On 7/19/12, Kartik Agaram <a...@akkartik.com> wrote:
>> SPLIT *between* symbols is one of those things that is not needed in many
>> cases, but in those few cases where you want it, you REALLY want it.
>
> Absolutely. But if it's an uncommon use case then I'm less concerned
> about what y'all choose.
>
> Unfortunately, Alan's response suggests I'd end up seeing SPLIT
> between symbols a lot if sweetexprs take off :/
>
> I don't think there's a way to resolve this. Both our styles are
> internally consistent, but mutually contradictory. (I tend to use
> dense lines more often than him.) And we're two points on a spectrum
> that's wider and more well-populated throughout than I realized, even
> among experienced lispers.
Well, you could try formalizing ENLIST, which is what you really want,
and which I support, but don't promote.
Assuming ~ = ENLIST:
let ~ x value-for-x(exr)
. y value-for-y(expr)
. {x * y}
This lets you get the density you want.
Like I said, the basic informal concept is there, but I have some
reservations about how it can be formally specified. An informal spec
is good but we need to tell the computers how to read it, and that's
best done formally so that we can be sure that the computer is indeed
doing what we want (and to make sure that we didn't forget some minor
detail).
There's no need to completely dismiss sweet-expressions just because
you see SPLIT winning and dislike the consensus of the symbol being
used. I'd support ENLIST too if there was a formal specification of
what it does, or an implementation: my position is basically \ for
SPLIT and ~ for ENLIST, on the proviso that ENLIST actually gets
formally specified. Heck, I'd like denser code myself, but I don't
mind having to loosen it up given the limitations of SPLIT (basically,
the difference between SPLIT-by-itself, followed by an indented line,
and SPLIT-at-the-start). But if ENLIST does get implemented and
formalized - and if dwheeler can safely unify ENLIST+SPLIT to make
GRIT - I'd go for it.
But I don't have infinite time, so I suggested a spec freeze date of
July 31. And I'm pushing for getting something out: an actual reader
that I can experiment with, get a feel for, and eventually use for
production code. "The real tiger is never a match for the paper one,
unless actual use is wanted."
I'm sorry if I come across as belittling your suggestions - your input
is still welcome. What you want is something like ENLIST. I suggest
you start thinking about how to formalize it, as it seems to match
your preferred style more, and as I myself am preoccupied with SPLIT.
>
> I understand now why we need to introduce GROUP, SPLIT, etc. But
> they're as alien to lisp noobs as parens and prefix. So why bother
> with sweetexprs? Parens/infix + GROUP/SPLIT seems strictly more
> cognitive load than just parens/infix.
I agree. Drop the parens! (^^)v
------------------------------------------------------------------------------
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
Readable-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/readable-discuss
