Chain: white shield (part 1)

💭 I find that I script out an entire conversation, sometimes returning the interesting parts before the other person is ready to receive them.

💡 that is: she says her entire side of the conversation before you get a chance to talk.

🔥 this is how i do my emails.

⚙️ Claude is writing the README files for the React and Swift apps.  This means we are about to launch.

💭 My findings have been, that a policy of only LLM-written code is possible today.

✨ It is the equivalent of a system where the human only writes the comments; the actual code is written by the machine.

💡 where the boundaries between code and comment are, remains unclear.

🔥 trust not in those who work in PHP.

💡 However: systems where one cannot READ the code are beyond our current ability.  Humans are better at debugging problems on production-size codebases.

🔥 give it three years, tops.

How does the Atacama project interact with the production efforts of Yevaud Platforms LLC?

⚙️ it would be great to have a useful "reply-to" button.

Is it feasible to adopt a policy where ALL code must be written by an LLM?

I think the answer is yes. 💡 but, it is not yet possible to adopt a policy where humans cannot read the code.

⚙️ a mis-matched parenthesis will make the block-quote not render. I'm not sure if this is a bug or a feature.

✨ trust not in those who do not proof their posts.

Claude is writing the README for the files.

This means that we might achieve a Project Milestone. The ability to require all changes relate to a change to the README file. ⚙️ This refers to the fact that we are operating on a "comments only" approach to the written React / Swift code. 🔥 all edit requests must be phrased in the form of a question.

Once a code base is mature enough, this becomes a reality.

But, initially, the goalposts are too remote.