Re: yy_state_t

[Paul Eggert]

On 10/13/19 12:54 AM, Akim Demaille wrote:

> > yy_state_num values are typically narrower than int, so following Leibniz's 
> > notational principle that notation should reflect reality, the former type name 
> > should be narrower than the latter, which means your preference is better.
>
> :)  I was unaware of this principle of his.

It was a principle of Leibniz’s characteristica universalis project, where he 
took up the challenge of constructing a new language or notation for 
representing human thought. Descartes had suggested the project in a letter to 
Mersenne but immediately dismissed it as utopian; Leibniz, possibly the best 
notation designer in history, knew about Descartes’s letter and most likely 
figured he could pull it off even if Descartes couldn’t.

Unfortunately there isn’t much written about that project, at least in English. 
And I can’t resist mentioning that later in life Kurt Gödel became obsessed with 
the project and believed that its publication had been suppressed by a hostile 
conspiracy. Although Gödel was mentally disturbed and his obsession has been 
dismissed, it is true that most of Leibniz’s writings have never been published.

But I digress....