I've thought and experimented with {x} and friends. If we accept {x} as
meaning x, as seems to be the case, then we just don't need the (. e) mapping
in neoteric expressions. {x} is easier and clearer, and is simply superior in
every way. So let's DROP (. e) from neoteric-expressions... we need an escape
mechanism, but we only need one, and {x} is amazingly good. The (. e) was
always complicated, it feels good to get rid of it.
The new proposed changes look good overall:
* {} maps to ()... eliminating a likely error
* {x} maps to x... a useful escape
* {- x} maps to (- x)... eliminating another likely error
* f{} maps to (f)... eliminating a likely error
* f{a} maps to (f a)... eliminating a likely error
* f{- x} maps to (f (- x))... a really useful extension.
Note that f{- x} is different from f(a b), but that's consistent with f{a + b}
being different from f(a b c). Basically, f{...} means that f is passed zero
or one parameter, no matter what.
Given these other additions to curly-infix, I think we should NOT accept
improper lists as basic sweet-expressions. We could, but that will make them
complicated to understand.
So the new list of proposed modifications is here:
https://sourceforge.net/p/readable/wiki/Modifications-0.5/
I'm surprised to be talking about changes at these levels, but I feel good
about these. These are additional capabilities that seem "obviously right".
--- David A. Wheeler
------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Readable-discuss mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/readable-discuss
