[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Improving aesthetics & readability of backquote
From: |
Paul W. Rankin |
Subject: |
Re: Improving aesthetics & readability of backquote |
Date: |
Tue, 21 May 2019 12:34:45 +1000 |
User-agent: |
mu4e 1.2.0; emacs 26.2 |
On Tue, May 21 2019, Richard Stallman wrote:
,@b is obviously something magic; people who don't know what it
does will look up what it means.
(splice b) looks like an expression, so people will assume it
calls the function 'splice', and get confused. I think we should
not implement that alias.
I think it would be a worthwhile experiment to run this theory by
a non-programmer, or a non-Lisp programmer, because this is just
not how people think. This distinction between "expression" and
"function" and "construct"... people who haven't spent most of
their lives programming Lisp just don't think in those terms.
Instead we think of the first element inside parentheses as "this
does a thing". We look at something like the above and think
"splice does a thing". The notion that confusion will arise
because (splice b) can only be called within a (quoteval ...), and
that this confusion will supersede the confusion of "what the heck
is ,@b??" is just not realistic.
And there are many times when the first element inside parentheses
actually does *not* do a thing, e.g.
(let ((a 1) (b 2)) ...) ;; a and b don't do things
(dolist (var list) ...) ;; var doesn't do a thing
But "quoteval", "unquote" and "splice" are just suggestions based
on the backquote documentation. If these suggestions are too
troubling, maybe divide this into two parts: 1. the backquote
syntax is ugly and hard to understand, and 2. here's a way it
could be better. Feel free to discard 2.
--
https://www.paulwrankin.com
- Re: Improving aesthetics & readability of backquote, (continued)
Re: Improving aesthetics & readability of backquote, Sam Steingold, 2019/05/22