Julian Rose’s Building Cultures, published last year by Princeton Architectural Press, contains 16 in-depth interviews with leading architects who have designed museums around the world. In 2022, András Szántó’s Imagining the Future Museum: 21 Dialogues with Architects, published by Hatje Cantz, offered a complementary glimpse into the sensibilities of a new generation of voices. (The titles share four interviewees: David Adjaye, David Chipperfield, Elizabeth Diller, and Kulapat Yantrasat) Rose and Szántó sat down with AN’s executive editor Jack Murphy to discuss the museum’s inexhaustible spatial variety and its capacity to shape civic and cultural space today. Julian Rose’s Building Cultures, published last year by Princeton Architectural Press, contains 16 interviews with architects. (Courtesy Princeton Architectural Press) AN: Julian, what are the major themes, concerns, and anxieties that you heard when interviewing architects about designing museums? Julian Rose (JR): The conversations in Building Culture grew out of my time at Artforum, so they began nearly 10 years ago in a pretty different world. In that context, one important theme was looking at the museum to understand how architecture relates to arts. Architects, either by choice or because the culture at large compels them to, are always defining what they do in relation to other cultural practices, especially the visual arts. This relationship goes back to the modernist avant-garde, and you could trace it even further. I was drawn towards architects who had deep connections to art, maybe they had even gone to art school or had a record of collaboration; not coincidentally, a lot of them have become known as museum specialists. The answers I heard were refreshing; people were not necessarily learning the lessons I expected. As an example: With Peter Zumthor, I thought we were going to have a focused conversation about the very architectural aesthetics and materials used by certain artists like Richard Serra or Donald Judd. No—he wanted to talk about the bigger picture, the emotional and philosophical connections. He’s obsessed with Walter de Maria’s landscape works like Lightning Field. Even if they don’t seem to have an obvious connection to architecture, he loves the scale and ambition. This kind of surprise happened in several conversations. The other key topic is the typological problem of the museum. As I write in my introduction, the museum refuses spatial optimization—there’s no “best” way to design one. In part, that’s because contemporary art is evolving. Look at the popularity of large-scale installations today, which require big open spaces, versus the more old-fashioned idea of a museum being the place you go to have a one-on-one moment with a masterpiece, which needs intimate galleries. Until recently, “public art” was a kind of forlorn category. It was something you might happen on in a park or a subway station, and it was separate from what most people thought of as real art, which of course was what you saw in the museum. And you went to the museum to have what was essentially a private experience of that art. Now you go to the museum to have an experience that’s both aesthetic and social—to look at art and to enjoy a public space—and I think that’s a huge part of why museums are so popular today. One of the fundamental takeaways from the book is that contemporary art is becoming more and more public, and the evolution of the art museum has been a crucial part of that shift. Artists are creating work that’s meant to be experienced by many people at once, and they need new spaces to do that. At the same time, all the architects wanted to talk about circulation, because there is a tension on some level between how we traditionally think of experiencing art and the crowds that certain museums are starting to receive. András Szántó’s Imagining the Future Museum: 21 Dialogues with Architects also features interviews with leading architects who design museums (Hatje Cantz) AN: András, how does this compare to how you approached your book? András Szántó (AS): One reason why the two books are quite complementary is that their genesis is so different. Julian, your book approaches its subjects with an interest in their relationship to art and their creative work. For me, the direction of travel was different. My talks came out of a previous book, which I did during the pandemic, for which I interviewed museum directors about how their institutions are changing. Rather than reviewing past projects, I was interested in the architects’ overall perspective on the museum as a form. Generally, there is the idea that architecture saved the visual arts from the fate of other forms of high art. And there has been a post-pandemic realization that you can do highly elitist and exclusive architecture in the language of modern design, just as you can using neoclassical architecture. We see a reckoning for how to realign museums to serve a wider segment of the population, not just the creation of these beautiful confections to attract the wealthy, highly educated cultural tourists of the world, but maybe the ability to send the message to someone who lives two miles away, “This is for you.” Venturi, Scott Brown and Associates, Sainsbury Wing, National Gallery, London, U.K., 1991 (Matt Wargo) Sainsbury Wing renovation by Selldorf Architects (Edmund Sumner/©The National Gallery London) AN: How did you go about selecting the architects you wanted to interview? AS: You consign yourself to a lifetime of apologies to people who you didn’t interview. I wanted to be global, so I didn’t stack my book with New York–based architects. I wanted to attempt a gender balance, which was difficult. Again, I think our books work well together, Julian, because you spoke with a lot of people on my dream list. JR: I agree that our books are a good pair; it was fun for me to read your book when mine was in progress. I was first educated as an architect, but I’m also coming at this as a historian, so the idea was trying to figure out how we got here: How did museums become so important? I think that the success of both the museum and contemporary art in general is a bit of a surprise to everyone. In this century, we’ve seen so many traditional “highbrow” forms of culture get pushed to the periphery, but museums are thriving. I thought about Building Culture as an oral history project. I almost did the opposite of András: I have a couple younger voices, but I wanted to speak with established figures because that generation has shaped the present and has ideas about the future, too. Frank Gehry was one of the first people I interviewed; he’s 96 and he still has important museums under construction. It was interesting to ask Renzo Piano what he thinks is next. People like Frank and Renzo have had plenty of media exposure, but I did feel like there was a certain depth missing from journalistic coverage. I wanted to do a relatively small number of longer conversations and cover the widest historical range I could. I was thrilled to have Denise Scott Brown in there, because the Sainsbury Wing [of the National Gallery, London] alone is a paradigm-shifting project. She’s part of a whole generation that had a huge impact through postmodern museum designs, although most of [her peers] are no longer with us. That felt important to capture. Gehry Partners, Guggenheim Museum Bilbao, Spain, 1997 (Courtesy Gehry Partners, LLP/© Frank O. Gehry) AS: We’re discussing the success of contemporary visual art, which for most people feels inscrutable and hard to access. You had an interesting thought experiment: What would the same art have done without the scaffolding of the museum around it? The art museum could have become a dusty, irrelevant thing—and often still is—but through the efforts of a new generation of museum experts, working together with architects, communicators, and other specialists, this form has been lifted up and made super contemporary through, frankly, a lot of the functions that were seen as somewhat secondary. This is where the rubber meets the road for architects: So many of the metrics, even the audience metrics, are related to the non-gallery functions of the museum. People flock to the museum as a place, and this is where architecture, landscape architecture, and interior design have been superb. Museums have become civic hubs, which was maybe a secondary concern initially. That’s why people like Piano and Gehry are interesting, because they came up having to work in both worlds. They created this highly successful institutional typology, which still has those art at its core, but it’s the civic infrastructure that is the most successful. JR: Museums have always had a civic function, but almost as a secondary part of the program. With an institution like the Centre Pompidou in Paris the civic aspect starts to dominate. Meanwhile, all of these other institutions that used to provide shared social space have largely disappeared, which has an isolating and alienating effect on culture. It’s funny: Civic engagement started out as almost an afterthought, but it has become a crucial function of the museum in the 21st century. AS: Another point to make about generations: Do not confuse age with being namby-pamby or conservative. Today’s older architects are people of the 1960s, absolutely. Many, like Elizabeth Diller and David Chipperfield, were more radical then than some of our younger architects are today. They did not necessarily expect to be multimillionaires. They were devoted to the public sphere. These “older” figures who now get giant commissions are, on a DNA level, super radical people. JR: Richard Gluckman is another important example. Like Chipperfield, he has a direct connection to modernism through his education. We can talk all day about modernism as a failed project, but the fact is that back when people like Richard and David were in school, architecture was still seen as a fundamental part of the progressive state. Gluckman went to school at Syracuse University in the late 1960s, and as a student he worked for his professors exclusively on projects like housing and university campuses. But by the time he got around to opening his own office, it was 1977. New York had almost gone bankrupt—no one was building that stuff anymore. Gluckman got involved in designing spaces for art, and this was his way of basically sneaking back into the public sphere. I think their generation was connected to a very different—and very powerful—understanding of what architecture meant for society, and you still see that in their work today. AS: We can think about the art museum as a scaffolding building around a core enterprise of artistic experience. But this means something different for collecting versus non-collecting institutions. Often, you find institutions places that are dedicating more and more of their space to social functions around the art, contemplative aspects of art, and so on. The best architects are absolutely capable of doing both things: One is creating transparency, porosity, ease of access, and landscape integration in a way that flows, and the other is delivering wonderful amenities like shops and cafes. We can question some old dichotomies: How hard do you have to separate gallery space and social space? How porous could those boundaries be? What everybody profoundly believes is that a successful museum experience must have a magic combination of three things: objects, humans, and architecture. And when those three things come together—incredible real objects with a social experience in the company of other people in a magisterial architectural space—that creates an enduring magic that you cannnot sacrifice. Shohei Shigematsu/OMA, rendering of the New Museum of Contemporary Art expansion, New York, New York, anticipated completion 2025 (Courtesy OMA/bloomimages.de) JR: It was interesting for me to think about how conservative the museum can be. My conversation with Shohei Shigematsu at OMA put that into relief for me. He was one of the lead architects for the Whitney Museum extension proposal. At the time, OMA’s whole thing was reinventing typologies for the 21st century—think CCTV twisting the skyscraper, or Seattle transforming the public library. They took that aggressive critical method to the museum too—in the 1990s for MoMA and the Tate Modern, and then to the Whitney in 2001, and didn’t win a single competition. The establishment was not interested! AS: I agree that architects are often more radical than their clients. Hopefully nobody misunderstands this, but there is often a profound disconnect between the veneration of rule-breaking, iconoclastic innovation in the gallery versus the conservatism of the museum organization. Organizationally speaking, most museums have not read an airport book on modern management. I see architects trying to push against that. An easy example: Why do these buildings still look like fortresses? Libraries have been redesigned to work for people while still accommodating books. All too often, art museums still feel like citadels with lots of walls. Why? Because walls are great for hanging art on the inside of the building. Is that really the singular goal? AN: How does the scale of the institution shape what it can do? AS: We have certainly seen the emergence of a lot of small institutes and institutions, because of the enormous expansion of private museums. I do think small scale is good. When you ask most people about their favorite museums, they will frequently mention places that are quite intimate, like the Fondation Beyeler, in Basel, by Renzo Piano, soon with a lovely modest expansion by Zumthor. Nobody likes a super tanker, which is easy to respect but hard to love. When it comes to big, we need to differentiate between the gigantic temple on the hill versus what I think could be the future: the SESC Pompéia model, an interdisciplinary, social-cultural hub that may be quite big in the aggregate, and where the visual arts play a role inside a larger matrix. Particularly in our big, sprawling cities, such multipurpose, campus-like configurations could be an ideal setting for a museum. JR: I agree that the future might be more like the biennale model: When done well, the whole city is activated. In that sense maybe the size of the institution itself is less important. But I worry that smaller institutions will be hurt as public funding dries up and all museums become increasingly reliant on philanthropy. The regional, kunsthalle-like spots will suffer because those aren’t glamorous places to give money, but those are often the locations the programming makes the biggest impact in the community. AN: What else should we discuss? AS: Globalization is worth mentioning. There is a parallel to be drawn, perhaps, to the evolution of art. At the end of the 20th century, an astonishing amount of liberation became available to artists as the master narrative of modernism splintered to a more pluralistic discourse where all kinds of positions were accepted as art. Today I think something similar has happened in museum architecture: With the proliferation of museums globally, the language of museum architecture has opened up into a new openness to difference and variation, often informed by regional, vernacular forms and needs. Museums can be built using local materials or respond to local typologies, versus the older ideas of the white cube or the enfilade gallery sequence. Anything can be a museum—not just because of reuse, which is important, but because architects can build some crazy stuff inside almost any kind of building: a power station, a prison, a hospital, an army barracks. And people will say, “That’s a museum.” JR: There’s a running joke in museum design that the Louvre is an adaptive reuse project. And it’s true: The world’s first public art museum started out as a palace. This speaks to the museum’s typological flexibility. Its program is very architectural in the sense that it’s about how people and artworks interact in space, but it’s not like an airport or a hospital with a hyper-specialized program that is understandably difficult to fit into an existing structure. I’m optimistic that museums will stay on the cutting edge of adaptive reuse even as it gets more and more important for the whole architectural profession. Another thing that came out of my book is how much museum architects pay attention to the spaces artists are working in. The New York loft is the classic example. Once upon a time, not every gallery looked like a renovated postindustrial space, but artists moved into defunct industrial spaces decades ago and eventually exhibition spaces followed. This exchange goes both ways—its dialectical. As museum buildings have gotten more varied, artists have had a lot of fun learning how to use these new spaces. The Guggenheim in New York is an example. For decades, [Frank Lloyd] Wright’s design has been criticized because it’s hard to show most traditional art forms on the spiral ramps. But the best things I’ve seen in that museum in the past ten years have been installations in the atrium. Artists can do something wild with that space. After seeing that, do you really want to look at a little painting on a curvy wall? Julian Rose is a designer, critic, and historian. He is currently completing a PhD at Princeton on the origin and evolution of museums of contemporary art. András Szántó advises museums, foundations, educational institutions, and corporations on cultural strategy and program development, worldwide. This post contains affiliate links. AN may have a commission if you make a purchase through these links.

May 21, 2025Ravie LakshmananCyber Espionage / Vulnerability Russian cyber threat actors have been attributed to a state-sponsored campaign targeting Western logistics entities and technology companies since 2022. The activity has been assessed to be orchestrated by APT28 (aka BlueDelta, Fancy Bear, or Forest Blizzard), which is linked to the Russian General Staff Main Intelligence Directorate (GRU) 85th Main Special Service Center, Military Unit 26165. Targets of the campaign include companies involved in the coordination, transport, and delivery of foreign assistance to Ukraine, according to a joint advisory released by agencies from Australia, Canada, Czechia, Denmark, Estonia, France, Germany, the Netherlands, Poland, the United Kingdom, and the United States. "This cyber espionage-oriented campaign targeting logistics entities and technology companies uses a mix of previously disclosed TTPs and is likely connected to these actors' wide scale targeting of IP cameras in Ukraine and bordering NATO nations," the bulletin said. The alert comes weeks after France's foreign ministry accused APT28 of mounting cyber attacks on a dozen entities including ministries, defense firms, research entities, and think tanks since 2021 in an attempt to destabilize the nation. Then last week, ESET took the wraps off a campaign dubbed Operation RoundPress that it said has been ongoing since 2023 by exploiting cross-site scripting (XSS) vulnerabilities in various webmail services like Roundcube, Horde, MDaemon, and Zimbra to single out governmental entities and defense companies in Eastern Europe, as well as governments in Africa, Europe, and South America. According to the latest advisory, cyber attacks orchestrated by APT28 are said to have involved a combination of password spraying, spear-phishing, and modifying Microsoft Exchange mailbox permissions for espionage purposes. The primary targets of the campaign include organizations within NATO member states and Ukraine spanning defense, transportation, maritime, air traffic management, and IT services verticals. No less than dozens of entities in Bulgaria, Czechia, France, Germany, Greece, Italy, Moldova, the Netherlands, Poland, Romania, Slovakia, Ukraine, and the United States are estimated to have been targeted. Initial access to targeted networks is said to have been facilitated by leveraging seven different methods - Brute-force attacks to guess credentials Spear-phishing attacks to harvest credentials using fake login pages impersonating government agencies and Western cloud email providers that were hosted on free third-party services or compromised SOHO devices Spear-phishing attacks to deliver malware Exploitation of Outlook NTLM vulnerability (CVE-2023-23397) Exploitation of Roundcube vulnerabilities (CVE-2020-12641, CVE-2020-35730, CVE-2021-44026) Exploitation of internet-facing infrastructure such as corporate VPNs using public vulnerabilities and SQL injection Exploitation of WinRAR vulnerability (CVE-2023-38831) Once the Unit 26165 actors gain foothold using one of the above methods, the attacks proceed to the post-exploitation phase, which involves conducting reconnaissance to identify additional targets in key positions, individuals responsible for coordinating transport, and other companies cooperating with the victim entity. The attackers have also been observed using tools like Impacket, PsExec, and Remote Desktop Protocol (RDP) for lateral movement, as well as Certipy and ADExplorer.exe to exfiltrate information from the Active Directory. "The actors would take steps to locate and exfiltrate lists of Office 365 users and set up sustained email collection," the agencies pointed out. "The actors used manipulation of mailbox permissions to establish sustained email collection at compromised logistics entities." Another notable trait of the intrusions is the use of malware families like HeadLace and MASEPIE, to establish persistence on compromised hosts and harvest sensitive information. There is no evidence that malware variants like OCEANMAP and STEELHOOK have been used to directly target logistics or IT sectors. During data exfiltration, the threat actors have relied on different methods based on the victim environment, often utilizing PowerShell commands to create ZIP archives to upload the collected data to their own infrastructure, or employing Exchange Web Services (EWS) and Internet Message Access Protocol (IMAP) to siphon information from email servers. "As Russian military forces failed to meet their military objectives and Western countries provided aid to support Ukraine's territorial defense, Unit 26165 expanded its targeting of logistics entities and technology companies involved in the delivery of aid," the agencies said. "These actors have also targeted internet-connected cameras at Ukrainian border crossings to monitor and track aid shipments." The disclosure comes as Cato Networks revealed that suspected Russian threat actors are leveraging Tigris Object Storage, Oracle Cloud Infrastructure (OCI) Object Storage, and Scaleway Object Storage to host fake reCAPTCHA pages that make use of ClickFix-style lures to trick users into downloading Lumma Stealer. "The recent campaign leveraging Tigris Object Storage, OCI Object Storage, and Scaleway Object Storage builds upon earlier methods, introducing new delivery mechanisms aimed at evading detection and targeting technically proficient users," researchers Guile Domingo, Guy Waizel, and Tomer Agayev said. Found this article interesting? Follow us on Twitter  and LinkedIn to read more exclusive content we post. SHARE    

Spotify was one of the first companies to take advantage of recent changes to the App Store Guidelines. The company swiftly submitted a new version of its app with links that make it easier for users to subscribe to Spotify Premium using external payment methods. In an amicus brief filed in support of Epic Games this week, Spotify touts that it has already seen a measurable increase in Premium subscription conversions thanks to these changes. Spotify says: “Apple claims that only it is harmed by the 2025 Order without any corresponding benefit to the public. This is wrong. By forcing Apple’s long-overdue compliance with the Injunction, the 2025 Order has created substantial benefits for developers and—more importantly—consumers. Spotify’s internal data show that its iOS app updates have already resulted in a significant increase in iOS users upgrading to a Premium subscription. Apple’s compliance has also enabled new product innovations that would not have been possible without the Injunction.” The company adds that it is confident this increase is due to the changes to the iOS app: “In the two weeks since Spotify updated its iOS app in accordance with the 2025 Order, the rate of conversions from the Free- to Premium-tier service has remained relatively constant on Android, while conversion among iOS users has increased substantially. This strongly suggests that the increase is due to Apple finally complying with the Injunction thanks to the 2025 Order.” Apple was forced to change its App Store Guidelines after an injunction was handed down in its legal battle with Epic Games. Judge Yvonne Gonzalez Rogers barred Apple from charging commission or otherwise interfering with developers’ ability to direct customers to payment methods outside of the App Store. The latest version of the Spotify app for iOS shows pricing details and promotional information with a link that lets users subscribe via Safari. It’s also now easier to purchase audiobooks in the app. Apple has requested a pause of the initial injunction. It is also formally appealing that decision. My favorite iPhone accessories: Follow Chance: Threads, Bluesky, Instagram, and Mastodon.  Main Document Brief – 011926- 00004 – Ninth CircuitDownload Add 9to5Mac to your Google News feed.  FTC: We use income earning auto affiliate links. More.You’re reading 9to5Mac — experts who break news about Apple and its surrounding ecosystem, day after day. Be sure to check out our homepage for all the latest news, and follow 9to5Mac on Twitter, Facebook, and LinkedIn to stay in the loop. Don’t know where to start? Check out our exclusive stories, reviews, how-tos, and subscribe to our YouTube channel

The Jet Li Collection (4K Blu-ray) $107 (was $130) | Releases July 29 Preorder at Amazon Preorder at Walmart If there was a Mount Rushmore of action stars, you can bet that Jet Li's face would be on it. A legend of Hong Kong's golden age of filmmaking and the star of several cult-classic Hollywood movies, Jet Li's movies are as enjoyable today as they were over 30 years ago. Five of the best Jet Li movies are being collected in a new 4K Blu-ray box set. Preorders for The Jet Li Collection opened in April, and now Amazon and Walmart are offering a nice discount that drops the price to $107 (was $130).If you already preordered a copy at full price, the new discount will automatically be applied to your order when Amazon or Walmart charges your payment method. Both retailers wait until preorders ship to charge, so if the price drops even lower before The Jet Li Collection's July 29 release, you'll be eligible for additional discounts thanks to the preorder price guarantee. The Jet Li Collection (4K Blu-ray) $107 (was $130) | Releases July 29 This collection is part of Shout Factory's new Hong Kong Cinema Classics line, and inside, you'll find the following movies on 4K Blu-ray and 1080p Blu-ray.Fist of LegendTai Chi MasterFong Sai Yuk AKA The LegendFong Sai Yuk 2 AKA The Legend 2The Bodyguard from BeijingEach movie is a new 4K restoration from the original film negatives, and they all come with brand-new extras. Not every movie here offers a massive number of bonus materials, but there's still a lot to dive into overall. You can check out all of the special features at the end of this story. Preorder at Amazon Preorder at Walmart These movies are essentially a highlight reel of Li's career during the '90s, with three of them being filmed in 1993 alone. Earlier films like Once Upon a Time in China and Swordsman 2 helped turn him into a rising star thanks to his talent and well-choreographed fight scenes, and the Fong Sai Yuk movies--known as The Legend and The Legend 2 in the US--alongside Tai Chi Master showcased more of his graceful ability to smash faces. Fast-forward to 1994, and Li's work on The Bodyguard from Beijing and Fist of Legend were two more instant classics that garnered international praise.Continue Reading at GameSpot

Posted on : May 21, 2025 By Tech World Times Business  Rate this post In this era where there is technological business evolution, it is very important to have some sort of edge over the counterparts. Regardless of whether operating a large-scale operation or a small-line manufacturing operation, your company requires operational efficiency in the manufacturing department. Attaining good performance and profitability in operational business entities is important for success. It can assist your manufacturing company in maximizing performance coordination and eliminating wastage. Keeping this scenario under consideration, we are presenting to you 5 strategies to maximize FMCG production efficiency. Applying Lean Manufacturing Principles   Lean manufacturing is a technique of waste elimination that is practiced within the manufacturing system. This is done to decrease waste amount so that efficiency is maximized. It enables the incorporation of statistical methods in removing articles at each step of the manufacturing process. Here are some ways to apply lean manufacturing principles: ·       Value stream mapping ·       In time production Investing in The Latest Cutting Edge Technologies Quick investment in advanced technologies creates new opportunities for enhancing operational efficiency in manufacturing. Here are some methods of relying on the techniques of advanced technologies: ·       Production process automation ·       Execution of advanced technologies  in manufacturing ·       Investing in the Internet of Things (IoT)  Methods of Supply Chain Management and Optimization Effective optimization and management of the supply chain play a vital role in operational manufacturing. This is because an optimized and well-managed supply chain guarantees shorter timelines for manufacturing. It reduces labor costs and generates enhanced product quality and much more. Some of the techniques to optimize supply chain management are: ·       Collaborating with the suppliers ·       Management of inventory and logistics Promotion of Workforce Engagement and Training The utilization of funds for the development and training of the workforce to guarantee that they attain an appropriate level of training to meet the needed industrial skills set, allows manufacturers to obtain a capable workforce that can raise the speed of production to attain operational efficiency in manufacturing.   This can be achieved through the adoption of different techniques: ·       Training sessions ·       Engagement ·       Applying Health and Safety measures Analysis and Monitoring of the Performance Metrics    The analysis and monitoring of the performance metrics daily is very important to pinpoint the important areas for improvement. This is the point where you can analyze and monitor the performance flawlessly. It can be attained through the implementation of the following techniques: ·       Effectiveness of the equipment ·       Cycle time of the production process ·       Decreasing the downtime period  Conclusion Maximizing an FMCG (Fast-Moving Consumer Goods) manufacturing company entails an in-depth evaluation of different processes, systems, and strategies to guarantee optimal efficiency, productivity, and profitability. Regardless of whether operating a large-scale operation or a small-line manufacturing operation, your company requires operational efficiency in the manufacturing department. Attaining good performance and profitability in operational business entities is important for success. It can assist your manufacturing company in maximizing performance coordination and eliminating wastage. After viewing the discussion above, it can be said that these five strategies can change your manufacturing ecosystem. This will boost profitability, efficiency, and productivity. Tech World TimesTech World Times (TWT), a global collective focusing on the latest tech news and trends in blockchain, Fintech, Development & Testing, AI and Startups. If you are looking for the guest post then contact at techworldtimes@gmail.com

This is the second in a multi-part series on creating web applications with generative AI integration. Part 1 focused on explaining the AI stack and why the application layer is the best place in the stack to be. Check it out here. Table of Contents Introduction I thought spas were supposed to be relaxing? Microservices are for Macrocompanies Ruby and Python: Two Sides of the Same Coin Recent AI Based Gems Summary Introduction It’s not often that you hear the Ruby language mentioned when discussing AI. Python, of course, is the king in this world, and for good reason. The community has coalesced around the language. Most model training is done in PyTorch or TensorFlow these days. Scikit-learn and Keras are also very popular. RAG frameworks such as LangChain and LlamaIndex cater primarily to Python. However, when it comes to building web applications with AI integration, I believe Ruby is the better language. As the co-founder of an agency dedicated to building MVPs with generative AI integration, I frequently hear potential clients complaining about two things: Applications take too long to build Developers are quoting insane prices to build custom web apps These complaints have a common source: complexity. Modern web apps have a lot more complexity in them than in the good ol’ days. But why is this? Are the benefits brought by complexity worth the cost? I thought spas were supposed to be relaxing? One big piece of the puzzle is the recent rise of single-page applications (SPAs). The most popular stack used today in building modern SPAs is MERN (MongoDB, Express.js, React.js, Node.js). The stack is popular for a few reasons: It is a JavaScript-only stack, across both front-end and back-end. Having to only code in only one language is pretty nice! SPAs can offer dynamic designs and a “smooth” user experience. Smooth here means that when some piece of data changes, only a part of the site is updated, as opposed to having to reload the whole page. Of course, if you don’t have a modern smartphone, SPAs won’t feel so smooth, as they tend to be pretty heavy. All that JavaScript starts to drag down the performance. There is a large ecosystem of libraries and developers with experience in this stack. This is pretty circular logic: is the stack popular because of the ecosystem, or is there an ecosystem because of the popularity? Either way, this point stands.React was created by Meta. Lots of money and effort has been thrown at the library, helping to polish and promote the product. Unfortunately, there are some downsides of working in the MERN stack, the most critical being the sheer complexity. Traditional web development was done using the Model-View-Controller (MVC) paradigm. In MVC, all of the logic managing a user’s session is handled in the backend, on the server. Something like fetching a user’s data was done via function calls and SQL statements in the backend. The backend then serves fully built HTML and CSS to the browser, which just has to display it. Hence the name “server”. In a SPA, this logic is handled on the user’s browser, in the frontend. SPAs have to handle UI state, application state, and sometimes even server state all in the browser. API calls have to be made to the backend to fetch user data. There is still quite a bit of logic on the backend, mainly exposing data and functionality through APIs. To illustrate the difference, let me use the analogy of a commercial kitchen. The customer will be the frontend and the kitchen will be the backend. MVCs vs. SPAs. Image generated by ChatGPT. Traditional MVC apps are like dining at a full-service restaurant. Yes, there is a lot of complexity (and yelling, if The Bear is to be believed) in the backend. But the frontend experience is simple and satisfying: all the customer has to do is pick up a fork and eat their food. SPAs are like eating at a buffet-style dining restaurant. There is still quite a bit of complexity in the kitchen. But now the customer also has to decide what food to grab, how to combine them, how to arrange them on the plate, where to put the plate when finished, etc. Andrej Karpathy had a tweet recently discussing his frustration with attempting to build web apps in 2025. It can be overwhelming for those new to the space. The reality of building web apps in 2025 is that it's a bit like assembling IKEA furniture. There's no "full-stack" product with batteries included, you have to piece together and configure many individual services:– frontend / backend (e.g. React, Next.js, APIs)– hosting…— Andrej Karpathy (@karpathy) March 27, 2025 In order to build MVPs with AI integration rapidly, our agency has decided to forgo the SPA and instead go with the traditional MVC approach. In particular, we have found Ruby on Rails (often denoted as Rails) to be the framework best suited to quickly developing and deploying quality apps with AI integration. Ruby on Rails was developed by David Heinemeier Hansson in 2004 and has long been known as a great web framework, but I would argue it has recently made leaps in its ability to incorporate AI into apps, as we will see. Django is the most popular Python web framework, and also has a more traditional pattern of development. Unfortunately, in our testing we found Django was simply not as full-featured or “batteries included” as Rails is. As a simple example, Django has no built-in background job system. Nearly all of our apps incorporate background jobs, so to not include this was disappointing. We also prefer how Rails emphasizes simplicity, with Rails 8 encouraging developers to easily self-host their apps instead of going through a provider like Heroku. They also recently released a stack of tools meant to replace external services like Redis. “But what about the smooth user experience?” you might ask. The truth is that modern Rails includes several ways of crafting SPA-like experiences without all of the heavy JavaScript. The primary tool is Hotwire, which bundles tools like Turbo and Stimulus. Turbo lets you dynamically change pieces of HTML on your webpage without writing custom JavaScript. For the times where you do need to include custom JavaScript, Stimulus is a minimal JavaScript framework that lets you do just that. Even if you want to use React, you can do so with the react-rails gem. So you can have your cake, and eat it too! SPAs are not the only reason for the increase in complexity, however. Another has to do with the advent of the microservices architecture. Microservices are for Macrocompanies Once again, we find ourselves comparing the simple past with the complexity of today. In the past, software was primarily developed as monoliths. A monolithic application means that all the different parts of your app — such as the user interface, business logic, and data handling — are developed, tested, and deployed as one single unit. The code is all typically housed in a single repo. Working with a monolith is simple and satisfying. Running a development setup for testing purposes is easy. You are working with a single database schema containing all of your tables, making queries and joins straightforward. Deployment is simple, since you just have one container to look at and modify. However, once your company scales to the size of a Google or Amazon, real problems begin to emerge. With hundreds or thousands of developers contributing simultaneously to a single codebase, coordinating changes and managing merge conflicts becomes increasingly difficult. Deployments also become more complex and risky, since even minor changes can blow up the entire application! To manage these issues, large companies began to coalesce around the microservices architecture. This is a style of programming where you design your codebase as a set of small, autonomous services. Each service owns its own codebase, data storage, and deployment pipelines. As a simple example, instead of stuffing all of your logic regarding an OpenAI client into your main app, you can move that logic into its own service. To call that service, you would then typically make REST calls, as opposed to function calls. This ups the complexity, but resolves the merge conflict and deployment issues, since each team in the organization gets to work on their own island of code. Another benefit to using microservices is that they allow for a polyglot tech stack. This means that each team can code up their service using whatever language they prefer. If one team prefers JavaScript while another likes Python, this is no issue. When we first began our agency, this idea of a polyglot stack pushed us to use a microservices architecture. Not because we had a large team, but because we each wanted to use the “best” language for each functionality. This meant: Using Ruby on Rails for web development. It’s been battle-tested in this area for decades. Using Python for the AI integration, perhaps deployed with something like FastAPI. Serious AI work requires Python, I was led to believe. Two different languages, each focused on its area of specialty. What could go wrong? Unfortunately, we found the process of development frustrating. Just setting up our dev environment was time-consuming. Having to wrangle Docker compose files and manage inter-service communication made us wish we could go back to the beauty and simplicity of the monolith. Having to make a REST call and set up the appropriate routing in FastAPI instead of making a simple function call sucked. “Surely we can’t develop AI apps in pure Ruby,” I thought. And then I gave it a try. And I’m glad I did. I found the process of developing an MVP with AI integration in Ruby very satisfying. We were able to sprint where before we were jogging. I loved the emphasis on beauty, simplicity, and developer happiness in the Ruby community. And I found the state of the AI ecosystem in Ruby to be surprisingly mature and getting better every day. If you are a Python programmer and are scared off by learning a new language like I was, let me comfort you by discussing the similarities between the Ruby and Python languages. Ruby and Python: Two Sides of the Same Coin I consider Python and Ruby to be like cousins. Both languages incorporate: High-level Interpretation: This means they abstract away a lot of the complexity of low-level programming details, such as memory management. Dynamic Typing: Neither language requires you to specify if a variable is an int, float, string, etc. The types are checked at runtime. Object-Oriented Programming: Both languages are object-oriented. Both support classes, inheritance, polymorphism, etc. Ruby is more “pure”, in the sense that literally everything is an object, whereas in Python a few things (such as if and for statements) are not objects. Readable and Concise Syntax: Both are considered easy to learn. Either is great for a first-time learner. Wide Ecosystem of Packages: Packages to do all sorts of cool things are available in both languages. In Python they are called libraries, and in Ruby they are called gems. The primary difference between the two languages lies in their philosophy and design principles. Python’s core philosophy can be described as: There should be one — and preferably only one — obvious way to do something. In theory, this should emphasize simplicity, readability, and clarity. Ruby’s philosophy can be described as: There’s always more than one way to do something. Maximize developer happiness. This was a shock to me when I switched over from Python. Check out this simple example emphasizing this philosophical difference: # A fight over philosophy: iterating over an array # Pythonic way for i in range(1, 6): print(i) # Ruby way, option 1 (1..5).each do |i| puts i end # Ruby way, option 2 for i in 1..5 puts i end # Ruby way, option 3 5.times do |i| puts i + 1 end # Ruby way, option 4 (1..5).each { |i| puts i } Another difference between the two is syntax style. Python primarily uses indentation to denote code blocks, while Ruby uses do…end or {…} blocks. Most include indentation inside Ruby blocks, but this is entirely optional. Examples of these syntactic differences can be seen in the code shown above. There are a lot of other little differences to learn. For example, in Python string interpolation is done using f-strings: f"Hello, {name}!", while in Ruby they are done using hashtags: "Hello, #{name}!". Within a few months, I think any competent Python programmer can transfer their proficiency over to Ruby. Recent AI-based Gems Despite not being in the conversation when discussing AI, Ruby has had some recent advancements in the world of gems. I will highlight some of the most impressive recent releases that we have been using in our agency to build AI apps: RubyLLM (link) — Any GitHub repo that gets more than 2k stars within a few weeks of release deserves a mention, and RubyLLM is definitely worthy. I have used many clunky implementations of LLM providers from libraries like LangChain and LlamaIndex, so using RubyLLM was like a breath of fresh air. As a simple example, let’s take a look at a tutorial demonstrating multi-turn conversations: require 'ruby_llm' # Create a model and give it instructions chat = RubyLLM.chat chat.with_instructions "You are a friendly Ruby expert who loves to help beginners." # Multi-turn conversation chat.ask "Hi! What does attr_reader do in Ruby?" # => "Ruby creates a getter method for each symbol... # Stream responses in real time chat.ask "Could you give me a short example?" do |chunk| print chunk.content end # => "Sure! # ```ruby # class Person # attr... Simply amazing. Multi-turn conversations are handled automatically for you. Streaming is a breeze. Compare this to a similar implementation in LangChain: from langchain_openai import ChatOpenAI from langchain_core.schema import SystemMessage, HumanMessage, AIMessage from langchain_core.callbacks.streaming_stdout import StreamingStdOutCallbackHandler SYSTEM_PROMPT = "You are a friendly Ruby expert who loves to help beginners." chat = ChatOpenAI(streaming=True, callbacks=[StreamingStdOutCallbackHandler()]) history = [SystemMessage(content=SYSTEM_PROMPT)] def ask(user_text: str) -> None: """Stream the answer token-by-token and keep the context in memory.""" history.append(HumanMessage(content=user_text)) # .stream yields message chunks as they arrive for chunk in chat.stream(history): print(chunk.content, end="", flush=True) print() # newline after the answer # the final chunk has the full message content history.append(AIMessage(content=chunk.content)) ask("Hi! What does attr_reader do in Ruby?") ask("Great - could you show a short example with attr_accessor?") Yikes. And it’s important to note that this is a grug implementation. Want to know how LangChain really expects you to manage memory? Check out these links, but grab a bucket first; you may get sick. Neighbors (link) — This is an excellent library to use for nearest-neighbors search in a Rails application. Very useful in a RAG setup. It integrates with Postgres, SQLite, MySQL, MariaDB, and more. It was written by Andrew Kane, the same guy who wrote the pgvector extension that allows Postgres to behave as a vector database. Async (link) — This gem had its first official release back in December 2024, and it has been making waves in the Ruby community. Async is a fiber-based framework for Ruby that runs non-blocking I/O tasks concurrently while letting you write simple, sequential code. Fibers are like mini-threads that each have their own mini call stack. While not strictly a gem for AI, it has helped us create features like web scrapers that run blazingly fast across thousands of pages. We have also used it to handle streaming of chunks from LLMs. Torch.rb (link) — If you are interested in training deep learning models, then surely you have heard of PyTorch. Well, PyTorch is built on LibTorch, which essentially has a lot of C/C++ code under the hood to perform ML operations quickly. Andrew Kane took LibTorch and made a Ruby adapter over it to create Torch.rb, essentially a Ruby version of PyTorch. Andrew Kane has been a hero in the Ruby AI world, authoring dozens of ML gems for Ruby. Summary In short: building a web application with AI integration quickly and cheaply requires a monolithic architecture. A monolith demands a monolingual application, which is necessary if your end goal is quality apps delivered with speed. Your main options are either Python or Ruby. If you go with Python, you will probably use Django for your web framework. If you go with Ruby, you will be using Ruby on Rails. At our agency, we found Django’s lack of features disappointing. Rails has impressed us with its feature set and emphasis on simplicity. We were thrilled to find almost no issues on the AI side. Of course, there are times where you will not want to use Ruby. If you are conducting research in AI or training machine learning models from scratch, then you will likely want to stick with Python. Research almost never involves building Web Applications. At most you’ll build a simple interface or dashboard in a notebook, but nothing production-ready. You’ll likely want the latest PyTorch updates to ensure your training runs quickly. You may even dive into low-level C/C++ programming to squeeze as much performance as you can out of your hardware. Maybe you’ll even try your hand at Mojo. But if your goal is to integrate the latest LLMs — either open or closed source — into web applications, then we believe Ruby to be the far superior option. Give it a shot yourselves! In part three of this series, I will dive into a fun experiment: just how simple can we make a web application with AI integration? Stay tuned.  If you’d like a custom web application with generative AI integration, visit losangelesaiapps.com The post Building AI Applications in Ruby appeared first on Towards Data Science.

Thousands of people have preregistered and now it’s finally here: Our biggest e-book yet, User interface design and implementation in Unity, is available to download. Get ready to dive into over 130 pages of advanced instruction in UI design.Your game’s user interface is perhaps the most direct way you can communicate with and guide your players – like a folded map you hand to them that reveals clues, key details, and directions as they progress. Whether you’re using more traditional elements like health bars and pop-up messages, or elements completely embedded in the game world, such as showing stats on the back of a player’s survival suit, the UI is integral to immersing players in your game’s story, realm, and artistic style.We’re thrilled to announce that our latest technical e-book, User interface design and implementation in Unity, is available to download for free. Thousands of people have already signed up for it, and just as many have downloaded its companion piece, the demo project, UI Toolkit sample – Dragon Crashers, to date. Now it’s your turn.The interest in this e-book is understandable. As it says in the introduction, “User interface is a critical part of any game… a solid graphical user interface (GUI) is an extension of a game’s visual identity… [and] modern audiences crave refined, intuitive GUIs that seamlessly integrate with your application.”The guide begins by covering UI design and art creation fundamentals, and then moves on to in-depth instructional sections on UI development in Unity. Written and reviewed by technical and UI artists – external and Unity professionals alike – the e-book unpacks both Unity UI, the default solution, and the newer UI Toolkit.The emphasis, however, is on the latter toolset, as UI Toolkit now provides many benefits for projects with complex, fullscreen interfaces. Think of projects that require a scalable and performant system for runtime UI. To help you choose the right solution for your project, please refer to this section of the Unity manual.The e-book is a treasure trove of information for professional UI designers, artists, and other Unity creators who want to deepen their knowledge of UI development. Here’s a snapshot of what’s inside.The first section aims to inspire with foundational tips for making effective UI. It looks at examples of diegetic UIs, where UI elements can be found right in the story, making parts of the game world function as a user interface. It explains how elements can either contribute to or break the immersion that a player experiences. We even turned this section into a blog post that you can read here.The guide then turns to the roles and responsibilities of a UI designer, and what tools and methods they employ such as UI wireframing, art creation through mockups, fonts, and grey-boxing. There’s also a chapter on asset preparation and exporting graphics from Digital Content Creation (DCC) tools. These earlier sections in the guide are helpful no matter what game engine and UI solution you’re using.An extensive chapter is devoted to Unity UI. Unity UI is our longtime system for creating in-game UIs, and currently the go-to solution for positioning UI in a 3D world or using GameObject-based Unity systems.This section outlines Unity UI fundamentals for prototyping and integrating assets in-Editor: the Canvas, prebuilt UI elements, TextMesh Pro, and Prefabs, among others. We recently updated an article on advanced optimization techniques for Unity UI, where you can find tips on related topics.UI Toolkit is made for maximum performance and reusability with workflows and authoring tools informed by standard web technologies. UI designers and artists will likely find it familiar, especially with prior experience designing web pages.Three major sections of the guide highlight instructions for developing runtime UI with UI Toolkit. There’s a thorough explanation of the parts that comprise UIs made with UI Toolkit, including the Unity Extensible Markup Language (UXML) and Unity Style Sheet (USS) using UI Builder.You’ll explore how UI Toolkit positions visual elements based on Yoga, an HTML/CSS Layout engine that implements a subset of Flexbox. Flexbox architecture provides advantages, such as responsive UI, enabling you to adapt your UI to different screen resolutions and sizes. Through both UXML and USS, you can decouple the styles applied to UI layouts (and switch those styles up as needed), while logic and functionality continue to live in code. Workflows for visual elements, the fundamental building blocks of each interface, are also discussed in great detail – from positioning, size, and alignment settings, to margins and padding.The chapter on styling shows you how to define reusable styling for visual elements with Selectors, override styles and define unique attributes with inline styles, and create animations and effects with USS animation and a Camera Render Texture. It also demonstrates how you can thematize UI elements for holidays and other special events.That’s when the e-book gets into UI Toolkit sample – Dragon Crashers, with different sections that depict how the UI was made; from the menus and custom controls like radial counters or tabbed views, to embedded UXML templates and more.Finally, the guide concludes with a mini profile of the studio Mechanistry’s UI migration to UI Toolkit for their new game, Timberborn. This brief study showcases how their lean team managed to scale and keep their game consistent across various menus and screens.At 137 pages, the UI e-book is not a light read. As with the other technical e-books released this past year, use it as a reference on an ongoing basis.Along with the e-book, check out a couple of recently released resources filled with useful tips for leveraging Unity UI and UI Toolkit:The Unite 2022 session, Extending the Unity Editor with custom tools using UI Toolkit, shows programmers how to use UI Builder to create a custom Inspector for real-time Play Mode debug data visualization.The webinar, Best practices for mobile UI design, gathers experts from Outfit7, Samsung, and Unity to share strategies for maximizing the creativity and flow of your mobile games.Bookmark one or both of these pages. They compile all of our technical e-books and advanced content:Unity best practicesAdvanced best practices – Unity ManualWe hope that you enjoy this latest e-book and look forward to your feedback in this forum.

Sandeep Kashyap, CEO, ProofHubMay 21, 20254 Min ReadEugene Sergeev via Alamy StockIt’s normal to feel nervous when you have to make big decisions at work. After all, you never know how things will turn out. Fortunately, decision-making frameworks can help lessen those nerves and boost your confidence. They bring structure and clarity by bringing practical, proven methods that turn chaos into clarity.  For IT leaders, these frameworks support critical thinking, confident action, and smarter choices -- even under pressure. Most importantly, they help you cut through the noise and ensure every decision stays aligned with your long-term business goals. This blog post will walk you through the five frameworks for effective decision-making that can help IT leaders make more informed decisions. Each one is designed to help you simplify complexity and lead with greater impact.  Importance of Decision-Making FrameworksDecision-making frameworks bring consistency and logic to the decision-making process. They help you break things down and focus on the essentials. Here are the benefits of using these frameworks. Make your objectives clear: Structured decision-making frameworks help you cut through the noise and focus on what matters most, ensuring every decision aligns with your objectives.  Bring teams together: The frameworks allow you to involve the right people and ensure everyone is on the same page. Related:Avoid costly mistakes: IT decisions often involve significant investments, such as new software and infrastructure upgrades. The framework helps you assess potential risk upfront and make deliberate choices.  5 Decision-Making Frameworks Every Leader Should KnowA decision-making framework provides clarity and consistency to make better decisions. Here are five frameworks that can sharpen your thinking and strengthen your leadership. 1. RAPID (recommend, agree, perform, input, decide) RAPID is a decision-making framework that helps clarify who is responsible for what when multiple stakeholders are involved.  Each letter in RAPID represents a key role in the decision-making process:  Recommend: The person in this role leads the effort by gathering data, analyzing options, and proposing a well-informed recommendation. Agree: These stakeholders have to work closely with the recommender to shape the best possible decision.  Perform: This is the individual or team responsible for executing the decision once it's made. Input: These contributors offer valuable insights, expertise, or context that inform the recommendation.  Decide: The final authority that makes the call and commits the organization to move forward. This role carries accountability for the outcome. Related:2. SPADE (setting, people, alternatives, decide, explain) The SPADE framework breaks down each step of the structured decision-making process so that you can reach an informed and critical conclusion. It’s especially helpful when decisions involve multiple teams, limited time, and high visibility.  Each letter in SPADE represents a crucial phase in the decision-making process: Setting: Define the decision’s scope, goal, and constraints.  People: Identify and engage relevant stakeholders such as decision-makers, influencers, and executors. Alternatives: Generate options related to the decision based on criteria like cost, security, and scalability.  Decide: Evaluate all options and select the best course of action. You can avoid negative consequences and bias through objective methods like private voting. Explain: Clearly document and explain the rationale behind a decision to ensure alignment across teams and maintain accountability for outcomes. 3. OODA loop (observe, orient, decide, act) The OODA loop is a four-step approach to decision-making that focuses on filtering available information, putting it in context, and quickly making the most appropriate decision.  Related:The word OODA stands for:  Observe: Monitor system performance, team dynamics, and industry trends to gather relevant and timely data. Orient: Analyze the information you have collected to understand the context, challenges, and opportunities. Decide: Based on your analysis, choose the most effective course of action. Act: Implement the decision quickly and efficiently. Once action is taken, the loop restarts—each decision and outcome creates new conditions to observe and evaluate. 4. Eisenhower MatrixThe Eisenhower Matrix is a task prioritization technique that helps make decisions related to tasks. It helps you organize tasks into four quadrants, based on the urgency and importance, and suggests appropriate action for tasks in each quadrant. It ensures that essential tasks are completed first, contributing to the success of projects and goals.  Here is what the Eisenhower matrix includes:  QuadrantDescription Action DoImportant and urgent Handle these immediately ScheduleImportant but not urgent Schedule these for later DelegateUrgent but not important Assign these to others if possible DeleteNeither urgent nor important Consider removing these altogether 5. Decision TreeA decision tree is a graphical representation that helps IT leaders map out the possible outcomes of different decisions. It helps leaders assess risks, rewards, and the potential consequences of each choice before committing to a path. Decision trees are most useful in complex decision-making processes where multiple scenarios are involved. ConclusionIT leaders deal with tough decisions every day. Which project should be prioritized? Should we adopt new tools or improve the existing ones? Who should get what tasks? To handle these challenges, leaders can use frameworks for effective decision-making like RAPID, SPADE, OODA, Eisenhower Matrix, and decision trees. These tools help bring structure and clarity to tough decisions, making it easier to move forward with confidence in a fast-changing business world. About the AuthorSandeep KashyapCEO, ProofHubSandeep Kashyap, the visionary CEO of ProofHub, boasts over 25 years of IT industry experience. He's a recognized luminary known for innovation and agility. His contributions extend to project management insights and leadership, growth and entrepreneurship. His practical expertise is evident in ProofHub's success. Recognized as Top Leadership Voice on Linkedin, Sandeep’s contributions provide invaluable insight for leaders and professionals seeking to create thriving workplaces.See more from Sandeep KashyapWebinarsMore WebinarsReportsMore ReportsNever Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.SIGN-UPYou May Also Like

Announced back at the Game Developers Conference (GDC) 2019, Havok Physics for Unity was initially distributed as an experimental package on the Unity Asset Store. Now, with the availability of ECS for Unity (Entity Component System) in the Unity 2022.2 Tech Stream, Havok Physics for Unity is officially supported for production. In fact, we’ve made this package available to all Unity Pro, Enterprise, and Industrial Collection subscribers for free.Havok Physics for Unity is built on the same foundation of technology that powers many of the world’s leading game franchises like Destiny and Assassin’s Creed, among others. When we first set out to define what the future of physics could look like with our Data-Oriented Technology Stack (DOTS), we sought a partner that shared the same core concepts and values as us. Through our partnership with Havok, we were able to leverage DOTS to deliver the highly optimized, stateless, entirely C#, and performant Unity Physics we know today.We also prepared for more complex simulation requirements for users who might need a stateful physics system. We knew that Havok would be the perfect solution to integrate into Unity for those high-end simulation needs.The Havok Physics for Unity package is written using the same C# ECS framework as Unity Physics, and is backed by the closed-source, proprietary Havok Physics engine, written in native C++. Havok Physics for Unity is heavily optimized for many typical gaming use cases. For example, core algorithms have been refined over many years with various automatic caching strategies (including the sleeping of inactive objects), meaning that CPU resources are spent only as needed.Since the experimental package, Unity and Havok have been working together with early users of the plug-in to drive improvements and add new features.Here’s a breakdown of what’s new:Havok Physics for Unity is now available for free for all Unity Pro, Enterprise, and Unity Industrial Collection subscribers.Havok Physics for Unity is based on the 2021.2 version of the original Havok SDK, and brings more stability and performance to the Unity plug-in.As part of the full release, we’ve incorporated support for motorized joints (motors), such as linear position, as well as linear, rotational, and angular velocity.We’ve also added new methods to the HavokSimulation API, which enables granular stepping for your simulations, plus methods for accessing Havok simulations more efficiently with singletons.Check out the complete changelog of updates to Havok Physics for Unity.Havok Physics is a robust physics engine designed to handle the performance demands of the most dynamic games, often involving complex scenes with lots of physical interaction. By working with partners across the industry for over 20 years, Havok has encountered, solved, and continued to iterate on some of the toughest problems facing real-time physics simulation. This investment has led to the stable stacking of physics bodies, minimal artifacts for fast-moving bodies, and generally more controlled behavior, especially when it comes to non-optimized collision geometry.Of course, physics means action, so let’s see how these two creators are currently taking action with Havok Physics for Unity.Title:Hostile Mars Studio:Big Rook Games Studio size:Individual Platforms: Windows PC, console Genre: Open-world, base-building automation tower defense Players: Single playerFirst presented at PAX – East in 2022, Hostile Mars drew in quite an audience with its dedicated use of physics and unique blend of genres. The vast Martian landscape is brought to life in an open-world factory base-building and automation game. Players confront each other at the ground level in close combat to defend their factories through both third-person shooting and programmable defenses.Jake Jameson, founder of Big Rook Games, began by blending genres across multiplayer and single-player games – adding elements from puzzlers along with first-person shooters (FPS) and wave-based shooters. The more he evolved this idea, the more he narrowed his focus on creating a single-player base builder. He soon discovered the ideal balance between building, strategy, and combat, finally leading up to the one and only Hostile Mars.When completing that final iteration of Hostile Mars, it was clear the game needed to use some sort of data-oriented programming model. Jake wanted to have large enemy waves while still achieving high-end visuals. In order to provide the players with the best possible experience, the game had to run performantly on a massive scale, supporting thousands of enemies simultaneously onscreen.To meet these demands, Jake turned to Unity’s Data-Oriented Technology Stack. In leveraging ECS for Unity, every enemy in Hostile Mars could run real-time Mesh Physics/Collisions, A* Pathfinding, and Local Avoidance, in addition to robust state systems, animations, weapon and projectile systems, high-quality VFX, particle systems, and more.“Without DOTS, I wouldn’t have been able to provide the experience that I imagined in my original design. It just wouldn’t have been possible without implementing my own ECS framework, and as a solo dev, this isn’t viable given my timeline and budget.” – Jake Jameson, founder of Big Rook GamesAlthough Jake is not a game developer by trade, he is an avid gamer and was already familiar with the original Havok Physics engine. Knowing how trusted Havok’s technology has been among AAA studios, Jake felt confident implementing it as soon as it became available in Unity via the experimental package.Hostile Mars is a physics-intensive experience. The player uses physics-based traps and turrets to manipulate the physics properties of enemies. By applying different physical properties to enemies, the goal is to stop them in their tracks or drive them toward more dangerous traps.Not to mention that Hostile Mars involves an incredible number of enemies. More specifically, there are up to 5,000 individual physics-based enemies that can flood the player’s Martian factories, which leads to hundreds of collisions and projectiles onscreen at once.All of the physics in Hostile Mars utilizes Havok Physics for Unity; from the collisions between projectiles and enemies, players and enemies, even enemies and other enemies, to enemies and the landscape itself, and hovering enemies who require a constant state of force to stay afloat. These distinct physics interactions take place in real-time, with physical simulation that allows for believable hovering, gravity, and mesh point collisions. So when players strike their enemies at a particular point, they will see them spin away just as Havok Physics for Unity intended them to.Not only are the enemies entirely physics-based, but the traps that are necessary for gameplay are physics-based too. There are gravity traps that push and pull enemies, which in turn, slow down or speed up traps that smash enemies. There are even traps that spring spikes up into an enemy’s path, not just pushing them aside, but realistically simulating the force and velocity of the spikes so that the enemy reaction appears authentic.Once the players advance, the enemy waves become varied and increasingly complex, which requires players to build traps more strategically, combining their different physics-based interactions to herd enemies with different weaknesses toward stronger traps and turrets that can impact them most. For example, while a frost trap might slow some enemies down, an explosive trap can deal the most damage possible against a highly concentrated wave.Plans for releasing Hostile Mars on consoles are still in the works, but in the meantime, you can add Hostile Mars for Windows PC to your Wishlist on Steam today.Title: Robocraft 2 Studio:Freejam Games Studio size:25 Platforms:Windows PC, console Genre: Online vehicular combat Players: 5v5 online multiplayerRobocraft 2 is the free-to-play sequel to 2017’s award-winning Robocraft, where players build customizable robot battle vehicles that drive, hover, walk, and fly in an open-world multiplayer environment. Since this initial success, the team at Freejam Games has refined a fully customizable experience for Robocraft 2, so that players can bring their own creations to competitive multiplayer gameplay.As Freejam Games experimented with projects following the success of Robocraft, they focused on providing exciting new building tools. This way, players could enjoy more freedom to design complex, physics-simulated creations.The team evaluated how moving the physics from the client side to dedicated multiplayer game servers could enhance the physical interactions between the vehicles and robots created by the players. They discovered that relegating the physics to the server created a range of fun gameplay moments, wherein weight, inertia, momentum, friction, mass, and bounciness were all accurately simulated. In other words, the heavy vehicles could easily push lighter ones, or be combined with joints like pistons, servos, and rotating platforms. Even weapons and explosions applied realistic kickback and force when they hit. All of these experiments, alongside community testing and feedback, culminated in Robocraft 2.In Robocraft 2, players now get to create complex vehicles, take them into battle, and destroy them in 5v5 team battles online. From their experience with the first Robocraft game, the Freejam team knew how competitive and creative their players could be, finding new ways to optimize the provided building tools in order to win battles.This meant that the team had to rely on three key features from their physics engine in order to provide a fair experience for all:Fast performanceRobust, non-glitchy physics simulationAccess to low-level areas of the physics engine to make modifications in the gamePrior to the 2017 announcement of DOTS and ECS for Unity, Freejam Games explored the possibility of building their ECS framework in-house. They then quickly adopted ECS for Unity in its experimental release, starting with Unity Physics. For experiments with server-side tech, they used determinism as a solution for keeping the player clients and simulation on the server in sync, while Unity Physics (which is deterministic) provided the performance.As the game evolved, they moved away from a stateless approach and became early adopters of Havok Physics for Unity. As a stateful system, Havok Physics for Unity ultimately powered the performance of the simulation within the gameplay requirements for Robocraft 2.“The high performance of Havok [Physics for Unity] allows us to have accurate server-side physics in our online game. In turn, that provides several significant benefits including giving all clients an equal representation of the physics for a better quality experience. The fact that the server is authoritative over the simulation also has the added benefit of reducing the opportunity for cheaters.” – Ed Fowler, principal programmer and cofounder of FreejamHavok Physics for Unity helped Freejam Games solve complicated issues. For example, as players defeat opponents, the robots and vehicles fall apart block by block, which can create hundreds of Rigidbodies in the environment. To free up CPU and maintain high frame rates, those inactive Rigidbodies can be put to sleep via Deactivation, a feature that effectively removes the physics from broken pieces temporarily.Player creations in Robocraft 2 consist of many Rigidbodies with Compound Colliders constrained together by joints, which can be smashed or stacked on top of one another. Further improvements to overall performance of the physics simulation can be gained with Collision Caching, which additionally allows for refined simulation of joints/constraints, such as in those stacking situations.Lastly, the Havok Visual Debugger was used to visualize the collision in the game world in real-time. It enabled Freejam Games to identify glitches, snags, and efficiently spot instances where rogue contacts arise. This accelerated their workflow and prompted fast fixes.Want to see Robocraft 2 in action? Add it to your Wishlist on Steam.To help you get started, check out the ECS Physics Samples on GitHub.If you need more guidance, we created a tutorial to help you learn more about Unity’s physics options, including Havok Physics for Unity.By the end of the tutorial, you will be able to do the following:Describe the key benefits of Havok Physics for Unity and Unity PhysicsExplain the relationship between Unity Physics and Havok Physics for UnityIdentify situations where physics solutions for ECS are the right fit for a projectThis tutorial is an introduction to physics solutions in ECS for Unity, tailored to users with an intermediate or advanced level of experience with the Unity Editor. As mentioned earlier, DOTS is Unity’s Data-Oriented Technology Stack, a suite of data-oriented technologies for users looking to make complex projects with highly optimized performance. If you want to learn more about DOTS, we recommend the newly-released DOTS Guide on GitHub.We are also actively engaging with many of you on the DOTS channel, part of the Unity Discord, and in the forums. We look forward to learning more about the projects you’re building with Havok Physics for Unity.

This is such exciting work about environmental forecasting, so we’re happy to have the two of you join us today.   Megan and Wessel, welcome.  MEGAN STANLEY: Thank you. Thanks. Great to be here.  WESSEL BRUINSMA: Thanks.  TINGLE: Let’s jump right in. Wessel, share a bit about the problem your research addresses and why this work is so important.  BRUINSMA: I think we’re all very much aware of the revolution that’s happening in the space of large language models, which have just become so strong. What’s perhaps lesser well-known is that machine learning models have also started to revolutionize this field of weather prediction. Whereas traditional weather prediction models, based on physical laws, used to be the state of the art, these traditional models are now challenged and often even outperformed by AI models.   This advancement is super impressive and really a big deal. Mostly because AI weather forecasting models are computationally much more efficient and can even be more accurate. What’s unfortunate though, about this big step forward, is that these developments are mostly limited to the setting of weather forecasting.   Weather forecasting is very important, obviously, but there are many other important environmental forecasting problems out there, such as air pollution forecasting or ocean wave forecasting. We have developed a model, named Aurora, which really kicks the AI revolution in weather forecasting into the next gear by extending these advancements to other environmental forecasting fields, too. With Aurora, we’re now able to produce state-of-the-art air pollution forecasts using an AI approach. And that wasn’t possible before!  TINGLE: Megan, how does this approach differ from or build on work that’s already been done in the atmospheric sciences?  STANLEY: Current approaches have really focused training very specifically on weather forecasting models. And in contrast, with Aurora, what we’ve attempted to do is train a so-called foundation model for the Earth system. In the first step, we train Aurora on a vast body of Earth system data. This is our pretraining step.   And when I say a vast body of data, I really do mean a lot. And the purpose of this pretraining is to let Aurora, kind of, learn some general-purpose representation of the dynamics that govern the Earth system. But then once we’ve pretrained Aurora, and this really is the crux of this, the reason why we’re doing this project, is after the model has been pretrained, it can leverage this learned general-purpose representation and efficiently adapt to new tasks, new domains, new variables. And this is called fine-tuning.  The idea is that the model really uses the learned representation to perform this adaptation very efficiently, which basically means Aurora is a powerful, flexible model that can relatively cheaply be adapted to any environmental forecasting task.    TINGLE: Wessel, can you tell us about your methodology? How did you all conduct this research?  BRUINSMA: While approaches so far have trained models on primarily one particular data These datasets are a combination of estimates of the historical state of the world, forecasts by other models, climate simulations, and more. We’ve been able to show that training on not just more data but more diverse data helps the model achieve even better performance. Showing this is difficult because there is just so much data.   In addition to scaling to more and more diverse data, we also increased the size of the model as much as we could. Here we found that bigger models, despite being slower to run, make more efficient use of computational resources. It’s cheaper to train a good big model than a good small model. The mantra of this project was to really keep it simple and to scale to simultaneously very large and, more importantly, diverse data and large model size.  TINGLE: So, Megan, what were your major findings? And we know they’re major because they’re in Nature. [LAUGHS]  STANLEY: Yeah, [LAUGHS] I guess they really are. So the main outcome of this project is we were actually able to train a single foundation model that achieves state-of-the-art performance in four different domains. Air pollution forecasting. For example, predicting particulate matter near the surface or ozone in the atmosphere. Ocean wave forecasting, which is critical for planning shipping routes.   Tropical cyclone track forecasting, so that means being able to predict where a hurricane or a typhoon is expected to go, which is obviously incredibly important, and very high-resolution weather forecasting.   And I’ve, kind of, named these forecasting domains as if they’re just items in a list, but in every single one, Aurora really pushed the limits of what is possible with AI models. And we’re really proud of that.   But perhaps, kind of, you know, to my mind, the key takeaway here is that the foundation model approach actually works. So what we have shown is it’s possible to actually train some kind of general model, a foundation model, and then adapt it to a wide variety of environmental tasks. Now we definitely do not claim that Aurora is some kind of ultimate environmental forecasting model. We are sure that the model and the pretraining procedure can actually be improved. But, nevertheless, we’ve shown that this approach works for environmental forecasting. It really holds massive promise, and that’s incredibly cool.  TINGLE: Wessel, what do you think will be the real-world impact of this work?  BRUINSMA: Well, for applications that we mentioned, which are air pollution forecasting, ocean wave forecasting, tropical cyclone track forecasting, and very high-resolution weather forecasting, Aurora could today be deployed in real-time systems to produce near real-time forecasts. And, you know, in fact, it already is. You can view real-time weather forecasts by the high-resolution version of the model on the website of ECMWF (European Centre for Medium-Range Weather Forecasts).  But what’s remarkable is that every of these applications took a small team of engineers about four to eight weeks to fully execute. You should compare this to a typical development timeline for more traditional models, which can be on the order of multiple years. Using the pretraining fine-tuning approach that we used for Aurora, we might see significantly accelerated development cycles for environmental forecasting problems. And that’s exciting.  TINGLE: Megan, if our listeners only walk away from this conversation with one key talking point, what would you like that to be? What should we remember about this paper?  STANLEY: The biggest takeaway is that the pretraining fine-tuning paradigm, it really works for environmental forecasting, right? So you can train a foundational model, it learns some kind of general-purpose representation of the Earth system dynamics, and this representation boosts performance in a wide variety of forecasting tasks. But we really want to emphasize that Aurora only scratches the surface of what’s actually possible.  So there are many more applications to explore than the four we’ve mentioned. And undoubtedly, the model and pretraining procedure can actually be improved. So we’re really excited to see what the next few years will bring.  TINGLE: Wessel, tell us more about those opportunities and unanswered questions. What’s next on the research agenda in environmental prediction?  BRUINSMA: Well, Aurora has two main limitations. The first is that the model produces only deterministic predictions, by which I mean a single predicted value. For variables like temperature, this is mostly fine. But other variables like precipitation, they are inherently some kind of stochastic. For these variables, we really want to assign probabilities to different levels of precipitation rather than predicting only a single value.  An extension of Aurora to allow this sort of prediction would be a great next step.   The second limitation is that Aurora depends on a procedure called assimilation. Assimilation attempts to create a starting point for the model from real-world observations, such as from weather stations and satellites. The model then takes the starting point and uses it to make predictions. Unfortunately, assimilation is super expensive, so it would be great if we could somehow circumvent the need for it.  Finally, what we find really important is to make our advancements available to the community. [MUSIC]  TINGLE: Great. Megan and Wessel, thanks for joining us today on the Microsoft Research Podcast.  BRUINSMA: Thanks for having us.  STANLEY: Yeah, thank you. It’s been great.  TINGLE: You can check out the Aurora model on Azure AI Foundry. You can read the entire paper, “A Foundation Model for the Earth System,” at aka.ms/abstracts. And you’ll certainly find it on the Nature website, too.   Thank you so much for tuning in to Abstracts today. Until next time.   [MUSIC FADES]