← Back to context

Comment by anthk

5 days ago

Learn to live with 1GB of RAM. Either CLI/TUI tools, or software like Dillo or Sxiv. Mainline dillo from their Git and not the outdated one from the repos. The new "dilloc" it's amazing, you need Dillo compiled with socket control support. Then you will be able to use dilloc as a client to spawn commands con scripts. For instance, to rewrite URL's from a list. Hint: the ones from https://farside.link . Youtube to Invidious, X.com to Nitter and so on.

Connect to https://farside.link, https://lite.cnn.com, https://text.npr.org or gopher://magical.fish as a services portal or news source.

Use mpv+yt-dlp if you are under Unix to watch online videos. Complex? Just a little bit first. Incredibily rewarding later. No JS will be needed to play videos on most websites. Also with yt-dlp you are able to save them for later usage.

Try programming with small languages from https://t3x.org and doing Math/Intro to Statistics book with Klong and its manual. s9fes can be a good enough Scheme Lisp to complete the exircies Concrete Abstractions and maybe SICP if you know how to reimplement (frame) and the missing functions. An easy task after CACS.

Consider SQLite+Python+TkInter or TCK+Tk as the DDBB UI on top.

Golang can be great too with 1GB of RAM and a simple n270 netbook, I run Yggdrasil on that, and NNCP too among other tools. Everything with nvi as the editor (basically vi+UTF-8+some status line for help), simple Makefiles git://bitreich.org/english_knight and entr(1) as a tool to watch a directory and spawn 'make' on file changes.

6 comments

anthk

Reply
pjmlp  5 days ago

Yeah, as someone that started with 48 KB, I think most people lost track of what is possible.

Even ESP32 is better than most MS-DOS PCs that were available at the time, and MS-DOS had a rich software environment to chose from.

  • anthk  5 days ago

    Some languages from T3X target even DOS as CP/M, such as T3X0 itself. OFC with lesser capabilities and being slower on older 16 bit machines, but it's an interesting exercise. And, as I said, s9fes it's very capable, it has Unix syscall and Curses support:

    https://www.t3x.org/s9fes/

    Now, Klong and MLite can look like toys; but once you grab a piece of paper and a pen and begin jotting down some trivial Math and algo sketches for both languages, even if you are just a self-called "Software Engineer" (in order to be that in Europe you need a Bachelors+Master degree at least, here it would be an Advanced Vocational Trade), your Math skills (and OFC programming ones) will skyrocket.

    Also, well, Math and Statistics are great to understand the 100% of papers from STEM. Not just the procedures, but in order to parse the results, Statistics are used to check the validity of some experiment.

    Klong https://www.t3x.org/klong/

    Intro to Statistics with Klong https://t3x.org/klong-stat/index.html

    MLite https://t3x.org/mlite/index.html

    And, finally, contrary to LLM's, by using books and solid grounds on both Math and CS you can expect that everything you write it's reproducible. That's it, input always matches the expected output in any case, forever. Not the case for LLM's.