2025-04-12 13:58:41 <KipIngram   > I always preferred starting with general principles.
2025-04-12 14:00:58 <KipIngram   > I guess it depends on the particular material you're looking to learn.
2025-04-12 14:05:22 <KipIngram   > Like, history.  Really no way to start except for specific examples - specific events.  Then maybe over time you start to see common patterns in how things unfold.  But physics - principles first is way better.
2025-04-12 16:55:11 <cleobuline  > forth: LOAD "test.fth"
2025-04-12 16:55:46 <cleobuline  > mforth: 1 2 3 5 7 11 13 17 19  * * * * * * * * PRIME-FACTORS
2025-04-12 16:55:46 <mforth      > Unknown word: PRIME-FACTORS
2025-04-12 16:55:58 <cleobuline  > mforth: LOAD "test.fth"
2025-04-12 16:56:00 <cleobuline  > mforth: 1 2 3 5 7 11 13 17 19  * * * * * * * * PRIME-FACTORS
2025-04-12 16:56:00 <mforth      > 2 3 5 7 11 13 17 19
2025-04-12 16:56:30 <cleobuline  > revised some bugs ...
2025-04-12 21:38:52 <veltas      > KipIngram: In my experience only Physics really works that way
2025-04-12 21:41:34 <veltas      > I definitely agree with the Gelfand Principle though
2025-04-12 21:44:09 <veltas      > E.g. arguably it would have been better to not start with Hello World here: https://lua.org/pil/1.html
2025-04-12 21:45:01 <veltas      > Especially as PIL is aimed at programmers, not novices.  Although that's funny because Lua is often people's first language, and there aren't good novice-focused books for it
2025-04-12 21:45:05 <veltas      > Not as far as I know anyway
2025-04-12 21:46:47 <veltas      > I think it applies more to computing than maths, in maths it's quite typical not to bother with those kinds of examples
2025-04-12 21:47:07 <veltas      > Dijkstra's opinion: discarded
2025-04-12 21:49:42 <veltas      > Has anyone got into Vibe Forthing?
2025-04-12 21:52:54 <veltas      > This is a great thread https://boards.4chan.org/g/thread/105030271
2025-04-12 22:32:58 <cleobuline  > mforth:  LOAD "test.fth"
2025-04-12 22:32:58 <mforth      > Unknown word in definition: 0=
2025-04-12 22:32:58 <mforth      > Error: Definition discarded due to error
2025-04-12 22:32:58 <mforth      > Unknown word in definition: ZELLER
2025-04-12 22:32:59 <mforth      > Error: Definition discarded due to error
2025-04-12 22:32:59 <mforth      > Unknown word in definition: FIRST-DAY
2025-04-12 22:33:00 <mforth      > Error: Definition discarded due to error
2025-04-12 22:34:22 <xentrac     > poor mforth
2025-04-12 22:34:33 <xentrac     > you really need to fix it so it doesn't produce 10 lines of output for one line of input though
2025-04-12 22:34:51 <xentrac     > veltas: I haven't found an LLM that's good at Forth yet
2025-04-12 22:35:53 <identity    > "There's not enough [Forth] around for LLMs to train on"
2025-04-12 22:36:41 <xentrac     > KipIngram: I think that even for physics it's better to start with specific examples (an apple falling, a planet orbiting, billiard balls colliding) than general principles (F = Gm₁m₂/r² = ma)
2025-04-12 22:36:43 <thrig       > and the forth differ in many ways, which may not help
2025-04-12 22:37:49 <xentrac     > usually the differences are in pretty recondite corners of the language
2025-04-12 22:38:10 <xentrac     > well, not if you try to write 4* instead of cells, I guess