mhagiwara

Hi! Author here. You're right that music and speech carry more than just raw symbols such as intonation, pitch, dynamics, etc. all add important information. But in the framework I'm using, both are still fundamentally “1D” because their core structure unfolds sequentially in time: notes in a score, words and phrases in speech. Even when additional features like accent or dynamics are present, they're layered onto a linear stream.

Recto, by contrast, makes two dimensions part of the syntax itself: elements and rects can expand horizontally or vertically, and they can nest recursively. That's the difference I wanted to highlight.

I redesigned Morse code with modern statistical techniques, including neural networks. NeuralMorse dynamically tokenizes input text and encodes it as sequences of eight tonal alphabets, optimized by word embeddings and the assignment problem.

I'm interested!

This has never been a big problem for me. Some clients have strict policies (e.g., only allowing you to work on their servers) while others don't care as much. In some cases, as a freelancer, you only work on a "proof of concept" prototype assuming you have access to their private data. It is their FTEs job to actually train/implement/deploy this prototype using their private data.

This has never happened to me (yet).

I do have a CPA and we did an S-corp election for 2019 tax return. This is something I didn't get to mention in the post because we hadn't filed the tax yet.

If your clients are choosing you based on where you are and how your rate compares with other freelancers, I think you have wrong clients.