"
First, [: [: in a tacit expression is "probably a whistleblower"
(which I interpret as: something that matters more immediately than
anything else). Of course, there might be exceptions (: ]) so it's
worth reading more closely. Meanwhile, [: [: [: is almost certainly a
"whistleblower". There could still be exceptions, but they would be
complex to construct. But...
"

I do not have to wonder: if there are no teasers; the remaining caps (if
any) are whistle-blowers.

"
While this is a valid reading, I'll note that there can be exceptions
here, also. For example, how do we know whether (-) means "subtract"
or "negate", for example? In the general case we cannot know until we
can see a representative set of examples. Similarly, in the general
case we will not know when the subtraction takes place (or even if it
will take place).
"

In my mind there is an important difference: this ambivalence of - is the
rule rather than the exception for primitive verbs; most of them are
ambivalent and for good reasons.  A teaser cap is the exception to the
rule, [: is the only verb that is not invoked when is the leading verb in a
fork, for no compelling reason (again, from my viewpoint).

"
Anyways, I think overall I am sympathetic to your point, but I'm also
just a wee bit uncomfortable with it (and I'm also just a wee bit
uncomfortable with Linda's point).

For me, I mostly think of left-to-right reading as "structural" or
maybe "functional" and right-to-left reading as "procedural" or maybe
"imperative". Both can be valid, except sometimes you need the other.
"

I am also sympathetic to your point.  I did not mean to reignite a
controversy that has been discussed too many times.  I was just pointing
out some of my reasons to Linda for avoiding [: teasers since she has
previously expressed her own reasons for avoiding @: .  I think one should
adopt a style that makes oneself more comfortable and presumably more
productive: avoiding none, avoiding one but not the other, or even avoiding
both.

On Sun, May 12, 2013 at 2:29 PM, Raul Miller <[email protected]> wrote:
>
> On Sun, May 12, 2013 at 2:14 PM, Jose Mario Quintana
> <[email protected]> wrote:
> >    ',{:+ [: {:}:y'(13 : )
> > [: , [: {: [: + [: [: [: {: }:
> > In the good old days, from my perspective, @: was used instead of [: and
> > the whistle-blower would have been as visible in the tacit as it is in
the
> > explicit expression.
>
> This reasoning sounds tantalizing, but I am uncomfortable with it.
>
> First, [: [: in a tacit expression is "probably a whistleblower"
> (which I interpret as: something that matters more immediately than
> anything else).  Of course, there might be exceptions (: ]) so it's
> worth reading more closely.  Meanwhile, [: [: [: is almost certainly a
> "whistleblower". There could still be exceptions, but they would be
> complex to construct. But...
>
> > “Once you found a tacit expression for a block, wouldn't you use it in
next
> > definition?
> >
> > ' }. (<:0{x) |. y' (13 :) NB. Finding a tacit form for the for block...
> > [: }. ] |.~ [: <: 0 { [“
> >
> > I read the last line more or less as follows: trigger a domain error,
just
> > kidding; forget the usual fork interpretation, make the middle verb to
act
> > monadic-ally on the ending verb; so, drop the... trigger a domain error,
> > just kidding; forget the usual fork interpretation, make the middle
verb to
> > act monadic-ally on the ending verb; so, decrement the first item on the
> > left (argument).
>
> While this is a valid reading, I'll note that there can be exceptions
> here, also.  For example, how do we know whether (-) means "subtract"
> or "negate", for example? In the general case we cannot know until we
> can see a representative set of examples. Similarly, in the general
> case we will not know when the subtraction takes place (or even if it
> will take place).
>
> Anyways, I think overall I am sympathetic to your point, but I'm also
> just a wee bit uncomfortable with it (and I'm also just a wee bit
> uncomfortable with Linda's point).
>
> For me, I mostly think of left-to-right reading as "structural" or
> maybe "functional" and right-to-left reading as "procedural" or maybe
> "imperative".  Both can be valid, except sometimes you need the other.
>
> --
> Raul
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to