Promoting open platform strategy, Qt Group accelerates ecological transformation

In the era of software definition, continuously reducing software development, testing and production costs, improving development efficiency, and shortening the development cycle is an endless topic that is always on the road. At present, the technology boom caused by generative artificial intelligence has lasted for a year. How will AI affect software development and testing, and how will software development tools and platforms change? At the end of 2023, the Qt Global Summit 2023 China Station held in Shanghai released some important trend information.

The last Qt Global Summit China was held in 2021. Two years later, the development of the global economy and technology has undergone tremendous changes. Qt Company acquired two quality assurance tools, the company logo was updated to Qt Group, and the technology strategy also changed Something new has happened. At the just-concluded CES 2024, Qt announced that it would provide solutions through the Amazon Cloud Service (AWS) App Store and support developers to develop on AWS. From the Qt Global Summit 2023 to the just-held CES, Qt released strategic changes Important signal: "Embracing AI from the perspective of augmented intelligence; accelerating platform and ecological transformation through greater openness" will become the core direction of Qt's strategic evolution in the future.

Embracing AI from the perspective of augmented intelligence 

In the past year, AI has become popular all over the world, AI coding assistants have emerged one after another, and the programming paradigm has undergone unprecedented profound changes; programming languages ​​have changed from Java to C, C++ to today's Prompt project characterized by natural language, and the programming threshold has been continuously lowered. These changes have prompted everyone from developers to development tool and platform manufacturers to think about the future evolution direction of programming paradigms and development platforms.

"Having a coding assistant means that developers can generate more code in less time, reduce development workload, and significantly improve development efficiency. So is the quality of the generated code guaranteed? Does it meet system-level and integration requirements?" Qt Group Product Director Luo Miao asked the following questions to developers during his speech at the summit.

Luo Miao believes that although generative AI can help developers generate code, it does not mean that the software developed can achieve the expected quality. Because the code mechanism used to train LLM (large language model) and generative AI is not necessarily suitable for the software mechanism the user wants to create, whether the code generated by the AI ​​meets the standard, whether there is a problem with the standard itself, and whether the generated code conforms to the system Level requirements and whether they meet integration requirements are unknown. In this case, the more code you generate, the more tests you need.

"Whether it is from the perspective of system design, architecture control or software quality, Qt believes that AI, especially augmented intelligence (Augmented Intelligence), can assist humans as a virtual partner, rather than trying to replace humans. And with the The continuous deepening of AI in the field of programming will inevitably involve more and more challenges." Luo Miao said that Qt recommends carefully evaluating the terms and conditions related to the use of any generative AI, whether for open source or commercial use, and hopes that Make software development and testing easier by eliminating boilerplate code and repetitive work through enhanced intelligence while protecting intellectual property and software security.

Based on this, Qt is actively promoting from several levels to resolve the problems encountered by developers using AI programming and release the potential of AI. The first is to introduce more coding assistants into Qt Creator and make them easier for developers to use. Qt Creator IDE has launched a ready-made GitHub Copilot plug-in to support various coding assistant use cases; the second is to use more than 20 years of ability to build stable and secure cross-platform applications, fine-tune large language models, and integrate the intellectual property rights of Qt and Qt developers Safely introduce coding assistants. Recently, Qt is trying to implement smarter API auditing using GPT-4. The third is to explore opportunities for edge AI, including small language models in this field.

Be more open and accelerate platform-based ecological transformation

In an era full of changes, "user needs as the center" is the eternal foundation stone. For software development, faster speed and higher quality are the constant demands of users. When it is confirmed that software development is still dominated by people and AI is auxiliary enhanced intelligence, then development tools, development environments, and software testing need to Make greater improvements around development speed and software quality.

Focusing on user needs to achieve development speed, quality and efficiency, Qt proposed two key technology strategies at the summit: first, to optimize the entire development process through product portfolio and increase product creation speed; second, to realize Qt through greater openness Platform and ecological transformation.

From the perspective of process optimization, the software development process needs to be directly and seamlessly adapted to the cloud in different teams. Cross-functional teams and integrated tool chains will undoubtedly speed up product creation and iteration. For example, user interface design and development often encounter such problems. The design team designs the UI and has a bunch of design documents, but the development team cannot use them directly. When the two teams take over, they will face data docking problems. In order to minimize the loss of information during team handover, Qt has added various bridges to its design and construction tool Qt Design Studio. Regardless of whether designers use mainstream design tools such as Adobe Photoshop, Sketch, Figma, etc., they can use the corresponding bridge to Materials are seamlessly imported and automatically converted to generate UI application code for direct use by the development team.

"Qt has always believed in the power of open source and the advantages of an open tool strategy. An open technical approach can create first-class products that meet needs. For a long time, Qt has supported different third-party build systems, packaging managers and version control systems. Next, Qt will establish a plug-in ecosystem that is interconnected with Qt Creator to provide more extended functions." Luo Miao said several "openings" in a series, announcing Qt Creator's platform strategy.

In Qt's full-stack open strategy, the "open framework strategy" also needs special mention. On the one hand, after 20 years of development, Qt tools and modules have become more and more complete, and the framework has become more and more powerful. Many developers mistakenly believe that Qt tools and modules must be used for development and construction under the Qt framework. In fact, the Qt framework supports the mixture of multiple tools, multiple environments, and multiple technologies, and is an open and extensible framework.

On the other hand, open frameworks are important because users want to reuse a variety of software assets, or expand the original desktop to web or mobile, or reuse part of the system-level code into new projects. Among them, the use of hybrid technologies to create new products has become the norm.

Just like GE's "Vivid" series of ultrasound heart scanners, its background is based on Windows system and uses ActiveX plug-ins. Its interface development uses Qt technology, integrating the Qt interface part with the background C++ program to realize the original code reuse integration.

And such "mixed demands" not only exist in the medical field, but also in all walks of life. To solve the mixed demands, openness is the only and better choice.

It can be said that the open strategy is the core message of this Qt Summit. The launch of this strategy means that Qt has transformed from a software development tool, development environment, and testing and verification manufacturer to a platform manufacturer and ecosystem manufacturer. According to Luo Miao, Qt's first step to build a huge ecosystem using a "platform + plug-in" approach is to integrate other tools in the form of plug-ins. It is possible to integrate other frameworks in the future. As a software development tool, platform, and testing lifecycle manufacturer with a history of more than 20 years, Qt already has a sizable ecosystem. By integrating third-party tools and third-party frameworks, it can speed up the entire ecosystem. Extension.

At present, software development is facing huge development opportunities. If software development can truly realize "thinking without boundaries" and "doing well", we cannot rely solely on Qt. We need to integrate more industrial forces to do better. This is true. On January 9, Qt joined forces with Amazon Cloud Service (AWS) to add its tool suite to the AWS Marketplace. Developers can complete 90% of development and testing in the cloud, and in the final stage of development, the remaining 10% Portable to physical hardware for final testing and verification. This will greatly save development and testing time, reduce the workload of iterative prototyping, and ultimately alleviate OEM supply chain challenges.

It should be said that Qt's open strategy is in line with the trend, and how to attract more development tools and development frameworks to integrate with the Qt ecosystem may become a challenge for Qt in the next round of reform.