2025-09-11 02:44:10 <cleobuline  > forth:  1 2 3 4 5 3 ROLL .S
2025-09-11 02:44:18 <cleobuline  > forthBot:  1 2 3 4 5 3 ROLL .S
2025-09-11 02:44:19 <forthBot    > <5> 1 3 4 5 2
2025-09-11 04:44:19 <forthBot    > Environment for cleobuline inactive, freeing...
2025-09-11 09:59:18 <cleobuline  > forthBot:  1 2 3 4 5 4 ROLL .S
2025-09-11 09:59:19 <forthBot    > <5> 2 3 4 5 1
2025-09-11 11:59:18 <forthBot    > Environment for cleobuline inactive, freeing...
2025-09-11 12:02:06 <cleobuline  > yet forthBot   is still a shit that is useless to pollute the planet
2025-09-11 13:18:41 <cleobuli_   > https://en.wikipedia.org/wiki/Dan_Bricklin
2025-09-11 14:42:47 <KipIngram   > https://imgur.com/a/SHjTofP
2025-09-11 14:44:17 <KipIngram   > https://www.youtube.com/@HansBezemer/featured
2025-09-11 18:21:25 <KipIngram   > So guys - I had a thought this morning and wonder if anyone's see it in play anywhere.
2025-09-11 18:21:37 <KipIngram   > Normally when Forth encounters an unrecognized word, it immediately throws an error.
2025-09-11 18:22:26 <KipIngram   > The thought was that instead of doing that, it could remember the details of that word and mark a pending error, but not throw an actual error until later.  That would give the intervening words an opportunity to "do something" with that symbol and clear the error condition.
2025-09-11 18:22:49 <KipIngram   > So you could put symbol handlers after words instead of before, if that was for some reason of any value.
2025-09-11 18:23:16 <KipIngram   > You could manage that delay period and other details a lot of ways, offering varying amounts of flexibility as to what you could accomplish.
2025-09-11 18:28:09 <vdupras     > do you have an example of interesting semantics that this mechanism would enable? I struggle to imagine them
2025-09-11 18:29:02 <KipIngram   > Not one that I necessarily think I'd WANT to do, but the simplest example I can imagine is that you could say <name> : ... instead of : name ...  I don't see that as of much value, but you COULD - the : could pick the name up from the staging place rather than the input stream.
2025-09-11 18:33:10 <vdupras     > without a compelling use case, a simpler Forth is strictly better.
2025-09-11 18:33:42 <KipIngram   > I agree - I was mostly just wondering if anyone had ever played with such an idea.  After all - all of us as a collective seem to eventually get around to every idea, good or bad.
2025-09-11 21:53:27 <MrMobius    > KipIngram: if I understand what you mean, that's what I'm doing with my forth
2025-09-11 21:53:56 <MrMobius    > a word can call a word that hasn't been defined yet and it doesn't throw an error until you try to run it
2025-09-11 23:55:57 <cleobuline  > MrMobius:  bizarre
2025-09-11 23:56:17 <cleobuline  > forthBot:  : test undefined ;
2025-09-11 23:56:17 <forthBot    > Unknown word in definition: undefined
2025-09-11 23:56:18 <forthBot    > Error: Definition discarded due to error