[alicebot-archcomm] Attribute var

[Dr. Richard S. Wallace]

Yes, even John McCarthy, the inventor of Lisp, says this.

We learned recently that McCarthy was at Princeton during time when Church
was working on Lambda calculus there.

Martin Davis said, "Church was my thesis
advisor. McCarthy wrote his thesis on ordinary differential equations under
Lefschetz. He developed LISP years later when he was at MIT. When asked, he
insisted that he had never studied Church's lambda calculus."

I could have written AIML in Lisp syntax from the beginning, but would it
have become so popular?  When you see the phenomenal growth of something
like HTML, it makes you wonder whether the simplest, most elegant syntax is
what most people really want.   Tim Berners Lee could have used Lisp syntax
too for that matter.  For whatever reason, people seem to like XML syntax on
average better than Lisp syntax, or any other syntax, even though people
like you and I know there is a better alternative.

I was betting on the fact that more people are familiar now with expressions
like <fun>...</fun> than (fun...).  Especially because, it is a markup
language.  To me that means, in the simplest form, no markup at all.  For
instance, an unmarked FAQ document in ASCII could be considered an
"unmarked" AIML document.   Many templates in any case are just plain,
unmarked text.  So I was appealing to that "branding" if you will of the
"markup language" concept.  For good or bad, we are now living in a world
where millions of people associate "markup" with <this>.  We are aiming for
wide adoption of the standard.

But this is all getting very [OT], unless we turn it into that "Vision"
thread, which I think might be better outside the Arch Comm list, until they
have something to propose, even if many of us are on the Vision
Subcommittee.

Rich

----- Original Message -----
From: "John Foderaro" <jkf@franz.com>
To: <alicebot-archcomm@list.alicebot.org>
Sent: Thursday, March 14, 2002 10:34 AM
Subject: Re: [alicebot-archcomm] Attribute var


>
>  It's true that lambda calculus inspired having functions as first
> class objects, but the idea of programs as data and data as programs
> is something computer scientists people came up and perfected.
>
>  XML is a data representation and people are trying to add
> evaluation semantics to XML data (AIML does this as well of course).
> _______________________________________________
> alicebot-archcomm mailing list
> alicebot-archcomm@list.alicebot.org
> http://list.alicebot.org/mailman/listinfo/alicebot-archcomm