Windows Developer Day Review, too Review

Windows Developer Day Review, too Review

The first Windows Developer Day of this year was officially held at one o'clock on January 1, March 8, Beijing time.

Because it was too late to watch the live video, I also watched the replay at the company the next morning. There is a lot of information worth studying throughout the conference. Let's sort out several important points first. If you are interested, you can study them together. There will also be articles to analyze each key point in detail later.

This Windows Developer Day demonstrates the next-generation Windows 10 SDK capabilities. You can download the Windows Insider Preview Downloads, therefore, you must register a Windows Insider account to access the service properly.

The following describes the highlights of Windows Developer Day:

• Modern Application Experience
• Windows AI Platform
• Windows IoT

 

Modern Application Experience

1. MSIX

GitHub MSIX Packaging SDK

The new packaging tool after MSI can package UWP, WPF, Windows Forms, and Win32 programs. It is already open source on GitHub and deserves further research.

In addition, MSIX also includes a cross-platform solution:

2. Advanced Installer Express Edition

Advanced Installer

A tool for converting Desktop exe msi to UWP Appx files is provided free of charge and by a third party. With this tool, you can quickly generate an Appx file from the Desktop App and publish it to Microsoft Store.

In addition, Advenced Installer also integrates virtual machines to automatically start the virtual machine to run and transpose the exe msi file to generate an Appx file and return it to your development machine.

In addition, a friendly UI allows developers to edit the tile, application information, dependencies, and other information of UWP Appx. The editing experience is similar to that of Appx of Visual Studio. the manifest graphic editing interface is clear and intuitive.

3. Fluent Design

Fluent Design System

Design and UI for UWP apps

Fluent Design is a key Design and development method recommended by Microsoft sdks in recent versions. Microsoft's own applications, such as setting interfaces and calculators, also applied this Design method immediately. Because it is not the focus of this release, I will not talk about it more, but I will explain it in detail later.

4. Adaptive cards

GitHub: Adaptive cards

It provides a more flexible way for developers to display content and interact with applications and users. Adaptive crads 1.0 is also open-source in GitHub. In addition, five sdks, namely JS,. Net, UWP, Android, and iOS, can be tested in Bot Framework WebChat, Cortana Skills, and Windows Timeline. According to the presentation, this year's Build 2018 will be further presented and expected!

 

Windows AI Platform

GitHub Windows Machine Learning

Windows Machine Learning samples

In the next-generation Windows 10 SDK, developers can evaluate the trained AI model on the Windows platform and conveniently Add the ONNX model file to the UWP application in Visual Studio, as part of Assets. In the Windows 10 SDK, NameSpace is "Windows. AI. MachineLearning. Preview". After the official release, the NameSpace should be adjusted accordingly to remove Preview. Sample can be seen in the second link, and I will share it in detail later.

Windows AI Platform has several notable advantages:

• Hardware acceleration is on hardware devices supporting DirectX12. Windows ML can use GPU to evaluate the model for acceleration.
• Local evaluation Windows ML can use local hardware for model evaluation, reducing the server traffic cost and server pressure caused by model uploading to the cloud. You can get results more quickly and conveniently.
• In the machine vision scenario, Windows ML simplifies and optimizes the processing of images, video files, and video streams, and processes the Input Source and the camera pipeline.

ONNX is an ONNX Model file format standard developed by Microsoft, Facebook, Amazon, and other companies. If you are not using ONNX, it doesn't matter. Windows ML tools can easily convert Core ML, Scikit-Learn, XGBoost, and LibSVM to ONNX. Apply to all devices of the Windows series.

Windows ML architecture:

Here is Microsoft's official Blog record for Windows AI Platform: Windows Blogs AI Platform for Windows Developers

 

Windows IoT

Windows 10 Internet of Things

Windows IoT has two versions:

• Windows IoT Core is suitable for small smart devices. The unit price of the NXP module is less than $50 for developers to create low-cost IoT solutions.
• Windows IoT Enterprise is applicable to Powerful PC and Server devices and can play a strong role on large devices, such as bank ATM, medical devices, and industrial devices.

We are delighted that Visual Studio, Nuget, remote debugging, and all other development methods we are familiar with, no matter which version we choose. Windows IoT relies on Intelligent Edge and Intelligent Cloud.

Intelligent Edge

It is an interface connecting the cloud and the real world. It is used to send data from the cloud and respond to smart devices in the real world. Microsoft also focuses on several Partner: VW's MQB platform, Misty One Robotics, and Xogo. If you are interested, refer to the link above for details.

Intelligent Cloud

It can process all the data we obtain in the real world, respond to and process the data, and does not rely on the developer's own server devices. Developers can easily apply cloud-based microservices on Windows. With the help of Azure IoT Edge, this feature is still in the preview version. I believe it will be officially released soon.

 

This release of Windows Developer Day has made us very much look forward to the future development of the Windows platform, whether it is Desktop App, UWP App, Windows IoT, or Windows AI, especially Windows AI applications in IoT and apps. I will also do more detailed research and sharing later.