A Look At The Data: Blogs, Forums, And The Rise Of LLM Tools | HackerNoon

Table of Links

Abstract and 1 Introduction

2. Prior conceptualisations of intelligent assistance for programmers

3. A brief overview of large language models for code generation

4. Commercial programming tools that use large language models

5. Reliability, safety, and security implications of code-generating AI models

6. Usability and design studies of AI-assisted programming

7. Experience reports and 7.1. Writing effective prompts is hard

7.2. The activity of programming shifts towards checking and unfamiliar debugging

7.3. These tools are useful for boilerplate and code reuse

8. The inadequacy of existing metaphors for AI-assisted programming

8.1. AI assistance as search

8.2. AI assistance as compilation

8.3. AI assistance as pair programming

8.4. A distinct way of programming

9. Issues with application to end-user programming

9.1. Issue 1: Intent specification, problem decomposition and computational thinking

9.2. Issue 2: Code correctness, quality and (over)confidence

9.3. Issue 3: Code comprehension and maintenance

9.4. Issue 4: Consequences of automation in end-user programming

9.5. Issue 5: No code, and the dilemma of the direct answer

10. Conclusion

A. Experience report sources

References

A. Experience report sources

This appendix contains a list of sources we draw upon for the quotes and analysis in Section 7. While all sources were included in our analysis, we did not draw direct quotes from every source in this list.

A.1. Blog posts and corresponding Hacker News discussions

1. Andrew Mayne, March 17 2022, “Building games and apps entirely through natural language using OpenAI’s code-davinci model”. URL: <https://andrewmayneblog.wordpress.com/2022/03/17/building-games-and-apps-entirely-through-natural -language-using-openais-davinci-code-model/>. Hacker News discussion: https://news.ycombinator.com/item?id=30717773

2. Andrew Mouboussin, March 24 2022, “Building a No-Code Machine Learning Model by Chatting with GitHub Copilot”. URL: https://www.surgehq.ai/blog/building-a-no-code-toxicity-classifier-by-talking-to-copilot. Hacker News discussion: https://news.ycombinator.com/item?id=30797381

3. Matt Rickard, August 17 2021, “One Month of Using GitHub Copilot”. URL: https://matt-rickard.com/github-copilot-a-month-in/.

   Nutanc, November 15 2021, “Using Github copilot to get the tweets for a keyword and find the sentiment of each tweet in 2 mins”. URL: https://nutanc.medium.com/using-github-copilot-to-get-the-tweets-for-a-keyword-and-find-the-sentiment-of-each-tweet-in-2-mins-9a531abedc84.

4. Tanishq Abraham, July 14 2021, “Coding with GitHub Copilot”. URL: https://tmabraham.github.io/blog/github_copilot.

5. Aleksej Komnenovic, January 17 2022, “Don’t fully trust AI in dev work! /yet”. URL: https://akom.me/dont-fully-trust-ai-in-dev-work-yet.

A.2. Miscellaneous Hacker News discussions

1. https://news.ycombinator.com/item?id=30747211

2. https://news.ycombinator.com/item?id=31390371

3. https://news.ycombinator.com/item?id=31020229&p=2

4. https://news.ycombinator.com/item?id=29760171

5. https://news.ycombinator.com/item?id=31325154

6. https://news.ycombinator.com/item?id=31734110

7. https://news.ycombinator.com/item?id=31652939

8. https://news.ycombinator.com/item?id=30682841

9. https://news.ycombinator.com/item?id=31515938

10. https://news.ycombinator.com/item?id=31825742