acmiyaguchi

The idea of using stenographic techniques to exfiltrate data is interesting, but I don't quite follow the general method outlined in the repository -- either through the generated documentation or code. The threat model and case studies seem contrived. I find it hard to believe that folks would expose data via RAG that they wouldn't want users of the underlying system to be privy to.

There's too much fluff here to be useful. I imagine having something that is concise and concrete would make it more appealing to others. But as-is, it's missing a good technical summary and demonstration.

The first paragraph seems self-explanatory, and the templating language is simple enough that you can infer what it does when it runs. There's nothing about this that seems JS ecosystem specific, other than being implemented in a language that itself is implemented in javascript. I can imagine building a stamp interpreter in Python for example, it's not doing very much other than touching and catting text into files.

Getting an interview for global entry is actually pretty easy, you can just do it on the return leg of international travel with no appointment needed.

Interesting, is this how they're currently structured? It sounds a lot like Mozilla with the Mozilla Foundation and Corporation.

A first or second year (undergrad) student doing theoretical science to this level is astounding to me. It was an upper-division course for me, albeit a prerequisite for things like compilers and cryptography, so I'm sure it could be put earlier in the journey.

Sipser's "Introduction to the Theory of Computation" was the book I had to read, and it certainly makes this post more accessible.

I agree, it changes the semantics of dictionaries in a way that doesn't feel right. There are several things that are off-putting, but another example in vein with the mixed key types is the implicit ordered dict.

Modeling is a process to help explain phenomena. I'm doubtful a particular deep neural network architecture can replace even the few examples provided in the article. It might be a useful tool to augment, but it won't replace the high level idea of being able to build and interpret computer models.