Raul Miller wrote (Dec 29):
(me) > I'm afraid I don't see what you're getting at.
> I will try again.
> I will show you an adverb, A, and I will show two examples
> of its use. Then I will show you how J parses this adverb.
> Then I will show you what I believe to be your proposed
> tacit equivalent, and I will ask you how you think J should
> parse that tacit expression.
> A=:1 :'1,2,u'
> 3 A
> 1 2 3
> >: A
> 1 , 2 , >:
We seem to be at cross-purposes.
My requirement is (and has long been) to able, when teaching
or introducing tacit J, to tacitly define operations (adverbs
and conjunctions) and place their operands with the same
freedom that I can place arguments (using [ and ]) when
tacitly defining functions. Or maybe I don't really mean tacit,
since [ and ] are quasiexplicit at least, but what is repellant
is the 1 : ' ' in your example above. This excrescence,
especially the enclosing apostrophes, is a complete
put-off for beginners.
My suggestion, an example rather than a direct requirement,
is to reintroduce primitive [. and ]. and make them work for
operands like [ and ] work for arguments. It seemed to me
that an explicit definition, such as the A above, could simply
have the excrescence removed and any u or v replaced by
[. or ]. respectively. If this doesn't work then my suggestion
is invalid, but I insist that my requirement remains valid, and
satisfying that requirement would enable me to do work
that I am otherwise not able to do.
> So, anyways, I think you want
> T=: 1, 2, [.
> to be equivalent to A.
My suggestion is for that.
> So, when J builds the tacit expression for T,
> should J be using the Dyad parsing rules or the
> Trident parsing rules?
So the answer to this is yes. In my ignorance, I would
imagine this could be done by delaying the decision.
But if that were impossible, and no other simple
suggestion is in the offing, then restricting [. and ].
to standing for functions would still be very useful.
After all, the rank conjunction can make a function
out of any noun.
Jose Mario Quintana wrote (Dec 29):
> If you are confident that your rule is indeed so simple and clear
then you
> should be able to easily implement a working model of your proposal.
You do not
> even have to mess with jijs,ijs; for example, . . .
Thank you very much for trying to show me how to
make a model of my proposal, but I've only ever used
tacit J in teaching and so your example is mysterious
to me. As I've been retired from teaching for quite
a few years now, I only look at the J forum for some
occasional nostalgia, and I haven't actually used J
since then except for the book I started writing, and
I haven't worked on that since a couple of J versions
ago. I just haven't got the time to learn enough to
do what you're suggesting.
> My point is that you might have a better chance that your proposal is
> implemented (so far the silence is deafening), if you show a working
model but even if
> it were not implemented, you and maybe some of us would have a handy
utility as
> a consolation. Again, I would had liked to see your ideas in action.
The silence is indeed deafening. The only reason I drop
this requirement and suggestion in when it seems relevant
to the forum is because I hope it might catch the
imagination of a deus in machina sooner or later.
If it doesn't, eheu fugaces . . .
Thank you very much for your interest and attempted help.
I'm only sorry to be so poor a student for you.
Neville Holmes, P.O.Box 404, Mowbray 7248, Tasmania
Normal e-mail: [EMAIL PROTECTED]
Make the switch to the world's best email. Get the new Yahoo!7 Mail now.
www.yahoo7.com.au/worldsbestemail
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
