. However, there is no centralized repository consisting of examples of using these strategies.
Meta AI researchers have recently released Mephisto. It is a new platform to collect, share, and iterate on the most promising approaches to collecting training datasets for AI models. Researchers can exchange unique collecting strategies with Mephisto in a reusable and iterable format. It also allows them to change out components and quickly locate the exact annotations required, minimizing the barrier to custom task creation.
The team uncovers many common pathways for driving a complex annotation activity from concept to data collection in Mephisto. In addition to improving the quality of datasets, Mephisto also enhances the experience of the researchers and annotators who created the data set.
A new approach to in-memory computing proposes a new set up to create an artificial synapse that can both store and process data.
In this blossoming era of AI, efficient computational approaches to processing and storing large amounts of data are required. However, current computer designs have inherent performance limitations.
In recent years, research has been focused on the development of alternative computing architectures that mimic the brain. These devices, called neuromorphic computers, circumvent many of the issues associated with the traditional von Neumann architecture, which has been around since 1945 and is composed of processing and memory units.
