[FRIAM] mental imagery

[Santafe]

It’s interesting; 

probably either common knowledge for decades, or even one of the inputs to the design.

Human languages have found their way to this kind of structure-level type checking with such devices as gender systems and measure words.

Gender systems: moscato is a grape, and noce moscata is nutmeg in italian.  Or testo and testa for the big and little heads (note the prioritization).

Measure words cover nearly everything in Chinese, and also have a large grammatical role in Japanese, though to me the latter seems less so.  Also odd, in the sense that Japanese is structurally nearer to Turkic languages than to anything Sino-Tibetan.  I haven’t learned whether Turkic languages have something categorical that would presage the things that appear as measure terms in Japanese, and it is not obvious to me how something structural and ubiquitous like measure words could be “borrowed” across such a large structural difference.  Vocabulary and a writing system is one thing; structure is another, though I can think of small examples (Arabic to Urdu, where one uses prefixes and the other suffixes, but those are tiny single instances).

Eric



> On Nov 25, 2025, at 17:46, Marcus Daniels <marcus at snoutfarm.com> wrote:
> 
> I agree that variable/field naming is a slippery slope, but the better the typing and typing system, the easier it is to use variable names like “obj” without it becoming unreadable.  The information that writers like to hide in the names is encoded in the types instead, and the compiler can enforce type consistency.  Names are a potentially misleading style.  
> 
> I think any of the frontier models can convert between natural language and a language like Lean 4.   The encoded semantics are still loaded with all the common biases of fluent users of English and western culture, but at least the result is definite, computer readable, and can be refined or shown to be wrong once in some context after adopting all that implicit meaning.
>  
> From: Friam <friam-bounces at redfish.com> on behalf of glen <gepropella at gmail.com>
> Date: Tuesday, November 25, 2025 at 9:59 AM
> To: friam at redfish.com <friam at redfish.com>
> Subject: Re: [FRIAM] mental imagery
> 
> Yeah, I like it too. But I maintain my worry that what's important here is *our* ability to [un|re]bind the symbols. And what Marcus' "literate" code does is cajole us into a particular binding. It's a classic confidence trick. (To be clear, that's a good thing.) Rather than name one's variables "x" or "P", we name them mnemonically so as to share subjectivity with others. Mostly, we use positive affect names. Few people would find it easy to read code where the names were cuss words, words like "nazi", or violent. So we name them not only so as to communicate their *intended* purpose (never mind that intention can be wrong/misleading), but also as a coercive/rhetorical act.
> 
> Changing gears a bit, I ran across the "as if" personality <https://en.wikipedia.org/wiki/Identity_disturbance>. I can't help but wonder about a room full of agents "concealing their inner emptiness, living *as if* they had genuine feelings and desires." >8^D I'm at risk of Get-Off-My-Lawn, here. But 90% of the time, when I'm in a room with more than, say, 5 people, it *feels* to me like they're all philosophical zombies, maybe me included.
> 
> Are we all *actually* "as if" personalities? And those who think they're not are delusional?
> 
> On 11/24/25 5:06 PM, Steve Smith wrote:
> > I'm finding your Lean4 fascinating for it's balance between intuitive enough to (almost) read and (known to be) formal enough to trust to be testable/executeable.
> > 
> > Reminds me vaguely of the semester I learned BNF and kept finding myself expressing (only to myself) observations about the world in that idiom... later Prolog captured that part of me (for a while) .
> > 
> 
> 
> -- 
> ¡sıɹƎ ןıɐH ⊥ ɐןןǝdoɹ ǝ uǝןƃ
> ὅτε oi μὲν ἄλλοι κύνες τοὺς ἐχϑροὺς δάκνουσιν, ἐγὰ δὲ τοὺς φίλους, ἵνα σώσω.
> 
> 
> .- .-.. .-.. / ..-. --- --- - . .-. ... / .- .-. . / .-- .-. --- -. --. / ... --- -- . / .- .-. . / ..- ... . ..-. ..- .-..
> FRIAM Applied Complexity Group listserv
> Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fbit.ly%2fvirtualfriam&c=E,1,CAP6FhYezq1caINQoALl7gq2g_585JwGRRzqEnJLiKPuPn6w65WJWT1Lyc7B_70VungnHDXRm9EynJgwmtPXxbUHYeDHox8fEij9zRAo-A,,&typo=1
> to (un)subscribe https://linkprotect.cudasvc.com/url?a=http%3a%2f%2fredfish.com%2fmailman%2flistinfo%2ffriam_redfish.com&c=E,1,QxybKU7ftFae_hkJtuas4QN3nAfzWJvasgOS4qPyv-y3VN6HytbmQiMlIN-eXS6AcuytOVYM0gxHYphpsW2B5-bwpRSBdNFXa09m6jMqNvpq5Kg,&typo=1
> FRIAM-COMIC https://linkprotect.cudasvc.com/url?a=http%3a%2f%2ffriam-comic.blogspot.com%2f&c=E,1,EhueRF2nyHL3r_WqoTs2bHU9qGwOy7LW3SD77pM1lnT2LFaHexeO7gBoU79NJw7MSWYBCEjrRlQgnSaaCPvDQjm1ZZltVxp9VuCUfDWrqqT7CaWdlxZv1Cai&typo=1
> archives:  5/2017 thru present https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fredfish.com%2fpipermail%2ffriam_redfish.com%2f&c=E,1,qtN9M9Pvvz-thHBwAfpjU2BsQdicuI-tW6we63_fX5FDtbWJuN9yKggtSJw-M-Og466hjUNbVVdbbYimGx80YYUmmluC_5G2t-4sOmob&typo=1
>  1/2003 thru 6/2021  http://friam.383.s1.nabble.com/

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://redfish.com/pipermail/friam_redfish.com/attachments/20251127/7b1199e3/attachment.html>