Design Decision: Choosing llm_do() as the Function Name #14
wu-changxing
started this conversation in
Ideas
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Date: 2024-01-29 (Updated 2024-01-30) | Status: Decided → Revised
Summary
We chose
llm_do()over the shorterllm()because functions need verbs to be immediately understandable. Whilellm()is shorter, it reads as a noun and confused users about what action it performs.The Problem with
llm()Initial user feedback revealed:
llmreads as a noun, not a verbllm()does at a glanceWhy
llm_do()Wonllm())Options We Considered
llm()- Too noun-like, unclear action ❌llm_oneshot()- Too verbose (11 chars), enterprise Java feel ❌llm_ask()- Only works for questions ❌llm_go()- Too vague ("go where?") ❌llm_gen()- Implies generation only, not extraction ❌llm_call()- Could confuse with function calls ❌ask()- Natural but too informal for all use cases ❌llm_do()- Clear, versatile, professional ✅Testing Against Real Use Cases
llm_do()reads naturally across ALL use cases.Lessons Learned
llm()only became clear through usageRead the full decision: docs/design-decisions/002-choosing-llm-function-name.md
Beta Was this translation helpful? Give feedback.
All reactions