Terrence,

When we're trying to be strict and unambiguous, we use the formal name of 
primitives (sometimes modified with an adjective
indicating the valence of interest), as in  "dyadic  i.  ".  When we're being 
informal, context is usually sufficient to determine
the primitive in question, and primitives have been denoted with informal names 
that don't even appear in the Dictionary. 

In particular, have a look at the last paragraph of Section III of the 
dictionary.  From
http://www.jsoftware.com/help/dictionary/dict3.htm  :


        Although a name (such as foreign for !:) is suggested for each word, 
    others can be used in addition to or instead of them. Thus, joy might 
    be used for ! since the exclamation mark derives from an I placed 
    above an o, an abbreviation of the Latin io. Similarly, iota might be 
    used instead of integers and index of for i. .

FYI, according to  http://mathworld.wolfram.com/WholeNumber.html  some authors 
use "whole number" to mean a number with a
fractional part of zero.  The article suggests the least ambiguous name for the 
result of monadic  i.  is  "non-negative
integers", and for monadic  i:  , just "integers".

OTOH, it might be a fun contest or exercise to come up with a concise, 
accurate, unique name for monadic  i:  .  The Dictionary
will probably not change, but word games are fun.  I'll start:  "integer 
neighborhood".

-Dan

PS:  The DoJ is a deeply considered, meticulously designed reference.  That 
doesn't mean that the DoJ it doesn't contain bugs and
ambiguities;  it has and it does.  But it does mean that unless the bug is 
obvious (e.g. a spelling error), a change will require
a very strong supporting argument.  

This argument is going to have to outweigh the cost of incorporating the change 
(evaluating whether the change alters the
semantics of the relevant entry, whether it makes the relevant entry less 
concise, whether it will have second order effects
(interactions with other entries), whether it introduces new vocabulary into 
the documentation, whether it conforms to the
existing tone and style, etc etc etc).  

In other words, you're going to have a hard time convincing Roger to make 
semantic changes to the DoJ.  Historically, he's only
made such changes at the suggestions veteran members of the J community, and 
usually he uses his own formulation for the change,
rather than the wording suggested (i.e. he publicly recognizes the problem, but 
uses his own solution).

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to