This marks the biggest change to the computer keyboard in decades. Although the tech giant isn’t the biggest seller of PCs – that distinction belongs to Lenovo – it runs the most popular Windows operating system. Lenovo has announced its own computers with Copilot keys, as have Dell and HP.

When users type a few words into Copilot, which uses artificial intelligence models from OpenAI, servers in remote data centers do the necessary computational work to produce a response. Microsoft named the new machines as AI PCs.

The key that allows users to open the Copilot panel at any time on the right side of the screen helps with this. Each computer has an Intel processor
Core Ultra, which contains a dedicated neural processing unit, or NPU.

Surface computers have had NPUs since 2019.

When these NPUs are in a computer, the benefits include faster responses and better security. Recently added artificial intelligence features in Windows 11, such as automatic audio transcription and eye contact simulation during video calls, are run by the NPU, which frees up the rest of the chip for other tasks.

“We’re excited to bring to market devices that deliver great AI experiences for our customers,” Microsoft said.

The company reports that this is just the first part of its quest to provide consumers and organizations with improved devices that will not only boost their productivity, but also save some of their time on certain tasks.

The button’s introduction comes two weeks after Apple unveiled the MacBook Air laptops, which feature an upgraded Neural Engine accelerator in the custom M3 chip.

JetBrains has officially launched the stable release of Kotlin Multiplatform. This innovative technology empowers developers to efficiently share code across an array of platforms, including iOS, Android, desktop, web, server-side, and many others. Kotlin Multiplatform has already gained traction among elite development teams at major corporations, such as Netflix, VMware, Cash App, McDonald’s, Electrolux, and Phillips. This groundbreaking release is poised to revolutionize the way we approach cross-platform development.

The framework offers a fully stabilized API, streamlined project configuration, enhanced interoperability with Objective-C and Swift, along with notable improvements in build times and
overall performance.

“Where other technologies abstract away or completely replace platform-specific app development, Kotlin Multiplatform is complementary to existing platform-specific technologies and is geared toward replacing platform-agnostic business logic. It’s a new tool in the toolbox as opposed to replacing the toolbox,” said David Henry and Mel Yahya, Developers at Netflix.

The foremost advantage of Kotlin Multiplatform lies in its flexibility. Developers have the liberty to selectively share a portion of their application’s logic, allowing them to craft native code when the need arises to interface with platform-specific APIs or construct native user interfaces. What’s even more enticing is the synergy achieved by combining Kotlin Multiplatform with Compose Multiplatform, a declarative framework designed for seamless UI sharing across diverse platforms.

“Now that Kotlin Multiplatform is stable, developers can feel much more confident about adopting it for their production projects, as JetBrains will be evolving Kotlin Multiplatform according to the strictest backward compatibility rules. Now is a great time to start using Kotlin Multiplatform for sharing code and to explore Compose Multiplatform’s capabilities for sharing UIs. We will continue improving the core technology, the libraries, and the tooling to provide the best multiplatform development experience to all,” said Ekaterina Petrova, Product Marketing Manager of Kotlin Multiplatform at JetBrains.

This synergy empowers developers with unparalleled creative freedom, enabling them to share as much as 100% of their codebase, all within the Kotlin ecosystem. While Compose Multiplatform stands as a stable solution for Android and desktop, it is being explored experimentally for web applications and is currently in its Alpha stage for iOS.

In addition to Kotlin’s enhanced cross-platform tools, future updates will improve the iOS developer experience by providing direct interoperability between Kotlin and Swift and support for SwiftPM.

All workloads with time series and all performance
There are several enhancements and updates being made available to the global user base. Although InfluxDB has always excelled at processing metrics, the new engine can not only process them, but also handle events and track with dramatic improvements in performance and efficiency.

There are currently no other time series data stores available that are optimized to index, store and analyze metrics, event and tracking data together in one place.

InfluxDB Cloud, with its new high-performance time series engine, can ingest, store and analyze high-volume, high-speed data in real-time and extract metrics on the fly from events at high granularity. This expands the number and variety of use cases that InfluxDB can handle, such as observability and distributed tracking that rely on high-cardinality data.

Lower cost of ownership
InfluxDB Cloud now uses Apache Parquet as its persistence format. This not only contributes to dramatic improvements in compression, but also allows users to extend the value of their time series data by passing it to other systems and ecosystems that use Parquet.

Optimized for low latency queries
Ingesting and storing data in InfluxDB is only half the story. The ability to efficiently search and use all that data is just as important. We’ve optimized the new engine for performance thanks to several under-the-hood updates. While concepts like parallelism, pushdown, and in-memory caching are critical aspects determining the performance of low-latency queries, what really matters to users is that fast and reliable queries improve the user experience.

The new database engine is available in two AWS regions (Virginia (us-east-1) and Frankfurt (eu-central-1)), but new availability zones are planned as well as adding availability for Azure and Google Cloud.

Users already know MediaElement from the Xamarin Community Toolkit, where it was added thanks to the amazing work of community member Peter Foote. While this version was good enough, it also had room for improvement, especially on Android.

That’s why when porting MediaElement to .NET MAUI, the eeup of developers built everything from scratch. This way they made sure they kept all the parts that were already good.

Under the hood
For Android, ExoPlayer was used as the platform analog, replacing Android MediaPlayer which was used for Xamarin. This automatically gets a lot of extra features that are available out of the box, such as HTTP Live Streaming (HLS) video playback, look and feel controls for platform transport, and many other things.

On iOS and macOS, the AVPlayer platform is used, as was done with Xamarin’s MediaElement. Also the one for Tizen is unchanged, using Tizen.Multimedia.Player.

Getting Started
Getting started with MediaElement is easy. First you need to install the CommunityToolkit.Maui.MediaElement NuGet package. This is a separate package from the main Community Toolkit package.

All current MediaElement features are available on the documentation page. Vesluis has also posted a video describing the basics of how to get started with MediaElement.

1. SLSA proof
BuildKit can now create SLSA Provenance attestations to trace a build back to source and make it easier to understand how a build was created. Images built with the new versions of Buildx and BuildKit include metadata such as links to source code, build timestamps, and materials used during the build. To attach the new provenance, BuildKit now creates OCI-compliant images by default.

2. Software material specification
While certificates of origin help record how an assembly was completed, software bills of materials (SBOMs) record what components were used. This is similar to tools like docker sbom, but instead of being required to perform your own scans, the image author can embed the results into the image.

3. SOURCE_DATE_EPOCH
Getting reproducible builds from Dockerfiles has been quite difficult in the past – a full reproducible build requires bit-for-bit precision that produces exactly the same result every time. Even fully deterministic builds will produce different timestamps between runs.

4. Locations of OCI images as named contexts
BuildKit has been able to export OCI image layouts for some time now. Since version 0.11 BuildKit can again import these results using named contexts. This makes it easier to build contexts entirely locally – without the need to send intermediate results to a registry.

5. Cloud cache backends
To achieve good build performance when building in ephemeral environments, such as CI pipelines, you need to store the cache in a remote backend. The latest BuildKit release supports two new storage backends: Amazon S3 and Azure Blob Storage.

When building images, you can provide data to an S3 bucket or Azure Blob storage to automatically store the build cache to be pulled into future builds. This build cache means that even though CI or local runners can be destroyed and re-created, you can still access your remote cache to get quick builds when nothing has changed.

6. OCI image annotations
OCI image annotations allow attaching metadata to container images at the manifest level. They are an alternative to labels, which are more generic, and can be more easily attached to cross-platform images.

7. Build inspection with –print
If you start working in a codebase with Dockerfiles, understanding how to use them can be tricky. Buildx supports the new –print flag for printing details about an assembly. This flag can be used to quickly and easily get information about the arguments and secrets needed for a build, as well as the build targets.

8. Firing functions
The Bake file format for orchestrating builds has been improved.

More release details can be found on the Docker blog and in the changelog. Questions and issues can be asked in the #buildkit channel on the Docker community Slack.

According to the research paper, VALL-E can not only create high-quality personalized speech with just a three-second recording of a tilted speaker acting as an acoustic stimulus. It does this without the need for additional structural engineering, pre-designed acoustic characteristics or fine-tuning. It supports contextual learning and TTS approaches based on zero-shot prompts.

Audio demonstrations of the AI model in action are provided by VALL-E. “Speaker Prompt”, one of the samples that VALL-E needs to duplicate. For comparison purposes, the “Ground Truth” is a pre-recorded excerpt from the same speaker that uses a particular phrase (sort of like the “control” in the experiment). The “Baseline” sample represents a typical example of text-to-speech synthesis, and the “VALL-E” sample represents the output of the VALL-E model.

TTS technology has been integrated into a wide range of applications and devices, such as virtual assistants like Amazon’s Alexa and Google Assistant, navigation applications, and e-learning platforms. It is also used in industries such as entertainment, advertising and customer service to create more engaging and personalized experiences.

Some of the new features in Elastic 8.6 are:

Elastic Enterprise Search is the standard for deploying powerful, modern search and discovery using Elasticsearch – enabling you to search anything, anywhere.

With Elastic 8.6, Elastic Enterprise Search introduces new tools for implementing and managing natural language processing (NLP) in search indexes – accelerating both time to value and higher quality search results. In addition, Elastic 8.6 gives users ingesting data from MongoDB more control and precision with custom filtering and greater overall ingestion flexibility with a new open source connector for network devices.

Elastic Observability is the most widely deployed solution for transforming metrics, logs and traces into actionable IT insights – enabling you to unify observability across your entire digital ecosystem.

With Elastic 8.6, Elastic Observability streamlines end-to-end incident management with alerting workflows through a new OpsGenie connector. Additionally, root cause analysis is accelerated through a new curated journey that helps identify performance or availability issues caused by application dependencies.

Elastic Security brings together SIEM, endpoint security and cloud security, helping practitioners prevent, detect and respond to threats quickly and at cloud scale.

With Elastic 8.6, Elastic Security improves incident investigation and response times across SIEM, cloud, and endpoint – delivering enhanced ransomware protection, improved analyst workflows, and advanced endpoint detection.

An Opsgenie connector is also added in this release. This allows alerts generated in Elastic to be sent to Opsgenie. Signals will be automatically closed in Opsgenie upon recovery in Elastic. Configuring the action to create an alert is supported via a form view or JSON editor.

Filestore Enterprise provides a fully managed, regional network attached storage (NAS) system for GKE. The storage is completely decoupled from the host and requires no additional infrastructure operations to join or decouple volumes. They can be read and written simultaneously from hundreds to thousands of containers.

When combined with the GKE Filestore Container Storage Interface (CSI) driver, Filestore Multishares can allocate up to ten partitions in a single Filestore Enterprise instance. PVs can be allocated from 100 GiB to 1 TiB. When storage is requested using Kubernetes Persistent Volume Claim (PVC), the GKE Filestore CSI driver will pack the volume requests into the Filestore Enterprise instance.

During this time, PVC creation or expansion operations for the same storage class can be blocked. Snapshots are not currently supported with Filestore Multishares. A shared Filestore StorageClass can be used to manually create snapshots through the Filestore API.

To access the Filestore Multishare feature, the GKE Filestore CSI driver must be enabled on version 1.23 or later. Note that this feature is limited to the Filestore Enterprise service level. More details about the Filestore Multishare feature can be found on the Google Cloud website.

Saikat Roychowdhury, senior software engineer at Google, and Akshay Ram, product manager at Google, caution about a few considerations when using this tool. Since the underlying Filestore storage is shared, there is a risk that some “noisy neighbor” will consume an excessive amount of IOPS and bandwidth.

The state map for the Step Function performs the same data processing steps. The existing state map is limited to 40 parallel iterations at a time. This limitation makes it difficult to scale data processing workloads to thousands of elements (or even more). To achieve higher parallel processing prior to the AWS innovation, complex workarounds had to be applied to the existing map state component.

The new distributed state map allows you to write Step Functions to coordinate large-scale parallel workloads within your server applications. You can now iterate over millions of objects, such as logs, images, or .csv files stored in Amazon Simple Storage Service (Amazon S3). The new distributed card state can run up to ten thousand parallel data processing workflows.

The Step Functions distributed card supports a maximum parallelism of up to 10,000 parallel executions, well above the parallelism supported by many other AWS services. You can use the maximum concurrency feature of the distributed map to ensure that you do not exceed the concurrency of a downstream service. There are two factors to consider when working with other services. First, the maximum concurrency supported by the service for your account. Second, the burst and ramp rates determine how fast you can achieve the maximum concurrency.

The new feature is generally available in a subset of AWS regions, including Ohio, Northern Virginia, Singapore, Frankfurt, and Ireland.

The embeddings in the new model are digital representations of concepts converted into sequences of numbers that make it easier for computers to understand the relationships between these concepts. Since the initial launch of the OpenAI/embeddings endpoint, many applications have incorporated embeddings for customizing, recommending, and searching content.

Model enhancements
Higher performance. text-embedding-ada-002 outperforms all legacy embedding models on text search, code search, and sentence similarity tasks and obtains comparable performance on text classification. For each task category, we evaluate the models on the datasets used in the old embeddings.

Merging capabilities. We have greatly simplified the interface of the /embeddings endpoint by merging the five separate models shown above (text-similarity, text-search-query, text-search-doc, code-search-text, and code-search-code) into one new model.

Longer context. The context length of the new model has been increased fourfold, from 2048 to 8192, making it more convenient for working with long documents.

Smaller embedding size. The new embeddings have only 1536 dimensions, which is one-eighth the size of the davinci-001 embeddings, making the new embeddings more cost-effective when working with vector databases.

Reduced cost. We have reduced the price of the new embeddings by 90% compared to the old embeddings of the same size.

With the introduction of text-embedding-ada-002, embedding technology has advanced significantly. It is an invaluable tool for a variety of applications and users due to its powerful combination of efficiency, affordability and usability.