oneearedrabbit

[oneearedrabbit @oneearedrabbit]

@borisjabes Just a fun fact that I may have mentioned to you before: in Toki Pona, "need" and "want" are essentially equivalent--you are not supposed to want something unless you also need it. A knee jerk reaction to "Make something people want".

[oneearedrabbit @oneearedrabbit]

@FSouchu Fascinating! No sprite stacking this time?

[oneearedrabbit @oneearedrabbit]

@borisjabes sudo_accountadmin_breakglass

[oneearedrabbit @oneearedrabbit]

@JohnMeuser Just ordered the 4th edition.

[oneearedrabbit @oneearedrabbit]

@JohnMeuser I appreciate that It is grokable.

[oneearedrabbit @oneearedrabbit]

@JohnMeuser I would appreciate an introduction to predicate functors and logic, as this is a gap in my knowledge indeed. While Google is helpful, it seems like you have been doing some interesting work in this area.

[oneearedrabbit @oneearedrabbit]

@JohnMeuser ...mental model is that rank and shape are helpful tools (feel free to correct me if there's a better term) to generalize functions to work on arrays of arbitrary dimensions.

[oneearedrabbit @oneearedrabbit]

@JohnMeuser I gotta hop, one needs to pay the rent. Lovely chat. We should do this more often.

[oneearedrabbit @oneearedrabbit]

@JohnMeuser That is, in fact, trivial. Rank, shape, depth etc are (ahem) metadata, but, more importantly, these are types, broadly speaking.

[oneearedrabbit @oneearedrabbit]

@JohnMeuser Why bother? The reason is that I need an executable whiteboard language that doesn't require a machine to crunch numbers to communicate ideas. It is similar to how you implement algorithms on a piece of paper, but in a terse way that eliminates redundant constructs.

[oneearedrabbit @oneearedrabbit]

@JohnMeuser This turned out to be a bit rambling. My apologies. My biggest worry is the linear representation of memory in Forth. I was not ambition enough to factor this out differently and ultimately decided to approach things from a different angle.

[oneearedrabbit @oneearedrabbit]

@JohnMeuser We learn best from our mistakes :-) I intuitively brought familiar concepts and patterns of software engineering to the APL universe. However, this approach sacrifices the suggestive generality principle, which is crucial in APL programming.

[oneearedrabbit @oneearedrabbit]

0 → 1

[oneearedrabbit @oneearedrabbit]

(Two days later) I understand now that this is a poorly written APL code.

[oneearedrabbit @oneearedrabbit]

)ed pic