upstart writes:The Future of the NTFS Linux Driver as Part of the Kernel is in Question:Support in the Linux kernel for NTFS, the primary filesystem for Windows systems, has always been important for people who use both operating systems. The existing Linux NTFS driver has been unmaintained and has always lacked proper write support. A filesystem in userspace (FUSE) driver, NTFS-3G, came along, but since it operates in userspace, it isn't considered particularly fast.
upstart writes:The Founder and CEO of Canonical, the company behind the popular Linux Distribution Ubuntu, indicated on Thursday that the company will probably have its IPO sometime in 2023:
Fnord666 writes:Apple is apparently removing applications from its App Store that haven't been updated recently. I personally have several applications published that are simple, free utilities. Like the developer in the article, my applications are complete and have no need to be updated. In fact, in order to update them at this point I would have to buy a new Apple developer license ($99US) in order to publish an update. Fortunately they won't need much, if any, code changes to bring them up to date. It's just irritating that I will need to pay again to keep my apps published.Devs Are Up in Arms After Apple Says It Will Remove Games That Haven't Been Updated
Freeman writes:(Apologies in advance for the Facebook link)https://arstechnica.com/information-technology/2022/04/the-first-meta-store-is-opening-in-california-in-may/
hubie writes:Beams of protons are again circulating around the collider's 27-kilometre ring, marking the end of a multiple-year hiatus for upgrade work
hubie writes:Hostile Media Perception theory (HMP) is a theory about mass communication that says a partisan perceives bias when presented with neutral coverage of news from a source deemed to be opposite to their political leanings. It also suggests that reading news from a source perceived as politically biased might decrease their willingness to share it with others and vice versa. A paper in Royal Society Open Scientist reports on tests conducted to measure this effect. They took two "hot button" topics, police conduct and COVID-19 restrictions, and presented them to people as a headline and short report. The news items presented were real stories and presented in a neutral manner, but they manipulated the banner graphic on top of the headline to appear that it came from either Fox News or CNN.Their results showed that perceptions that a news source is biased depends upon both the political leaning of the viewer as well as particular topics being reported:
hubie writes:A recent paper in the Proceedings of the National Academy of Sciences (PNAS) claims to have developed a machine learning model that can infer over 30 personality or psychological traits of a person from simply looking at a picture of them. They used deep generative image models to create photorealistic pictures of different faces and combined that with over one million judgements to infer physical traits such as age and happiness, but also personality traits such as trustworthiness, smart, liberal/conservative, Middle-Eastern, gay, and dorky.One of the authors (Joshua Peterson) announced the paper in a Twitter thread. He noted:
looorg writes:Since we mentioned that the C64 got middle age (or however you see 40 as) one might also note that the European rival the ZX Spectrum also just turned to (on the 23rd of April). While it might not have been big in America it was fairly popular over in Europe, and certainly then in the UK. More of a rival over here then all this talk about the Apple II etc.https://www.theregister.com/2022/04/22/spectrum_at_40/
upstart writes:Planting Undetectable Backdoors in Machine Learning Models:These days the computational resources to train machine learning models can be quite large and more places are outsourcing model training and development to machine-learning-as-a-service (MLaaS) platforms such as Amazon Sagemaker and Microsoft Azure. With shades of a Ken Thompson speech from almost 40 years ago, you can test whether your new model works as you expect by throwing test data at it, but how do you know you can trust it, that it won't act in a malicious manner using some built-in backdoor? Researchers demonstrate that it is possible to plant undetectable backdoors into machine learning models. From the paper abstract:
hubie writes:Google marked Earth Day 2022 with a Doodle consisting of animated GIFs showing time-lapse images of four scenes: glacial retreat at the peak of Mount Kilimanjaro in Tanzania between December 1986 and 2020 and in Sermersooq, Greenland between December 2000 and 2020, a coral bleaching event on the Great Barrier Reef between March 2016 and October 2017, and deforestation of the Harz forests in Elend, Germany, between December 1995 and 2020.
upstart writes:How Bitcoin mining devastated this New York town:When specialized ASICs optimized for crypto mining went on the market, a processor arms race began. Plattsburgh, in upstate New York, had some of the cheapest electricity rates in the country and crypto miners beat a path to their town to set up shop. In 2018 the town was receiving a major crypto mining application every week.
upstart found an interesting article written by former Commodore engineer Bil Herd over at Hackaday:Commodore C64: The Most Popular Home Computer Ever Turns 40: