Marceau, Fisler and Krishnamurthi, "On novices' interactions with error messages"

https://cs.brown.edu/~sk/Publications/Papers/Published/mfk-mind-lang-novice-inter-error-msg

A rare empirical study that I found extremely useful.

Outside of a univ course, my lessons are slightly different from the paper's:

• Maintain a map of error messages for a project, use this to minimize the vocabulary of error messages.
• Be enigmatic rather than risk misleading.
• Make affordances beyond error messages optional. Color-coding has a cost. Not highlighting may be better.

permalink

I'm tempted to blindly implement https://github.com/JonathanMEdwards/subtext10/blob/master/doc/language.md in Mu without questioning any of its design choices.

permalink

Hackiest possible support for rendering Unicode combining characters using GNU Unifont

This is the equivalent of backing up a typewriter by one character and overlaying a second letter on the same space.

Main project page: https://github.com/akkartik/mu

permalink

Unicode in Mu's network-less read-only browser

Main project page: https://github.com/akkartik/mu

Before:

After:

permalink

The Mu computer now loads 140KB of Unicode glyphs from its system font

Unicode blocks now supported: latin, greek, cyrillic, armenian, hebrew, arabic, syriac, thaana, n'ko, indian (ISCII), sinhala, thai, lao, tibetan, myanmar, georgian (< U+1100)

Caveats:

• No support for combining characters yet (https://en.wikipedia.org/wiki/Combining_character) This makes the other languages I know (Hindi, Tamil) well-nigh useless.
• Unifont's glyphs for the non-Latin languages I know turn out to be quite spectacularly ugly.

permalink

A network-less, read-only browser built up from machine code

https://archive.org/details/akkartik-mu-2021-08-15 (video; 5 minutes; includes instructions to try it out)

A lot gets said about simplicity in software, about essential vs accidental complexity. If you really want a simple stack that empowers everyone, it isn't enough to just eliminate accidental complexity (even if we could all agree on what it is). You need to also avoid other people's essential complexity.

Main project page: https://github.com/akkartik/mu

permalink

It's amazing how fast computers are. Before I took the trouble to build a search index I figured I'd try the simplest possible way to search every single post and comment of 5 years of archives of a fairly active community. 150MB of text.

It's instantaneous.

Running emulated on Qemu. Without any acceleration.

permalink

This talk really puts Mu's accomplishments and failures in context.

Timothy Roscoe, "It's Time for Operating Systems to Rediscover Hardware"

My next step maybe: assemble hardware. Some candidate specs:

• Screen with graphics
• Keyboard
• Persistent storage
• Network
• Touch screen
• Suspend/hibernation

No BIOS/UEFI/Linux/mobile.

Maybe an MNT Reform?

I worry about debugging flow. But Mu had the same problem at the start.

permalink

Images (kinda) and files (kinda) on the Mu computer

The Mu computer has only 256 colors by default, but approximates arbitrary RGB combinations using dithering.

https://archive.org/details/akkartik-mu-2021-07-30 (video; 9 minutes)

Main project page: https://github.com/akkartik/mu

permalink

A little game: guess the result of mixing two colors

Testimonial from 4 year old: this is the best program you've made.

Source: http://akkartik.github.io/mu/html/apps/color-game.mu.html

Main project page: https://github.com/akkartik/mu

permalink