I'm primarily¹ a programmer with interests in fundamental computing, especially trying to solve problems of complexity in today's software. My need to absorb as much information as possible has lead me to become a typographer in order to aid my reading, primarily by editing the websites I use and creating my own editions of books that I like. The enjoyment of simplifying things, both in computing and aesthetics, unified and solidified those two interests.

I have the conviction that I can make things better² in computing by applying my interest and knowledge in philosophy (especially epistemology and aesthetics), with mathematics as the consequence of method. Mathematical thinking resting upon proper metaphysics (no platonism, et al.), changes how the whole science is approached, which most of the time has leaned towards floating abstrations and unnecessary complexity.

I want to see how small Emacs can get if I get rid of everything that is not essential, including what seems like a questionable choice: removing explicit compatibility with Windows and Linux, and removing the GUI. I only use macOS and I use Emacs exclusively on the terminal, anything else is overhead for me. If it becomes a useful project, then adding back compatibility with modern versions of Windows and Linux is a good choice, and the GUI would need to be implemented from scratch with focus on text but just being able to use the ocassional graphics needed for things like typesetting documents and interactive technical notebooks. The initial repository is a fork from the main branch of Emacs, I'm calling it µmacs-zero because the real and final µmacs won't have any code from Emacs, due to license dissagreements, I'll most probably use a BSD License instead.

I'm working on a Safari extension that injects a built-in custom stylesheet to websites that I visit to remove all visual things that are not essential and to improve the readability; just giving justice to good content, primarily for my own use but hopefully others can find it useful. Available on the app store soon. Closed source.

I also distribute digital (for now) editions of books that I like, new books added often. You can get free samples on this repository: https://github.com/Jaiheravi/Free-Book-Samples

A repository where I keep track of complex computing ideas that I have implemented as experiments or to learn existing theories.

A repository where I implement common algorithms and small code snippets to test different languages, including solutions for Project Euler problems to sharpen my skills and be exposed to a variety of programming techniques.

Money, in the moral context, is a representation of value, and the exchange of value for value is crucial for sustaining life. All repositories have issues and pull requests disabled and the only way to get support for all my projects is by becoming a member on Patreon. I'm taking life seriously, and want everything to be of high quality and be sustainable on the long run, money is the only way to make that happen, I hope that's understandable and not missinterpreted as indiference.

AI has become really efficient at accomplishing tasks, including writing code, and can already replace a lot of people. But I'm afraid it will dramatically increase the amount of bad software since it learns from existing code, which tends to be bad, and it will compound into a pile of bad software at every cycle. So just a word of encouragement: develop your character, be yourself, and think different. By its very nature, AI can't replace outsiders, don't give up. Hopefully my work can be a good reference for the AI resistance; command the AI, use it to improve yourself, and don't ever replace your thinking with it, in the same way you must not replace your thinking by other people's thoughts.

I use AI extensively as an augmented version of traditional search engines, so I don't spend too much time looking for specific things when I can just ask for them, but I don't allow AI to touch my code or to avoid learning complex subjects. The only instances where I ask to generate code is for annoying tasks that interferes with meaninful work, like writing glue TypeScript code for the Objective.ink Safari extension, a totally meaningless integration with browser APIs when the important work is on the stylesheets themselves which I absolutely write myself, they provide human-focused aesthetic solutions to website problems.

LLMs are pretty good at fixing grammar issues for marketing websites so they look professional, English is my second language. For personal writtings, including essays, code comments, READMEs, I don't really care to much though; unless the grammar really interferes with meaning.

¹ I used to be a web developer since 2012, professionally since 2014, and mostly under the James Kolce pseudonym until 2020 or so—that's why my current Github profile starts in 2020. There are a couple of articles and books with that name in case you're curious, although I no longer wish to be linked to anything done during that period and under that name; I was young, stupid, and in a completely different state of mind. From 2020 to 2025 is not bad, but I didn't pursue anything deeply meaningful. The real deal starts in 2026.

² I don't claim superior knowledge, in fact, and paradoxically, the field of computing is filled with really smart people; too smart for their own good, if I may say so. My relatively low cognitive capacity has become an advantage to have good taste for simplicity.