SoylentNews

upstart writes:

Study shows how accounting practices distort economic reality:

Accounting standards don't properly reflect the difference between losses driven by investments and actual business performance shortfalls, according to new University at Buffalo School of Management research.

Forthcoming in the Review of Accounting Studies, the study found that Generally Accepted Accounting Principles (GAAP) mask the true value of companies by marking investments in intangibles like technology, brands and human capital as losses.

"Even as the economy boomed in 2019 following a decade of growth, about half of all public companies reported losses because accounting rules force them to report losses despite initial success," says Feng Gu, Ph.D., chair and professor of accounting and law in the UB School of Management. "This is just another indication of how broken accounting is. We need to treat intangible investments as real economic assets."

[...] "GAAP losers are less likely to decline, more likely to reverse their losses, and even have better future stock performance than real losers or profitable firms," says Gu. "This is an alarming consequence for a group of highly dynamic and innovative firms that are the driving force behind the growing intangible revolution in our economy."

Journal Reference:
Gu, Feng, Lev, Baruch, Zhu, Chenqi. All Losses Are Not Alike: Real versus Accounting-Driven Reported Losses, (DOI: 10.2139/ssrn.3847359)

Original Submission

Freeman writes:

https://arstechnica.com/information-technology/2023/07/openais-most-powerful-chatbot-api-rolls-out-for-all-paying-customers/

On Thursday, OpenAI announced that all paying API customers now have access to the GPT-4 API. It also introduced updates to chat-based models, announced a shift from the Completions API to the Chat Completions API, and outlined plans for deprecation of older models.

Generally considered its most powerful API product, the GPT-4 API first launched in March but has been under closed testing until now.
[...]
OpenAI also announced that "based on the stability and readiness of these models for production-scale use," it is also making APIs for Whisper, DALL-E, and GPT-3.5 Turbo "generally available." And the company expects to continue fine-tuning the models throughout the year.

Original Submission

upstart writes:

Researchers discover safe, easy, and affordable way to store and retrieve hydrogen:

Researchers at the RIKEN Center for Emergent Matter Science (CEMS) in Japan have discovered a compound that uses a chemical reaction to store ammonia, potentially offering a safer and easier way to store this important chemical.

This discovery, published in the Journal of the American Chemical Society on July 10, makes it possible not only to safely and conveniently store ammonia, but also the important hydrogen is [sic] carries. This finding should help lead the way to a decarbonized society with a practical hydrogen economy.

For society to make the switch from carbon-based to hydrogen-based energy, we need a safe way to store and transport hydrogen, which by itself is highly combustible. One way to do this is to store it as part of another molecule and extract it as needed. Ammonia, chemically written as NH₃, makes a good hydrogen carrier because three hydrogen atoms are packed into each molecule, with almost 20% of ammonia being hydrogen by weight.

The problem, however, is that ammonia is a highly corrosive gas, making it difficult to store and use. Currently, ammonia is generally stored by liquefying it at temperatures well below freezing in pressure-resistant containers. Porous compounds can also store ammonia at room temperature and pressure, but storage capacity is low, and the ammonia cannot always be retrieved easily.

The new study reports the discovery of a perovskite, a material with a distinctive repetitive crystal structure, which can easily store ammonia and also allows easy and complete retrieval at relatively low temperatures.

The research team led by Masuki Kawamoto at RIKEN CEMS focused on the perovskite ethylammonium lead iodide (EAPbI₃), chemically written as CH₃CH₂NH₃PbI₃. They found that its one-dimensional columnar structure undergoes a chemical reaction with ammonia at room temperature and pressure, and dynamically transforms into a two-dimensional layered structure called lead iodide hydroxide, or Pb(OH)I.

Journal Reference:
Jyorthana Rajappa Muralidhar, Krishnachary Salikolimi, Kiyohiro Adachi, et al., Chemical Storage of Ammonia through Dynamic Structural Transformation of a Hybrid Perovskite Compound, J. Am. Chem. Soc., 2023. DOI: 10.1021/jacs.3c04181

Original Submission

Freeman writes:

https://arstechnica.com/information-technology/2023/07/book-authors-sue-openai-and-meta-over-text-used-to-train-ai/

On Friday, the Joseph Saveri Law Firm filed US federal class-action lawsuits on behalf of Sarah Silverman and other authors against OpenAI and Meta, accusing the companies of illegally using copyrighted material to train AI language models such as ChatGPT and LLaMA.

Other authors represented include Christopher Golden and Richard Kadrey, and an earlier class-action lawsuit filed by the same firm on June 28 included authors Paul Tremblay and Mona Awad. Each lawsuit alleges violations of the Digital Millennium Copyright Act, unfair competition laws, and negligence.

[...] Authors claim that by utilizing "flagrantly illegal" data sets, OpenAI allegedly infringed copyrights of Silverman's book The Bedwetter, Golden's Ararat, and Kadrey's Sandman Slime. And Meta allegedly infringed copyrights of the same three books, as well as "several" other titles from Golden and Kadrey.

[...] Authors are already upset that companies seem to be unfairly profiting off their copyrighted materials, and the Meta lawsuit noted that any unfair profits currently gained could further balloon, as "Meta plans to make the next version of LLaMA commercially available." In addition to other damages, the authors are asking for restitution of alleged profits lost.

"Much of the material in the training datasets used by OpenAI and Meta comes from copyrighted works—including books written by plain­tiffs—that were copied by OpenAI and Meta without consent, without credit, and without compensation," Saveri and Butterick wrote in their press release.

Original Submission

upstart writes:

China curbs exports of key computer chip materials:

The Chinese government is tightening controls over exports of two key materials used to make computer chips.

From next month, special licenses will be needed to export gallium and germanium from China, which is the world's biggest producer of the metals.

It comes in response to Washington's efforts to curb Chinese access to some advanced microprocessors.

[...] On Monday, China's Ministry of Commerce said the restrictions were needed to "safeguard national security and interests".

[...] Semiconductors, which power everything from mobile phones to military hardware, are at the centre of a bitter dispute between the world's two largest economies.

[...] "I think we gain and China gains from trade and investment that is as open as possible, and it would be disastrous for us to attempt to decouple from China," [said US Treasury Secretary Janet Yellen] during an appearance before Congress last month.

upstart writes:

Expect the saga of this release to stretch out a bit over northern summer:

Linux kernel overseer Linus Torvalds has delivered the first release candidate for version 6.5 of the kernel, but warned this release may not go entirely smoothly.

Torvalds's headline assessment of rc1 is "none of it looks hugely unusual."

"The biggest single mention probably goes to what wasn't merged, with the bcachefs pull request resulting in a long thread (we didn't hit a hundred emails yet, but it's not far away)."

As The Register reported in 2022, bcachefs is a filesystem that's been in development for nigh on a decade without being added to the kernel.

[...] In his announcement post for rc1, Torvalds wrote "Let's calm this party down."

[...] The release candidate is yours for the downloading here.

Original Submission

upstart writes:

Plus: Facebook corp loses appeal on crossing data streams in Germany:

Elon Musk's Twitter can breathe easy when it comes to the European Union – Meta's "Threads" will be steering clear.

This is because Mark Zuckerberg's company – which owns Facebook, WhatsApp, Instagram, Messenger, and more – will not launch its new Twitter-like service anywhere in the EU for the foreseeable future. But yes, it has arrived in Britain, London vultures have informed us.

Ireland's data protection watchdog, which oversees all the tech giants headquartered in the corporate tax-light country, told The Register that Meta had "confirmed to the DPC that they have no plans to roll out the service in the EU at present."

While it's hitting the US and the UK tomorrow – British and American influencers have reportedly already been offered early access – this Berlin-based vulture can confirm there's none of that here, not unless you wish to pipe your data across via VPN.

[...] Data privacy laws in the US are very loose, especially on a federal level, which is part of the reason why the various frameworks for data exchange between the European Union's member states and the United States keep failing when challenged in court.

However, it might come as a surprise to some that Meta doesn't appear to anticipate any regulatory trouble in the UK.

It's a signal that at least one tech giant expects the UK's "replacement for GDPR" – the second take on the Data Protection and Digital Information Bill (DPDIB II) – to be more sympathetic than current British privacy legislation. Brexit dividend, anyone?

Meanwhile, over in Germany, Meta can't combine the data it collects about you on Facebook with the stuff you spill on Instagram and WhatsApp, or with other websites and apps you use, without getting your explicit permission.

[...] Back in January, Ireland's regulator stopped Meta from launching advertising services on WhatsApp that uses data from Facebook or Instagram.

The crossing of the data streams is not a problem for the Facebook giant in the US and UK, it seems.

Original Submission

upstart writes:

Plastic production is skyrocketing, pushing microplastic pollution to dangerous new levels:

Not even the Arctic Ocean is immune to the incessant growth of microplastic pollution. In a new study that analyzed sediment core samples, researchers quantified how many of the particles have been deposited since the early 1930s. As scientists have shown elsewhere, the team found that microplastic contamination in the Arctic has been growing exponentially and in lockstep with the growth of plastic production—which is now up to a trillion pounds a year, with the global amount of plastic waste projected to triple by 2060.

These researchers analyzed the seawater and sediment in the western part of the Arctic Ocean, which makes up 13 percent of its total area. But in just that region, they calculated that 210,000 metric tons of microplastic, or 463 million pounds, have accumulated in the water, sea ice, and sediment layers that have built up since the 1930s. In their study, published last week in the journal Science Advances, they cataloged 19 synthetic polymer types in three forms: fragments, fibers, and sheets. That reflects a dizzying array of microplastic sources, including fragments from broken bottles and bags and microfibers from synthetic clothing.

Overall, the team found that microplastic levels have been doubling in Arctic Ocean sediments every 23 years. That mirrors a previous study of ocean sediments off the coast of Southern California, which found concentrations to be doubling every 15 years. Other researchers have found an exponential rise in contamination in urban lake sediments.

[...] The atmosphere, too, is increasingly infested with microplastics. By one calculation, the equivalent of hundreds of millions of disintegrated plastic bottles could be falling on the United States alone. A study of a peatland area in the Pyrenees found that in the 1960s, less than five atmospheric microplastics were being deposited per square meter of land each day. It's now more like 180.

oumuamua writes:

[This is mix of an advertisement and a blog post, rather than an actual story, but it piqued my interest by including 3 different text-to-image generators in one package. Some of us are already learning the limits of, and skills required for, image generation, and I suspect that it will not be too long before collaborative mixes of AI language generation will be appearing, at least in the areas of research. Anyway, lets see what you make of it... JR]

Genolve.com is jumping into the generative AI game with a version upgrade that integrates three major players in AI art:

Genolve, a platform for creating professional designs, has just released a major version upgrade which integrates three popular AI text-to-image generators: DALL-E 2, Stability.ai and Midjourney! Just type in a description of the desired image and generate it in seconds for pennies per image. Still, it can be tricky to get exactly what you need and in this blog post you'll learn the tricks and techniques to coax the AI to generate mind-blowing results through prompt engineering. You'll also learn step-by-step how to AI edit your own photos to remove unwanted items or add something extra using a technique called inpainting.

Original Submission

hubie writes:

A Canadian judge ruled that a thumbs-up emoji could be considered an agreement to a contract:

A Canadian judge has ordered a farmer to pay more than $CAD82,000 ($92,000) in damages following a legal battle over what the thumbs-up emoji means.

Chris Achter, the owner of a farming company in Swift Current, Saskatchewan, had sent a thumbs-up emoji in response to a photograph of a flax-buying contract from a grains buyer in 2021.

Months later, the buyer — which had been doing business with Mr Achter for several years — did not receive the flax as expected.

That started a dispute that led to "a far-flung search" to unearth what the thumbs-up emoji means, according to the June court ruling that surfaced in local media this week.

The buyer, South West Terminal, argued that the emoji implied acceptance of contractual terms, while Mr Achter said he used it only to indicate that he had received the contract, but not to indicate his agreement.

[...] He [the judge] said: "I am satisfied on the balance of probabilities that Chris okayed or approved the contract just like he had done before except this time he used a thumbs-up emoji."

Original Submission

hubie writes:

GCC Adopts A Code of Conduct:

While a few years late compared to many other open-source projects adopting a Code of Conduct, the GCC Steering Committee has now adopted a Code of Conduct "CoC" for this open-source compiler project.

Passionate compiler developers and other GCC stakeholders are encouraged to remind themselves to be civil in their discussions and follow other recommendations to foster their community. Jason Merrill of the GCC Steering Committee wrote in their announcement of this CoC:

"The vast majority of the time, the GCC community is a very civil, cooperative space. On the rare occasions that it isn't, it's helpful to have something to point to to remind people of our expectations. It's also good for newcomers to have something to refer to, for both how they are expected to conduct themselves and how they can expect to be treated.

  More importantly, if there is offensive behavior that isn't corrected immediately, it's important for there to be a way to report that to the project leadership so that we can intervene.

  At this time the CoC is preliminary: the code itself should be considered active, but the CoC committee (and so the reporting and response procedures) are not yet in place."

The draft GCC CoC can be viewed here.

Original Submission

upstart writes:

Wildfires in Canada have broken records for area burned, evacuations and cost, official says:

Wildfires raging across Canada have already broken records for total area burned, the number of people forced to evacuate their homes and the cost of fighting the blazes, and the fire season is only halfway finished, officials said Thursday.

"It's no understatement to say that the 2023 fire season is and will continue to be record breaking in a number of ways," Michael Norton, director general, Northern Forestry Centre, Canadian Forest Service, said Thursday during a briefing.

[...] The fires have burned 8.8 million hectares (27.7 million acres) an area about the size of the state of Virginia. This already exceeds the record of 7.6 million hectares (18.7 million acres) set in 1989 and is 11 times the 10-year average experienced by this date.

"The final area burned for this season may yet be significantly higher," said Norton. "What we can say with certainty right now is that 2023 is a record-breaking year since at least since 1986 when accurate records started to be kept."

Allen said the fine particles found in fire smoke not only have the ability to penetrate deep into airways, they also can travel long distances meaning they could drift far into the U.S.

There have been reports that fires in Eastern Canada and Quebec are affecting air quality in Europe.

[...] There are about 3,790 provincial firefighters battling the blazes across the country being assisted by Canadian Armed Forces personnel. Another 3,258 firefighters from Australia, South Africa, New Zealand, the U.S., Chile, Costa Rica, Mexico, Spain, Portugal, South Korea and the European Union have travelled to Canada to fight fires.

Norton said the cost of fighting wildfires has steadily grown and is approaching about CDN$1 billion (US$750 million) a year.

Original Submission

upstart writes:

NHTSA reminds Tesla to cough up data for Autopilot probe:

An investigation by America's National Highway Traffic Safety Administration (NHTSA) into the safety of Tesla Autopilot has led to a threat of fines if Elon Musk's electric car company doesn't hand over the data requested.

If Tesla doesn't comply with the order in the NHTSA's July 3 letter [PDF], the agency said it could issue fines and penalties that could reach as high as $26,315 per violation per day – capping out at $131.5 million.

That's not to suggest that Tesla has been avoiding giving US highway regulators the data they've asked for. Documents from the investigation indicate Tesla has turned over information several times already. The NHTSA told The Register that fine warnings are a standard part of such letters no matter which manufacturer is getting them.

Among the data requested by the NHTSA is a full rundown of information on vehicles included in the investigation, which is a lot: "All Tesla vehicles, model years 2014–2023, equipped with [Autopilot] at any time."

The NHTSA wants to know the software, firmware and hardware versions of each and every Tesla that falls into its investigative purview, whether the vehicles have a cabin camera installed, when the vehicle was admitted into Tesla's full-self-driving beta, and dates of the most recent software/firmware/hardware updates.

upstart writes:

On a shoestring budget, Chandrayaan-3 aims to observe Luna, Earth, even exoplanets:

India's Space Research Organisation (ISRO) will next week launch Chandrayaan-3, a mission that aims to land on the moon and deploy a rover.

ISRO yesterday announced that Chandrayaan-3 had been tucked into its capsule and mated with the (LVM-3) launcher that will take it into space. Liftoff from the Satish Dhawan Space Centre has been scheduled for July 14 at 2:35pm IST (09:05 Friday UTC).

The ridiculously economical $74.5 million mission aims to land near Luna's south pole in August. From a ramped compartment, the lander will deploy a 26 kilogram rover outfitted with instruments including an Alpha Particle X-ray Spectrometer (APXS) and Laser Induced Breakdown Spectroscope (LIBS).

The lander contains an accelerometer, Ka-band and laser altimeters, Doppler velocimeter, star sensors, inclinometer, touchdown sensor, and cameras for hazard avoidance and positional knowledge.

The lander also boasts several instruments including Chandra's Surface Thermophysical Experiment (ChaSTE) to measure surface thermal properties, Lunar Seismic Activity (ILSA) to measure tremors around the landing site, Radio Anatomy of Moon Bound Hypersensitive ionosphere and Atmosphere (RAMBHA) to study the gas and plasma environment, and a NASA-provided passive Laser Retroreflector Array (LRA) for lunar ranging studies.

A propulsion module that carries the rover and lander will stay in a 100km lunar orbit, where it will act as a communication relay satellite, complete with a payload – known as the Spectro-polarimetry of Habitable Planet Earth (SHAPE) – that studies spectral and polarimetric measurements of Earth from roughly 362,000 to 405,000 kilometers away.

[...] Only three countries - the USA, Russia, and China, have successfully landed missions on the Moon. Good luck, Chandrayaan-3!

Original Submission

An Anonymous Coward writes:

Magic: The Gathering's most coveted collectible The One Ring Card has been found. There is only one copy and version of this card, making it highly collectable. The owner wishes to remain anonymous, and multiple resellers are already offering millions to buy it.

The One Ring, a singular, serialized, one-of-a-kind card for Magic: The Gathering, has been found. Proof comes via the grading company PSA, which posted an image of the card Friday morning.

Magic's latest set of cards, titled The Lord of the Rings: Tales of Middle-earth, was created in coordination with Middle-earth Enterprises (MEE) to celebrate the original novels by J.R.R. Tolkien. The set includes many copies of The One Ring, including perforated versions meant to be torn apart at the table. But publisher Wizards of the Coast also created a singular copy, covered in gold foil and etched with the original Elvish Black Speech inscription.

If this is up your alley then the owner is accepting serious offers via their attorney through an email address hello@thenotablegroup.com, noting that no offers above one million dollars have currently been tabled and accepted.

Original Submission