In the past, there were programming languages that went through 80 years of iteration, and then there were the endless emergence of AI-assisted programming tools such as GitHub Copilot and ChatGPT. What kind of changes have occurred in the development methods of developers? JetBrains, the software tool development company behind the industry's leading Java IDE IntelliJ IDEA and the Kotlin programming language, is based on a survey of 26,348 developers around the world. In the latest "2023 Developer Ecosystem Status" report, it joins the AI trend survey for the first time and answers solved the above problems.
1. What problems do developers mainly solve by using AI tools?
The most common way developers use AI tools is to use natural language to ask general questions about software development. The second is to generate code, code comments or code documentation.
2. What types of generative AI tools are familiar to developers?
69% of developers chose AI text generation tools, and 60% of developers chose code generation tools. Developers are more familiar with AI text generation tools than code generation tools, which may be due to the popularity and ease of use of ChatGPT.
3. Which of the following AI tools are you currently using?
Unsurprisingly, ChatGPT takes the first place with a share of 77%, and GitHub Copilot ranks second with a share of 46%.
"Will the emergence of generative AI tools replace developers?" This discussion has never stopped. As more and more developers try to use AI tools, how will their attitudes towards it change?
4. What’s stopping you from learning to use generative AI tools?
Happily, only 1.6% of respondents have never heard of generative AI tools, and of those who are not yet familiar with generative AI tools, just under half plan to try them in the near future. Security concerns are the most frequently cited barrier to adoption of generative AI tools.
For complex enterprise-level application development, we have to admit that even GPT-4, which has strong programming capabilities, still lacks industry and business know-how necessary for enterprise-level applications; because of its reliance on Prompt, code quality suffers. Great uncertainty; difficult to meet the security requirements of enterprise-level applications and other serious limitations.
This is not only a problem that ChatGPT needs to overcome, many domestic generative AI manufacturers are also working to break through the above limitations. Take the full-stack, fully-automatic software development tool SoFlu software robot launched by Feisuan, a domestic digital technology service provider, as an example:
In terms of professionalism, in addition to the universal components and models provided, users and Feisuan can jointly build personalized components and models through SoFlu software robots based on their long-term accumulation of a large number of industry and business know-how and continue to iterate, thereby producing Applications that are closest to business needs.
In terms of reliability , compared to GPT-4, which is highly uncertain, SoFlu software robots have achieved standardization of software development, from "rule by man" to "rule by law", maximizing constraints on human variables to ensure highly reliable product quality.
In terms of security , SoFlu software robots allow customers to flexibly choose how to store and deploy code according to their needs. The support of privatized deployment can fundamentally eliminate customers' concerns about security and intellectual property rights.
In addition, FuncGPT (Hui Function) launched by SoFlu software robot focuses on Java generation of AI functions. It describes Java function requirements through natural language and generates high-quality, highly readable Java function codes in real time. The generated code can be directly copied to IDEA, or imported into the Java fully automatic development tool function library with one click. During internal testing, many developers and users reported that compared to many code generation tools on the market, the code generated by FuncGPT was of higher quality and could meet their actual work needs. Click here to develop new experiences in seconds
5. Which of the following statements about the future of AI-powered code generation do you agree with?
The data shows that while three-fifths of respondents believe AI coding tools will fundamentally change the software development job market, only 13% believe AI will completely replace developers writing code. Despite this, about one-third believe that software engineering will definitely become prompt engineering.
Overall, developers believe that they will not be replaced by AI. Instead, AI will become a new tool to assist in writing code.
In the context of digitalization, AI tools, on the basis of helping developers improve both efficiency and quality, will free developers to focus more on 20% of their work, think deeply about problems from a higher perspective, and transform from " programmers " to " programmers " Architect " , from " writing code " to " designing programs " , completely unleashing the innovative potential of each developer and crossing the boundaries of capabilities.
For more software development-related consultation, follow the [SoFlu Software Robot] official account.
IntelliJ IDEA 2023.3 & JetBrains Family Bucket annual major version update new concept "defensive programming": make yourself a stable job GitHub.com runs more than 1,200 MySQL hosts, how to seamlessly upgrade to 8.0? Stephen Chow's Web3 team will launch an independent App next month. Will Firefox be eliminated? Visual Studio Code 1.85 released, floating window Yu Chengdong: Huawei will launch disruptive products next year and rewrite the history of the industry. The US CISA recommends abandoning C/C++ to eliminate memory security vulnerabilities. TIOBE December: C# is expected to become the programming language of the year. A paper written by Lei Jun 30 years ago : "Principle and Design of Computer Virus Determination Expert System"