-Sn!PeR-

AMD has listed three new Ryzen 7030 Series "Rembrandt" Zen 3+ CPUs with their iGPUs disabled: the Ryzen 5 7235H, Ryzen 5 7235HS, and Ryzen 7 7435H. All three were quietly listed on AMD's website and are listed for global availability. This news, first spotted by WccfTech and VideoCardz (for the Ryzen 7), comes following the Ryzen 8000 Series "F" desktop CPUs with their iGPUs removed, but those were exclusive to China. As their names imply, the Ryzen 5 7235H and 7235HS CPUs are astonishingly similar— their specs are effectively identical, so the difference may be for marketing purposes. The official specs listings being referred to don't list overclocking being enabled on either of these CPUs, though an earlier listing mistake showing iGPU support has been fixed. According to VideoCards, the Ryzen 7 7435H was previously used in laptops from Lenovo and was also poised to show up in gaming machines from Mechrevo. The gaming bent makes sense, as those systems almost always come with discrete graphics. Wccf speculated that these could be worked into a competitive place in mid-tier CPU offerings and contacted AMD for official word on pricing, but received nothing by time of writing. We've seen AMD use previous-gen cores before, so the practice isn't surprising. These CPUs are likely going to be targeted at mobile platforms, which in the PC space now means a combination of laptops and mini PCs. Some Mini PCs have already been seen putting Rembrandt to good work, with the Ryzen 6900HX present in the Sibolan SZBox S69 still providing good iGPU performance for today's currently-low standards. Improved competition from Intel will soon change that, of course. As always, it takes both price and performance to measure the value of a chip, so we can't really attest to either just yet. Source.

Voted

I'm not sure about that, but I think if you have something that proves you are the owner of that account, you can contact one of the administrators to find you a solution.

SMU Associate Professor Christoph Treude examines the foundations for studies on open-source software and protestware. "Software developers don't develop everything from scratch," he says. "Just like car manufacturing, you rely on pieces that have been manufactured by others. So, it's the same with software developers, whether in the open source world or industry. They tend to re-use a lot of stuff that others have done." Open source ecosystems can contain millions of individual items. So what happens if someone adds malware to their particular piece of software to protest, say, the war in Ukraine? Well, that has happened, with the result that some users in Russia and Belarus have had their computers hacked. For instance, the developer behind software library node-ipc with its more than a million weekly downloads tried to replace all the files on the computers of users in Russia and Belarus with a heart emoji back in March 2022. "Because of the interconnectedness of the software ecosystem, people who contribute or maintain just one piece of the gigantic puzzle can have quite a bit of power." Sometimes, a maintainer, the main person driving an open source project, may make an honest mistake when developing software, Professor Treude says. "But more recently, with the war in Ukraine, if maintainers want to raise awareness about something specific, they turn their open source project into malware." In extreme cases, he says, "they've re-programmed the library purposefully to attack machines located in Russia and Belarus." Others take less drastic action and merely introduce a message or document "urging support for whatever side they're on." Identifying the main types of protestware In a paper titled 'In War and Peace: The Impact of World Politics on Software Ecosystems', which was presented at a software engineering conference more than a year ago, Professor Treude and his co-researcher Raula Gaikovina Kula from Japan's Nara Institute on Science and Technology identified three main types of protestware: Malignant protestware—software that intentionally damages or takes control of a user's computer without their knowledge or consent. Benign protestware—software created to raise awareness of a political or social issue but does not take control of the user's device. Developer sanctions which affect a software ecosystem more broadly. For instance, MongoDB decided not to sell its products to Russian users, and GitHub suspended Russian accounts. 'A loss of trust' Professor Treude says the role of open source in software engineering has shifted over the past decade. In the early days, major corporations such as Microsoft were opposed to open source software "as they believed software should be sold for money and should not be available to everybody for free." However, Microsoft eventually became a major contributor to open source, maintaining its own libraries. "Of course, open source doesn't require you to pay anything when you use it. That's the whole idea. But then maintainers realized that companies are making money based on the code they've written." In addition, he says, companies are also attaching business licenses to the open-source software "to ensure that others can't make money from it." "It's just against all the fundamental ideas of open source, but maybe we're seeing a reconfiguration of how all this fits together, and a lot of it has to do with a loss of trust." And, presumably, the emergence of protestware has also led to a loss of trust in open-source software. "Absolutely. We're wondering if we're slowly seeing the signs of the heyday of open source ending," Professor Treude says. "Maybe not ending, but people are getting less excited about it because of incidents such as protestware." Perhaps, then, some sort of certification process may be required to establish the provenance of open-source software. "That's a really interesting question because, on the one hand, that goes against some of the core ideas of open source that everything is free and open to everybody. On the other hand, maybe we do need some sort of liability when things go wrong." However, open source licenses do not "necessarily tell you if you're allowed to develop protestware or not," and there is no legal liability. Another option is that because there is so much information about programmers available on the internet, end users should do some digging before downloading software. "You can find out which libraries they maintain and have contributed to, how long they've been active, how reliable they are, and how quickly they've fixed security vulnerabilities and so on. There's a whole research area in my field called mining software repositories, which focuses on questions such as these." "Because we do have the data available, we might be able to use that information to help establish trust. Just by looking at someone's history, you can't necessarily predict the future, but it helps." Promoting ethical responsibility Professor Treude's latest paper, "Ethical Considerations Towards Protestware", which has been accepted by a practitioner magazine, examines several different ethical frameworks in relation to protestware: Duty Ethics—effectively a sense of duty (linked to Kant's Categorical Imperative) which would mean, by implication, that the injection of malware would not be ethical. Consequential Ethics—you should take into account the consequences of your actions (and seek to maximize overall happiness). Although the damage caused by malware might be seen as short-term, its long-term effects would be much larger. Principlism—taking a more pragmatic approach by following a fixed set of principles such as respect for autonomy and justice. This paper, written with co-researchers Marc Cheong of The University of Melbourne and Raula Gaikovina Kula, explores various ethical frameworks, but Professor Treude says there is no right answer. "There is the whole concept of 'do no evil' which stuck with me when we were writing this up. If you know that what you're doing will have negative consequences for a lot of people, I would personally think twice about doing it, but obviously, there are people who think the greater good is more important." The role of education Education could play a crucial role here, Professor Treude says. "A lot of computer science and software engineering programs, even at universities, barely cover ethics these days, and even if they do, they're often dealing with topics with no connection to computer science. But in protestware, we have something concrete that anybody with a software development background would be able to relate to. So, I think using examples like this in education would be useful." "I was hired specifically to look at the human and social aspects of software engineering. Software development is changing quickly at the moment thanks to AI (artificial intelligence) programs like ChatGPT, and a lot of the progress has focused on the technical side. But very few look at how we can actually empower humans to write better software. How can we get that interface between humans and AI right? That's what I see myself working on for the next little while." Source.

Typically, the more charges a NAND flash cell can hold, the less durable it is in terms of program/erase cycles. But 3D NAND material innovations, advancements of NAND controllers, and error correction algorithms can significantly increase the number of P/E cycle a NAND flash cell can sustain. This is what happened with Yangtze Memory Technologies Co.'s X3-6070 3D QLC device that boasts the endurance of 3D TLC ICs, ITHome reports. YMTC's X3-6070 3D QLC NAND device belongs to the company's fourth Generation products and features 128 active layers as well as Xtacking 3.0 architecture with a 2400 MT/s interface. While 128 active layers do not seem like a record by today's standards, one of the key part about this 3D QLC NAND device is that the manufacturer claims rather significant endurance of 4,000 program/erase cycles for this IC. The fast interface supported by these devices makes them suitable for the best SSDs featuring a PCIe 4.0 or PCIe 5.0 interface. Back in the day, it was believed that 3D TLC NAND could sustain 1,000 to 3,000 P/E cycles, but advancements of materials, controllers, and ECC methods increased that number to 4,000 P/E cycles and beyond. 3D QLC NAND was originally thought to be capable of 100 to 1,000 P/E cycles, but all makers of memory increased that number, as well. YMTC is perhaps the only company to advertise that its 3D QLC NAND is as durable as 3D TLC NAND, other makers of flash memory are surely also making progress in this direction. Without any doubt, 4,000 P/E cycles is a great achievement for YMTC's X3-6070 3D QLC NAND memory devices. It remains to be seen how competitive these flash memory ICs are considering the fact that they have 128 active layers, whereas YMTC's competitors offer QLC 3D NAND devices with 176 or more active layers. YMTC currently uses its X3-6070 3D QLC NAND memory for its PC41Q consumer-grade solid-state drive. This SSD boasts a sequential read/write speed of 5500 MB/s and a data retention period of one year at 30 degrees Celsius, which is on par with TLC solid-state drives. With endurance of 4,000 P/E cycles, Yangtze Memory is looking forward to use 3D QLC NAND flash for enterprise-grade storage devices. Source.

Name of the game: Islets Price: $4.99 - FREE Link Store: https://store.epicgames.com/en-US/p/islets-5f2670?epic_affiliate=lowcygier Offer ends up after X hours: Sale ends 4/4/2024 at 3:00 PM Requirements:

Voted

Contra, as the boys said, you have absolutely zero activity in this project, neither other projects. try to be active before requesting to hold some sort of rank.

Music title: Future, Metro Boomin, Kendrick Lamar - Like That Signer: Future, Metro Boomin Release date: 3 day ago Official YouTube link: https://www.youtube.com/watch?v=N9bKBAA22Go Informations about the signer: - Your opinion about the track (music video): FIRE++

WhatsApp is currently working on a feature that will let users edit their images using an editing tool that is powered by artificial intelligence (AI), according to details shared by a feature tracker. Users might be able to quickly modify an image's background, restyle it, or 'expand' it using AI, when the feature is rolled out to users in the future. Meanwhile, the company is also working on a feature that will let users ask questions to the company's 'Meta AI' service directly from the search bar. According to feature tracker WABetaInfo, a feature tracker with a good track record of unearthing new features on the messaging app before they are rolled out, the latest WhatsApp beta for Android 2.24.7.13 update contains code for an AI-powered image editor. The feature, which is still in development, cannot currently be tested by users who have signed up to receive beta versions of the app. In a screenshot published by WABetaInfo, an early version of the feature is visible on the interface that is shown when sending images on WhatsApp for Android. A green icon located at the top, to the left of the HD icon is visible, and tapping it displays three options Backdrop, Restyle, and Expand. As the feature is still being developed, it is currently unclear what each of these features does. Meanwhile, the more recent WhatsApp beta for Android 2.24.7.14 version contains details of another feature discovered by WABetaInfo. The feature tracker has spotted the ability to use the search bar at the top of the app to ask queries to Meta AI — the company's generative AI assistant for Meta products designed to compete with OpenAI's ChatGPT. It's worth noting that both these features are still being developed so you won't be able to test them out, even after updating to the latest version of the app. These features are likely to be refined and improved, then rolled out to testers on the beta channel, before they are rolled out to all users. They are also expected to make their way to users on iOS, providing feature parity across both mobile platforms. Source.

ETRI's researchers have unveiled a technology that combines generative AI and visual intelligence to create images from text inputs in just 2 seconds, propelling the field of ultra-fast generative visual intelligence. Electronics and Telecommunications Research Institute (ETRI) announced the release of five types of models to the public. These include three models of 'KOALA,' which generate images from text inputs five times faster than existing methods, and two conversational visual-language models 'Ko-LLaVA' which can perform question-answering with images or videos. The 'KOALA' model significantly reduced the parameters from 2.56B (2.56 billion) of the public SW model to 700M (700 million) using the knowledge distillation technique. A high number of parameters typically means more computations, leading to longer processing times and increased operational costs. The researchers reduced the model size by a third and improved the generation of high-resolution images to be twice as fast as before and five times faster compared to DALL-E 3. ETRI has managed to reduce the model's size(1.7B (Large), 1B (Base), 700M (Small)) considerably and increase the generation speed to around 2 seconds, enabling its operation on low-cost GPUs with only 8GB of memory amidst the competitive landscape of text-to-image generation both domestically and internationally. ETRI's three 'KOALA' models, developed in-house, have been released in the HuggingFace environment. In practice, when the research team input the sentence "a picture of an astronaut reading a book under the moon on Mars," ETRI-developed KOALA 700M model created the image in just 1.6 seconds, significantly faster than Kakao Brain's Kallo (3.8 seconds), OpenAI's DALL-E 2 (12.3 seconds), and DALL-E 3 (13.7 seconds). ETRI also launched a website where users can directly compare and experience a total of 9 models, including the two publicly available stable diffusion models, BK-SDM, Karlo, DALL-E 2, DALL-E 3, and the three KOALA models. Furthermore, the research team unveiled the conversational visual-language model 'Ko-LLaVA,' which adds visual intelligence to conversational AI like ChatGPT. This model can retrieve images or videos and perform question-answering in Korean about them. The 'LLaVA' model was developed in an international joint research project with the University of Wisconsin-Madison and ETRI, presented at the prestigious AI conference NeurIPS'23, and utilizes the open-source LLaVA(Large Language and Vision Assistant) with image interpretation capabilities at the level of GPT-4. The researchers are conducting extension research to improve Korean language understanding and introduce unprecedented video interpretation capabilities based on the LLaVA model, which is emerging as an alternative to multimodal models including images. Additionally, ETRI pre-released its own Korean-based compact language understanding-generation model (KEByT5). The released models (330M (Small), 580M (Base), 1.23B (Large)) apply token-free technology capable of handling neologisms and untrained words. Training speed was enhanced by more than 2.7 times, and inference speed by more than 1.4 times. The research team anticipates a gradual shift in the generative AI market from text-centric generative models to multimodal generative models, with an emerging trend towards smaller, more efficient models in the competitive landscape of model sizes. The reason why ETRI is making this model public is to foster an ecosystem in the related market by reducing the model size, which traditionally would require thousands of servers, thereby facilitating usage among small and medium-sized enterprises. In the future, the research team expects high demand for Korean cross-modal models that integrate visual intelligence technology into prominent open-language models of generative AI. The team highlighted that the core patent of this technology is based on knowledge distillation, a technology that enables small models to perform the role of large models by accumulating knowledge using AI. After making this technology public, ETRI plans to transfer it to image generation services, creative education services, content production, and businesses. Lee Yong-Ju, director of ETRI's Visual Intelligence Research Section, stated, "Through various endeavors in generative AI technology, we plan to release a range of models that are small in size but excel in performance. Our global research aims to break the dependency on existing large models and provide domestic small and medium-sized enterprises with the opportunity to effectively utilize AI technology." Professor Lee Yong-Jae from the University of Wisconsin-Madison, who oversees the LLaVA project, mentioned, "In leading the LLaVA project, we conducted research on open-source-based visual-language models to make it accessible to more people, competing against GPT-4. We plan to continue our research on multimodal generative models through international joint research with ETRI." The research team aims to showcase world-class research capabilities, moving beyond the conventional types of generative AI that convert text inputs into textual responses. They plan to extend their research to types that respond with sentences to images or videos, and types that respond with images or videos to sentences. Source.

Huawei and presumably China's Semiconductor Manufacturing International Co. (SMIC) have submitted patents for a chip production method called self-aligned quadruple patterning (SAQP). The ultimate aim is to produce chips on a 5nm-class process technology, according to a Bloomberg report. The very same method was a major reason for the failure of Intel’s 1st Generation 10nm-class process technology, but Huawei and SMIC have no choice but to use quadruple patterning as they do not have access to leading-edge production tools due to U.S. export rules. Usage of SAQP could enable SMIC to build chips on sub-10nm technologies (we are talking about SMIC’s rumored 5nm fabrication process) despite U.S. efforts to limit China’s capabilities in the field of advanced semiconductor production. In Intel’s case, SAQP technology was meant to eliminate reliance on high-end lithography, specifically extreme ultraviolet (EUV) lithography machines made by ASML. In the Huawei and SMIC case, quadruple patterning is the only technique that increases transistor density using the tools that the contract chipmaker already has. The SAQP method described in Huawei's patent application involves etching lines on silicon wafers multiple times to boost transistor density, reduce power consumption, and potentially increase performance. This approach could allow for the production of more sophisticated chips than those that SMIC already produces for Huawei, including the Kirin 9000S. SiCarrier, a state-backed chipmaking gear developer working with Huawei, has also been granted a patent that involves multipatterning, according to Bloomberg, which again confirms SMIC's plans to use this technology for its next-generation nodes. Despite the potential of quadruple-patterning technology to enable China to manufacture 5nm-class chips, experts like Dan Hutcheson, vice-chairman at TechInsights, believe that China will eventually need to acquire or develop EUV machines for long-term competitiveness beyond 5nm-class nodes. If Huawei and its partners resort to alternative methods for semiconductor production, their cost per chip may exceed what is economically feasible for commercial devices like PCs and smartphones. Then again, processors used for supercomputers, which in turn could be used for developing weapons including weapons of mass destruction, don't need to pay heed to commercial economic restrictions. While developing weapons may be important for China, the advancements in chip technology are crucial for the country's economy. Nowadays it is largely based on demand from within the country and somehow buyers in China would like to possess competitive products, such as an iPhone from Apple. This is where Huawei may use the tech, for consumer system-on-chip design, but its chipmaking partner SMIC has to catch up and this ultimately involves acquiring advanced chipmaking process technologies. Source.

Name of the game: God of War Price: $49.99 - $24.99 Link Store: https://store.epicgames.com/en-US/p/atom-eve-ff090d?epic_affiliate=lowcygier Offer ends up after X hours: SPECIAL PROMOTION! Offer ends 28 March Requirements:

Live Performance Title: TRAVIS SCOTT FE!N GRAMMYS PERFORMANCE WITH PLAYBOI CARTI 2024 Signer Name: Travis Scott Live Performance Location: idk Official YouTube Link: https://www.youtube.com/watch?v=4lRFxruI3h8 Your Opinion About the Track (Music Video): good, kinda overrated

Music title: Future, Metro Boomin - We Don't Trust You Signer: Future, Metro Boomin Release date: 1 day ago Official YouTube link: https://www.youtube.com/watch?v=ntP8vvswhYM Informations about the signer: - Your opinion about the track (music video): FIRE

Dragon's Dogma 2 has finally arrived, and the reception has been… less than favourable. On Steam it now boasts more than 13,000 user reviews, and the majority of them are negative. Only 39% of players leaving reviews have given it a positive rating. It's not off to a good start. This "Mostly Negative" rating is down to a couple of things, the most dominant being the presence of microtransactions. As Harvey noted earlier today, the microtransactions are very silly, but they are largely for things that you can acquire very easily with hardly any work. They are entirely unnecessary. This, of course, makes them even more sus, because their existence tricks players into thinking they have to spend money to acquire more rift crystals or the ability to change their appearance, which is nonsense. The optics are bad, and players are understandably pissed. While I think some of the furore is overblown, and we've been living with microtransactions for a long time, Capcom has done the RPG no favours. This is not a cheap game, and giving players a long list of extra charges for stupid things like camping gear and wakestones (all easily found in-game) was obviously going to piss them off. Dragon's Dogma 2 and its players deserve better. Another issue players are disappointed with is the poor optimisation, which we noted in our Dragon's Dogma 2 review and the accompanying performance analysis. Capcom itself has acknowledged the issues with CPU load, but there's no word on how long we'll need to wait to see some improvement. Your mileage will vary here. In my two playthroughs, I only really encountered big frame rate dips in Vernworth, the busiest city, and while it was annoying, it really had little impact on my overall experience. That said, I've been playing on a rig with an RTX 4090 and Intel i9-13900k—it can usually handle pretty much anything. I confess I don't get too bent out of shape about frame rate dips, or even the odd stutter, if the rest of the game is good enough. One of my all-time faves, Dishonored 2, had a host of performance issues at launch, and it didn't stop me from becoming absolutely besotted with it. Games are more than their frame rates. But, again, it's perfectly reasonable to expect a £50 game from a major studio to run smoothly on modern hardware. These issues did not suddenly crop up on launch day, or when review code went out. Capcom knew what state the game was in. Source.

Nickname: -Sn!PeR- Video author: Tentango Name of the game: Minecraft Link video: Rate this video 1-10: 8

OnePlus has been delivering its so-called ‘Core Experience' for years with its Nord CE series of phones. While this experience ensured a smooth and lag-free software experience for the most part, it often fell short in other areas. Despite offering an average camera for years, it still made sense as an overall package. But with some rather competitive smartphones like the Realme 12 Pro+ sliding into this sub Rs. 30,000 segment, how can OnePlus keep its CE relevant? Some of these answers may be available with its newest Nord CE 4. For starters, this is a heavily-embargoed phone and so we aren't allowed to give out plenty of details (display, software, and more) as OnePlus is still in its “hype-building” phase for the smartphone. But we will try our best to give you an idea about what this smartphone is all about and what's new. For starters, there's a brand-new design. It looks and feels very different from the Nord CE 3, the Nord CE 3 Lite, and the more recent Nord 3 as well. It is available in two colourways – Dark Chrome and Celadon Marble. We received the Celadon Marble review unit which has a minty green finish that looks like marble. OnePlus states that both of these colours are meant to appeal to the youth, which is the target audience for its Nord series of devices. The phone has a fairly large footprint (as big as Pixel 8 Pro) and also feels quite heavy at 186 grams. While I appreciate the curved edge rear panel, it ends abruptly into the polycarbonate frame, which has flat sides and sharp edges. Just like the previous model, the Alert Slider is missing on the Nord CE 4 as well. And there are two more aqua-centric features about its design, which I'm not at liberty to disclose at the moment. Its display looks familiar to the previous model in terms of design but has a thick lipping around it which feels a bit obtrusive. Just like the previous model, the Nord CE 4's display is of the AMOLED variety. It sports a full-HD+ resolution and offers a 120Hz refresh rate as well. In the box, OnePlus offers a charger, a Type-A to Type-C charging cable and a rather premium-looking soft case. Coming to the factor that drives OnePlus' core experience. There's a new Qualcomm Snapdragon 7 Gen 3 SoC, which should give this smartphone quite the performance bump over previous models. The chipset is also quite power efficient, which should technically translate into better battery life. The phone has 8GB of RAM and 256GB of storage and it also offers up to 1TB of expandable storage via a microSD card slot. While I'm not allowed to disclose the battery's capacity I can tell you that it's a really large one. Surprisingly, OnePlus has also managed to deliver 100W wired charging which is the fastest for a Nord device till date. In theory, it should make for a proper road-warrior. Indeed, there's plenty more that needs to be explored until we can whole-heartedly recommend the new OnePlus Nord CE 4 over the existing competition which has also gotten quite fierce this year. So, do stay tuned for our detailed review which will be out soon. Source.

SMU Associate Professor Christoph Treude examines the foundations for studies on open-source software and protestware. "Software developers don't develop everything from scratch," he says. "Just like car manufacturing, you rely on pieces that have been manufactured by others. So, it's the same with software developers, whether in the open source world or industry. They tend to re-use a lot of stuff that others have done." Open source ecosystems can contain millions of individual items. So what happens if someone adds malware to their particular piece of software to protest, say, the war in Ukraine? Well, that has happened, with the result that some users in Russia and Belarus have had their computers hacked. For instance, the developer behind software library node-ipc with its more than a million weekly downloads tried to replace all the files on the computers of users in Russia and Belarus with a heart emoji back in March 2022. "Because of the interconnectedness of the software ecosystem, people who contribute or maintain just one piece of the gigantic puzzle can have quite a bit of power." Sometimes, a maintainer, the main person driving an open source project, may make an honest mistake when developing software, Professor Treude says. "But more recently, with the war in Ukraine, if maintainers want to raise awareness about something specific, they turn their open source project into malware." In extreme cases, he says, "they've re-programmed the library purposefully to attack machines located in Russia and Belarus." Others take less drastic action and merely introduce a message or document "urging support for whatever side they're on." Identifying the main types of protestware In a paper titled 'In War and Peace: The Impact of World Politics on Software Ecosystems', which was presented at a software engineering conference more than a year ago, Professor Treude and his co-researcher Raula Gaikovina Kula from Japan's Nara Institute on Science and Technology identified three main types of protestware: Malignant protestware—software that intentionally damages or takes control of a user's computer without their knowledge or consent. Benign protestware—software created to raise awareness of a political or social issue but does not take control of the user's device. Developer sanctions which affect a software ecosystem more broadly. For instance, MongoDB decided not to sell its products to Russian users, and GitHub suspended Russian accounts. 'A loss of trust' Professor Treude says the role of open source in software engineering has shifted over the past decade. In the early days, major corporations such as Microsoft were opposed to open source software "as they believed software should be sold for money and should not be available to everybody for free." However, Microsoft eventually became a major contributor to open source, maintaining its own libraries. "Of course, open source doesn't require you to pay anything when you use it. That's the whole idea. But then maintainers realized that companies are making money based on the code they've written." In addition, he says, companies are also attaching business licenses to the open-source software "to ensure that others can't make money from it." "It's just against all the fundamental ideas of open source, but maybe we're seeing a reconfiguration of how all this fits together, and a lot of it has to do with a loss of trust." And, presumably, the emergence of protestware has also led to a loss of trust in open-source software. "Absolutely. We're wondering if we're slowly seeing the signs of the heyday of open source ending," Professor Treude says. "Maybe not ending, but people are getting less excited about it because of incidents such as protestware." Perhaps, then, some sort of certification process may be required to establish the provenance of open-source software. "That's a really interesting question because, on the one hand, that goes against some of the core ideas of open source that everything is free and open to everybody. On the other hand, maybe we do need some sort of liability when things go wrong." However, open source licenses do not "necessarily tell you if you're allowed to develop protestware or not," and there is no legal liability. Another option is that because there is so much information about programmers available on the internet, end users should do some digging before downloading software. "You can find out which libraries they maintain and have contributed to, how long they've been active, how reliable they are, and how quickly they've fixed security vulnerabilities and so on. There's a whole research area in my field called mining software repositories, which focuses on questions such as these." "Because we do have the data available, we might be able to use that information to help establish trust. Just by looking at someone's history, you can't necessarily predict the future, but it helps." Promoting ethical responsibility Professor Treude's latest paper, "Ethical Considerations Towards Protestware", which has been accepted by a practitioner magazine, examines several different ethical frameworks in relation to protestware: Duty Ethics—effectively a sense of duty (linked to Kant's Categorical Imperative) which would mean, by implication, that the injection of malware would not be ethical. Consequential Ethics—you should take into account the consequences of your actions (and seek to maximize overall happiness). Although the damage caused by malware might be seen as short-term, its long-term effects would be much larger. Principlism—taking a more pragmatic approach by following a fixed set of principles such as respect for autonomy and justice. This paper, written with co-researchers Marc Cheong of The University of Melbourne and Raula Gaikovina Kula, explores various ethical frameworks, but Professor Treude says there is no right answer. "There is the whole concept of 'do no evil' which stuck with me when we were writing this up. If you know that what you're doing will have negative consequences for a lot of people, I would personally think twice about doing it, but obviously, there are people who think the greater good is more important." The role of education Education could play a crucial role here, Professor Treude says. "A lot of computer science and software engineering programs, even at universities, barely cover ethics these days, and even if they do, they're often dealing with topics with no connection to computer science. But in protestware, we have something concrete that anybody with a software development background would be able to relate to. So, I think using examples like this in education would be useful." "I was hired specifically to look at the human and social aspects of software engineering. Software development is changing quickly at the moment thanks to AI (artificial intelligence) programs like ChatGPT, and a lot of the progress has focused on the technical side. But very few look at how we can actually empower humans to write better software. How can we get that interface between humans and AI right? That's what I see myself working on for the next little while." Both articles mentioned in this story are published on the arXiv preprint server. Source.

Micron demonstrated its massive 256 GB MCR DIMM memory modules this week at the Nvidia-hosted GPU Technology Conference (GTC). These modules are designed for next-generation servers, including those running Intel's Xeon Scalable 'Granite Rapids' processors. Micron announced earlier this week that the new 256 GB MCRDIMMs are currently being sampled with its customers. Micron demonstrated one 'tall' 256 GB DDR5-8800 MCRDIMM at GTC (pictured) but also plans to offer MCRDIMMs of standard height for applications like 1U servers. Both 256 GB MCRDIMMs are based on monolithic 32 Gb DDR5 ICs, but the tall one places 80 DRAM chips on both sides of the module, whereas the standard one uses 2Hi stacked packages, which means that they run slightly hotter due to less space for thermal dissipation. In any case, the tall module consumes around 20W, which isn't bad as Micron's 128GB DDR5-8000 RDIMM consumes 10W at DDR5-4800. Multiplexer Combined Ranks (MCR) DIMMs are a type of dual-rank memory module that enables both ranks to operate concurrently using a specialized buffer. The buffer allows the two physical ranks to act as if they were two separate modules working in parallel, thereby doubling performance by enabling the simultaneous retrieval of 128 bytes of data from both ranks per clock, effectively doubling the performance of a single module. Meanwhile, the buffer works with its host memory controller using the DDR5 protocol, albeit at speeds beyond those specified by the standard, at 8800 MT/s in this case. Typically, modules with two physical ranks function as a single module, meaning that when the host CPU (or memory controller) retrieves data from such a module, it is limited to fetching 64 bytes of data at a time. MCRDIMMs double that, thus substantially increasing per-module capacity and performance. Considering the fact that Micron demonstrated its 256 GB MCRDIMMs at Nvidia-hosted GTC, it is likely that the company positions these products for AI servers based on Intel's Xeon Scalable 'Granite Rapid' processors. Since such machines tend to require a huge amount of memory for training, the new high-capacity modules will come in handy. Intel's Xeon Scalable 'Granite Rapid' CPUs have a 12-channel memory subsystem supporting two modules per channel. Using Micron's 256 GB MCRDIMMs, a Granite Rapid-based machine can accommodate 3 TB of DDR5 memory using 12 slots and 6 TB of DRAM using 24 slots. Source.

voted