Even so, AI will propose different things at different times and you still need an experienced developer to make the call. In the end it replaces documentation and typing.
Even so, AI will propose different things at different times and you still need an experienced developer to make the call. In the end it replaces documentation and typing.
For public facing projects - your documentation just became part of the LLM's training data, so its now extra important your documentation is thorough and accurate because you will have a ton of developers getting answers from that system.
For private projects, your documentation can now be fed into a finetuning dataset or a RAG system, achieving the same effect.
I remember that book explain the command, what the purpose of the command was, the typical scenario for why you would need such command.
It explained the options. No silly explanation like "-sort" sorts the output.
It explained the return values also in detail.
Explained the errors also in detail and what might cause the error.
Most functions today are "explained" by reordering the words in the function name: "canUpdate()" returns true if $x can update.