[What has changed with ChatGPT. 2023 3… | by V | Medium https://voluntas.medium.com/chatgpt-%E3%81%A7%E4%BD%95%E3%81%8C%E5%A4%89%E3%82%8F%E3%81%A3%E3%81%9F% E3%81%8B-dd78c27e218b] Voluntas.
This article describes how the author’s development style has changed with the advent of ChatGPT. The main points are as follows
- The author uses ChatGPT Plus primarily to ask questions about Go, TypeScript, W3C and IETF specifications. They are looking for hints and clues rather than accuracy.
- When implementing the WebAuthn server, we consulted with ChatGPT on issues such as the number of bytes in the signature, removing zero-byte padding, suppressing output when testing Zerolog, and resolving VSCode errors.
- Utilize ChatGPT for test code generation for property-based testing. fast-check and vitest are used, and ask ChatGPT to generate jest-based code and modify it to conform to vitest.
- When learning new techniques or developing areas of unfamiliarity, he asks questions on ChatGPT and learns through a combination of official documentation and actual coding. This has significantly reduced the frequency of searches on Google.
- For the type of developer like the author who needs timely information, ChatGPT dramatically improves efficiency. In particular, I find the generation of property-based tests very useful.
- On the other hand, I don’t really feel the need to ask ChatGPT about Erlang/OTP and WebRTC, which are areas of expertise.
- ChatGPT has changed our development style dramatically, allowing us to ask questions immediately if we don’t understand something; we have also introduced GitHub Copilot for Business, which uses a different configuration than our public code.
- areas of weakness are no longer troubling him, and he feels that his efficiency has improved significantly. However, he is unsure if he will be able to adapt to this change and believes that he may lose his job.
Overall, the report shows that the advent of ChatGPT has brought about significant changes in the authors’ development style. Positive impacts, such as increased efficiency and complementing areas of weakness, are emphasized, but at the same time, concerns about increased dependence and difficulty of adaptation are expressed.
This page is auto-translated from [/nishio/ChatGPT で何が変わったか](https://scrapbox.io/nishio/ChatGPT で何が変わったか) using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I’m very happy to spread my thought to non-Japanese readers.