Good Morning! Next.js 14 introduces partial pre-rendering, combining static site generation with server-side rendering without new APIs. This optimization provides significant performance improvements for dynamic content. IBM has launched WatsonX Code Assistant for Z, using generative AI to accelerate COBOL application modernization on the IBM Z mainframe. The solution aims to boost developer productivity and lower costs. Company X has reported a 60% reduction in costs after transitioning from cloud services to on-premises infrastructure. This #CloudExit strategy has allowed substantial savings and increased experimentation.

— Forrest Knight & Meghanadh Vasireddy

Next.js 14 Features

Vercel has released Next.js 14, a significant update to the popular React-based web development framework Next.js, focusing on delivering performance improvements. Here are the key features of this release:

• Partial Pre-Rendering: Next.js 14 introduces a partial pre-rendering ability, combining static site generation with server-side rendering, all without requiring new APIs for developers. This feature optimizes the compiler to handle dynamic content and offers significant performance boosts.

• Turbopack: The Turbopack Rust-based bundler, optimized for JavaScript and TypeScript, is expected to move to "stable" in an upcoming minor release. It has passed 90% of more than 5000 tests, including bugs and potential edge cases for all of the Next.js development features. It also provides a 35% faster local server startup and 80% faster code updates with Fast Refresh.

• Server Actions: Next.js 14 improves the developer experience by releasing Server Actions, allowing developers to define asynchronous server functions. Server Actions can be used for tasks such as revalidating cached data, redirecting to different routes, and setting/reading cookies.

• Metadata Improvements: In Next.js 14, metadata about the viewport, color scheme, and theme is sent to the browser before web page content can be streamed from the server. Blocking and non-blocking metadata have been decoupled, with only a small set of metadata options being blocking. The goal is to ensure that non-blocking metadata does not prevent a partially pre-rendered page from serving the static shell.

• Next.js Learn: Vercel has also announced an overhaul of its free Next.js Learn course, which now includes all the new Next.js features.

Read More Here

IBM Launches WatsonX Code Assistant for Z to Modernize COBOL Applications

IBM has recently launched WatsonX Code Assistant for Z, a generative AI-assisted product designed to accelerate the modernization of mainframe applications, particularly those written in COBOL. This innovative solution aims to address the shortage of COBOL developers and the need for modernization in the face of evolving business requirements.

COBOL, one of the most widely used programming languages for businesses worldwide, has been a cornerstone of enterprise computing for decades. However, the modernization of COBOL applications has been a challenging task due to the complexity of the language and the scarcity of developers with COBOL expertise.

WatsonX Code Assistant for Z is designed to help businesses overcome these challenges by leveraging generative AI and automated tooling to accelerate mainframe application modernization. It enables faster translation of COBOL to Java, enhancing developer productivity on the platform. The solution is purpose-built for targeted use cases and uses AI to support Application Modernization and IT Automation.

The product is designed to increase developer productivity by generating new code from natural language requests and allowing customization of models with your own data. It also provides pre-trained models based on specific programming languages to ensure trust and efficiency for accurate code generation. WatsonX Code Assistant for Z is being designed to assist businesses in leveraging generative AI and automated tooling to accelerate their mainframe application modernization, allowing clients to take advantage of the performance, security, and resiliency capabilities of IBM Z.

The solution includes an on-prem and as a service component, and when purchased together, WatsonX Code Assistant for Z includes the referenced generative AI capabilities. This product is expected to lower the cost and risk associated with mainframe application modernization.

Read More Here

X Celebrates 60% Savings from Cloud Exit

X has reported a 60% reduction in costs following its exit from cloud services, a strategy that has been termed #CloudExit. This shift has not only resulted in substantial savings but also allowed X to increase the pace of experimentation and feature introductions.

The decision to transition from cloud services to on-premises infrastructure was driven by the need to cut costs and streamline operations. The company's cloud spend has dropped from around $180,000 per month to less than $80,000, resulting in an estimated yearly saving of a million dollars. This saving is expected to increase to about $2 million per year, amounting to a staggering $10 million over five years.

However, the transition was not without its initial costs. X had to invest approximately half a million dollars in new machines to replace the cloud rentals. Despite this initial outlay, the long-term savings are expected to far outweigh the initial investment.

This move by X is part of a growing trend of companies leaving the cloud. A 2021 study involving 600 organizations revealed that only 52% of those companies who moved to the cloud decided to remain there. The reasons for this shift include rising costs and the challenges of managing a cloud-based business.

While the cloud offers numerous benefits such as increased scalability, cost efficiency, and agility, many organizations are finding that these benefits do not outweigh the costs and complexities involved. For companies with predictable and steady workloads, on-premises infrastructure can often be a more cost-effective solution.

Read More Here

Google Encourages Migration to Credential Manager API for Enhanced Sign-In Experience

Google has recently announced significant updates to its Google Identity Services (GIS) and is encouraging developers to migrate to the Credential Manager API. This move is part of Google's ongoing commitment to improving the sign-in experience across platforms for both developers and users.

The authentication APIs from GIS on Android, which include One Tap sign-in, Credential Saving, Sign in with Google button, and Sign-In for Android (GSI), can now all be implemented using the Credential Manager. This new development allows developers to integrate with a single API for their authentication journeys.

The Credential Manager API supports multiple sign-in methods, including passwords, passkeys, and federated sign-in methods, providing a more consistent user experience for authentication across apps and sign-in mechanisms. It also supports third-party credential providers on Android 14 and higher, allowing users to use their existing credentials from other providers to sign in to your app.

With the general availability of these APIs in Credential Manager, the individual APIs will be deprecated in GIS. Notably, Smart Lock for Passwords, which was deprecated in 2022, will be removed from the Google Play Services SDK in November 2023. However, to minimize breaking changes that may impact existing integrations, all existing apps in the Play Store will continue to work. New app versions compiled with the new SDK will not be able to access the Smart Lock for Password API, hence the encouragement for developers to migrate to Credential Manager as soon as possible.

Read More Here

Youtube Spotlight

128-core MONSTER: the Mac Pro Apple SHOULD've built!

In this video, Jeff Geerling discusses their upgrade to a 128-core ARM desktop computer, which they claim is the fastest in the world. He upgraded the RAM, CPU, and GPU to enhance its performance. While the system excels in certain benchmarks and tasks, Geerling acknowledges that individual cores on other processors may be faster, and having 128 cores may be overkill for applications that can’t fully utilize them. He also explores the system’s gaming capabilities, the limitations of RAM speed, and the potential for future improvements.

Was this forwarded to you? Sign Up Here